$<
in Ordinary Make Rules
make macro=value
and Submakes
SHELL
make -k
VPATH
and Make
This manual is for GNU Autoconf (version 2.60, 23 June 2006), a package for creating scripts to configure source code packages using templates and an M4 macro package.
Copyright © 1992, 1993, 1994, 1995, 1996, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation; with no Invariant Sections, with the Front-Cover texts being “A GNU Manual,” and with the Back-Cover Texts as in (a) below. A copy of the license is included in the section entitled “GNU Free Documentation License.”(a) The FSF's Back-Cover Text is: “You have freedom to copy and modify this GNU Manual, like GNU software. Copies published by the Free Software Foundation raise funds for GNU development.”
--- The Detailed Node Listing ---
The GNU Build System
Making configure Scripts
Writing configure.ac
Initialization and Output Files
Substitutions in Makefiles
Configuration Header Files
Existing Tests
Common Behavior
Alternative Programs
Library Functions
Header Files
Declarations
Structures
Types
Compilers and Preprocessors
Writing Tests
Writing Test Programs
Results of Tests
Caching Results
Programming in M4
M4 Quotation
Using autom4te
Programming in M4sugar
Writing Autoconf Macros
Dependencies Between Macros
Portable Shell Programming
Portable Make Programming
VPATH
and Make
Portable C and C++ Programming
Manual Configuration
Site Configuration
Transforming Program Names When Installing
Running configure Scripts
Obsolete Constructs
Upgrading From Version 1
Upgrading From Version 2.13
Generating Test Suites with Autotest
Using an Autotest Test Suite
Frequent Autoconf Questions, with answers
History of Autoconf
Copying This Manual
Indices
nature of God. “Surely a Physicist,” said the physicist, “because
early in the Creation, God made Light; and you know, Maxwell's
equations, the dual nature of electromagnetic waves, the relativistic
consequences...” “An Engineer!,” said the engineer, “because
before making Light, God split the Chaos into Land and Water; it takes a
hell of an engineer to handle that big amount of mud, and orderly
separation of solids from liquids...” The computer scientist
shouted: “And the Chaos, where do you think it was coming from, hmm?”
—Anonymous
Autoconf is a tool for producing shell scripts that automatically configure software source code packages to adapt to many kinds of Posix-like systems. The configuration scripts produced by Autoconf are independent of Autoconf when they are run, so their users do not need to have Autoconf.
The configuration scripts produced by Autoconf require no manual user intervention when run; they do not normally even need an argument specifying the system type. Instead, they individually test for the presence of each feature that the software package they are for might need. (Before each check, they print a one-line message stating what they are checking for, so the user doesn't get too bored while waiting for the script to finish.) As a result, they deal well with systems that are hybrids or customized from the more common Posix variants. There is no need to maintain files that list the features supported by each release of each variant of Posix.
For each software package that Autoconf is used with, it creates a configuration script from a template file that lists the system features that the package needs or can use. After the shell code to recognize and respond to a system feature has been written, Autoconf allows it to be shared by many software packages that can use (or need) that feature. If it later turns out that the shell code needs adjustment for some reason, it needs to be changed in only one place; all of the configuration scripts can be regenerated automatically to take advantage of the updated code.
The Metaconfig package is similar in purpose to Autoconf, but the scripts it produces require manual user intervention, which is quite inconvenient when configuring large source trees. Unlike Metaconfig scripts, Autoconf scripts can support cross-compiling, if some care is taken in writing them.
Autoconf does not solve all problems related to making portable software packages—for a more complete solution, it should be used in concert with other GNU build tools like Automake and Libtool. These other tools take on jobs like the creation of a portable, recursive makefile with all of the standard targets, linking of shared libraries, and so on. See The GNU Build System, for more information.
Autoconf imposes some restrictions on the names of macros used with
#if
in C programs (see Preprocessor Symbol Index).
Autoconf requires GNU M4 in order to generate the scripts. It uses features that some versions of M4, including GNU M4 1.3, do not have. You should use version 1.4.4 or later of GNU M4.
See Autoconf 1, for information about upgrading from version 1. See History, for the story of Autoconf's development. See FAQ, for answers to some common questions about Autoconf.
See the Autoconf web page for up-to-date information, details on the mailing lists, pointers to a list of known bugs, etc.
Mail suggestions to the Autoconf mailing list. Past suggestions are archived.
Mail bug reports to the Autoconf Bugs mailing list. Past bug reports are archived.
If possible, first check that your bug is not already solved in current development versions, and that it has not been reported yet. Be sure to include all the needed information and a short configure.ac that demonstrates the problem.
Autoconf's development tree is accessible via anonymous CVS; see the Autoconf Summary for details. Patches relative to the current CVS version can be sent for review to the Autoconf Patches mailing list. Past patches are archived.
Because of its mission, the Autoconf package itself includes only a set of often-used macros that have already demonstrated their usefulness. Nevertheless, if you wish to share your macros, or find existing ones, see the Autoconf Macro Archive, which is kindly run by Peter Simons.
Autoconf solves an important problem—reliable discovery of system-specific build and runtime information—but this is only one piece of the puzzle for the development of portable software. To this end, the GNU project has developed a suite of integrated utilities to finish the job Autoconf started: the GNU build system, whose most important components are Autoconf, Automake, and Libtool. In this chapter, we introduce you to those tools, point you to sources of more information, and try to convince you to use the entire GNU build system for your software.
The ubiquity of make means that a makefile is almost the
only viable way to distribute automatic build rules for software, but
one quickly runs into its numerous limitations. Its lack of
support for automatic dependency tracking, recursive builds in
subdirectories, reliable timestamps (e.g., for network file systems), and
so on, mean that developers must painfully (and often incorrectly)
reinvent the wheel for each project. Portability is non-trivial, thanks
to the quirks of make on many systems. On top of all this is the
manual labor required to implement the many standard targets that users
have come to expect (make install
, make distclean
,
make uninstall
, etc.). Since you are, of course, using Autoconf,
you also have to insert repetitive code in your Makefile.in
to
recognize @CC@
, @CFLAGS@
, and other substitutions
provided by configure. Into this mess steps Automake.
Automake allows you to specify your build needs in a Makefile.am
file with a vastly simpler and more powerful syntax than that of a plain
makefile, and then generates a portable Makefile.in
for
use with Autoconf. For example, the Makefile.am
to build and
install a simple “Hello world” program might look like:
bin_PROGRAMS = hello hello_SOURCES = hello.c
The resulting Makefile.in
(~400 lines) automatically supports all
the standard targets, the substitutions provided by Autoconf, automatic
dependency tracking, VPATH
building, and so on. make
builds the hello
program, and make install
installs it
in /usr/local/bin (or whatever prefix was given to
configure, if not /usr/local).
The benefits of Automake increase for larger packages (especially ones with subdirectories), but even for small programs the added convenience and portability can be substantial. And that's not all...
GNU software has a well-deserved reputation for running on many different types of systems. While our primary goal is to write software for the GNU system, many users and developers have been introduced to us through the systems that they were already using.
Gnulib is a central location for common GNU code, intended to be shared among free software packages. Its components are typically shared at the source level, rather than being a library that gets built, installed, and linked against. The idea is to copy files from Gnulib into your own source tree. There is no distribution tarball; developers should just grab source modules from the repository. The source files are available online, under various licenses, mostly GNU GPL or GNU LGPL.
Gnulib modules typically contain C source code along with Autoconf
macros used to configure the source code. For example, the Gnulib
stdbool
module implements a stdbool.h header that nearly
conforms to C99, even on old-fashioned hosts that lack stdbool.h.
This module contains a source file for the replacement header, along
with an Autoconf macro that arranges to use the replacement header on
old-fashioned systems.
Often, one wants to build not only programs, but libraries, so that other programs can benefit from the fruits of your labor. Ideally, one would like to produce shared (dynamically linked) libraries, which can be used by multiple programs without duplication on disk or in memory and can be updated independently of the linked programs. Producing shared libraries portably, however, is the stuff of nightmares—each system has its own incompatible tools, compiler flags, and magic incantations. Fortunately, GNU provides a solution: Libtool. Libtool handles all the requirements of building shared libraries for you, and at this time seems to be the only way to do so with any portability. It also handles many other headaches, such as: the interaction of Make rules with the variable suffixes of shared libraries, linking reliably with shared libraries before they are installed by the superuser, and supplying a consistent versioning system (so that different versions of a library can be installed or upgraded without breaking binary compatibility). Although Libtool, like Autoconf, can be used without Automake, it is most simply utilized in conjunction with Automake—there, Libtool is used automatically whenever shared libraries are needed, and you need not know its syntax.
Developers who are used to the simplicity of make for small projects on a single system might be daunted at the prospect of learning to use Automake and Autoconf. As your software is distributed to more and more users, however, you otherwise quickly find yourself putting lots of effort into reinventing the services that the GNU build tools provide, and making the same mistakes that they once made and overcame. (Besides, since you're already learning Autoconf, Automake is a piece of cake.)
There are a number of places that you can go to for more information on the GNU build tools.
See Automake, for more information on Automake.
The book GNU Autoconf, Automake and Libtool1 describes the complete GNU build environment. You can also find the entire book on-line.
The configuration scripts that Autoconf produces are by convention called configure. When run, configure creates several files, replacing configuration parameters in them with appropriate values. The files that configure creates are:
#define
directives (see Configuration Headers);
To create a configure script with Autoconf, you need to write an
Autoconf input file configure.ac (or configure.in) and run
autoconf on it. If you write your own feature tests to
supplement those that come with Autoconf, you might also write files
called aclocal.m4 and acsite.m4. If you use a C header
file to contain #define
directives, you might also run
autoheader, and you can distribute the generated file
config.h.in with the package.
Here is a diagram showing how the files that can be used in configuration are produced. Programs that are executed are suffixed by ‘*’. Optional files are enclosed in square brackets (‘[]’). autoconf and autoheader also read the installed Autoconf macro files (by reading autoconf.m4).
Files used in preparing a software package for distribution:
your source files --> [autoscan*] --> [configure.scan] --> configure.ac configure.ac --. | .------> autoconf* -----> configure [aclocal.m4] --+---+ | `-----> [autoheader*] --> [config.h.in] [acsite.m4] ---' Makefile.in -------------------------------> Makefile.in
Files used in configuring a software package:
.-------------> [config.cache] configure* ------------+-------------> config.log | [config.h.in] -. v .-> [config.h] -. +--> config.status* -+ +--> make* Makefile.in ---' `-> Makefile ---'
To produce a configure script for a software package, create a file called configure.ac that contains invocations of the Autoconf macros that test the system features your package needs or can use. Autoconf macros already exist to check for many features; see Existing Tests, for their descriptions. For most other features, you can use Autoconf template macros to produce custom checks; see Writing Tests, for information about them. For especially tricky or specialized features, configure.ac might need to contain some hand-crafted shell commands; see Portable Shell. The autoscan program can give you a good start in writing configure.ac (see autoscan Invocation, for more information).
Previous versions of Autoconf promoted the name configure.in, which is somewhat ambiguous (the tool needed to process this file is not described by its extension), and introduces a slight confusion with config.h.in and so on (for which ‘.in’ means “to be processed by configure”). Using configure.ac is now preferred.
Just as for any other computer language, in order to properly program configure.ac in Autoconf you must understand what problem the language tries to address and how it does so.
The problem Autoconf addresses is that the world is a mess. After all, you are using Autoconf in order to have your package compile easily on all sorts of different systems, some of them being extremely hostile. Autoconf itself bears the price for these differences: configure must run on all those systems, and thus configure must limit itself to their lowest common denominator of features.
Naturally, you might then think of shell scripts; who needs autoconf? A set of properly written shell functions is enough to make it easy to write configure scripts by hand. Sigh! Unfortunately, shell functions do not belong to the least common denominator; therefore, where you would like to define a function and use it ten times, you would instead need to copy its body ten times.
So, what is really needed is some kind of compiler, autoconf, that takes an Autoconf program, configure.ac, and transforms it into a portable shell script, configure.
How does autoconf perform this task?
There are two obvious possibilities: creating a brand new language or
extending an existing one. The former option is attractive: all
sorts of optimizations could easily be implemented in the compiler and
many rigorous checks could be performed on the Autoconf program
(e.g., rejecting any non-portable construct). Alternatively, you can
extend an existing language, such as the sh
(Bourne shell)
language.
Autoconf does the latter: it is a layer on top of sh
. It was
therefore most convenient to implement autoconf as a macro
expander: a program that repeatedly performs macro expansions on
text input, replacing macro calls with macro bodies and producing a pure
sh
script in the end. Instead of implementing a dedicated
Autoconf macro expander, it is natural to use an existing
general-purpose macro language, such as M4, and implement the extensions
as a set of M4 macros.
The Autoconf language differs from many other computer languages because it treats actual code the same as plain text. Whereas in C, for instance, data and instructions have different syntactic status, in Autoconf their status is rigorously the same. Therefore, we need a means to distinguish literal strings from text to be expanded: quotation.
When calling macros that take arguments, there must not be any white space between the macro name and the open parenthesis. Arguments should be enclosed within the M4 quote characters ‘[’ and ‘]’, and be separated by commas. Any leading blanks or newlines in arguments are ignored, unless they are quoted. You should always quote an argument that might contain a macro name, comma, parenthesis, or a leading blank or newline. This rule applies recursively for every macro call, including macros called from other macros.
For instance:
AC_CHECK_HEADER([stdio.h], [AC_DEFINE([HAVE_STDIO_H], [1], [Define to 1 if you have <stdio.h>.])], [AC_MSG_ERROR([Sorry, can't do anything for you])])
is quoted properly. You may safely simplify its quotation to:
AC_CHECK_HEADER([stdio.h], [AC_DEFINE([HAVE_STDIO_H], 1, [Define to 1 if you have <stdio.h>.])], [AC_MSG_ERROR([Sorry, can't do anything for you])])
because ‘1’ cannot contain a macro call. Here, the argument of
AC_MSG_ERROR
must be quoted; otherwise, its comma would be
interpreted as an argument separator. Also, the second and third arguments
of ‘AC_CHECK_HEADER’ must be quoted, since they contain
macro calls. The three arguments ‘HAVE_STDIO_H’, ‘stdio.h’,
and ‘Define to 1 if you have <stdio.h>.’ do not need quoting, but
if you unwisely defined a macro with a name like ‘Define’ or
‘stdio’ then they would need quoting. Cautious Autoconf users
would keep the quotes, but many Autoconf users find such precautions
annoying, and would rewrite the example as follows:
AC_CHECK_HEADER(stdio.h, [AC_DEFINE(HAVE_STDIO_H, 1, [Define to 1 if you have <stdio.h>.])], [AC_MSG_ERROR([Sorry, can't do anything for you])])
This is safe, so long as you adopt good naming conventions and do not define macros with names like ‘HAVE_STDIO_H’, ‘stdio’, or ‘h’. Though it is also safe here to omit the quotes around ‘Define to 1 if you have <stdio.h>.’ this is not recommended, as message strings are more likely to inadvertently contain commas.
The following example is wrong and dangerous, as it is underquoted:
AC_CHECK_HEADER(stdio.h, AC_DEFINE(HAVE_STDIO_H, 1, Define to 1 if you have <stdio.h>.), AC_MSG_ERROR([Sorry, can't do anything for you]))
In other cases, you may have to use text that also resembles a macro call. You must quote that text even when it is not passed as a macro argument:
echo "Hard rock was here! --[AC_DC]"
which results in:
echo "Hard rock was here! --AC_DC"
When you use the same text in a macro argument, you must therefore have an extra quotation level (since one is stripped away by the macro substitution). In general, then, it is a good idea to use double quoting for all literal string arguments:
AC_MSG_WARN([[AC_DC stinks --Iron Maiden]])
You are now able to understand one of the constructs of Autoconf that has been continually misunderstood... The rule of thumb is that whenever you expect macro expansion, expect quote expansion; i.e., expect one level of quotes to be lost. For instance:
AC_COMPILE_IFELSE([char b[10];], [], [AC_MSG_ERROR([you lose])])
is incorrect: here, the first argument of AC_COMPILE_IFELSE
is
‘char b[10];’ and is expanded once, which results in
‘char b10;’. (There was an idiom common in Autoconf's past to
address this issue via the M4 changequote
primitive, but do not
use it!) Let's take a closer look: the author meant the first argument
to be understood as a literal, and therefore it must be quoted twice:
AC_COMPILE_IFELSE([[char b[10];]], [], [AC_MSG_ERROR([you lose])])
Voilà, you actually produce ‘char b[10];’ this time!
On the other hand, descriptions (e.g., the last parameter of
AC_DEFINE
or AS_HELP_STRING
) are not literals—they
are subject to line breaking, for example—and should not be double quoted.
Even if these descriptions are short and are not actually broken, double
quoting them yields weird results.
Some macros take optional arguments, which this documentation represents as [arg] (not to be confused with the quote characters). You may just leave them empty, or use ‘[]’ to make the emptiness of the argument explicit, or you may simply omit the trailing commas. The three lines below are equivalent:
AC_CHECK_HEADERS([stdio.h], [], [], []) AC_CHECK_HEADERS([stdio.h],,,) AC_CHECK_HEADERS([stdio.h])
It is best to put each macro call on its own line in configure.ac. Most of the macros don't add extra newlines; they rely on the newline after the macro call to terminate the commands. This approach makes the generated configure script a little easier to read by not inserting lots of blank lines. It is generally safe to set shell variables on the same line as a macro call, because the shell allows assignments without intervening newlines.
You can include comments in configure.ac files by starting them with the ‘#’. For example, it is helpful to begin configure.ac files with a line like this:
# Process this file with autoconf to produce a configure script.
The order in which configure.ac calls the Autoconf macros is not
important, with a few exceptions. Every configure.ac must
contain a call to AC_INIT
before the checks, and a call to
AC_OUTPUT
at the end (see Output). Additionally, some macros
rely on other macros having been called first, because they check
previously set values of some variables to decide what to do. These
macros are noted in the individual descriptions (see Existing Tests), and they also warn you when configure is created if they
are called out of order.
To encourage consistency, here is a suggested order for calling the Autoconf macros. Generally speaking, the things near the end of this list are those that could depend on things earlier in it. For example, library functions could be affected by types and libraries.
Autoconf requirementsAC_INIT(
package,
version,
bug-report-address)
information on the package checks for programs checks for libraries checks for header files checks for types checks for structures checks for compiler characteristics checks for library functions checks for system servicesAC_CONFIG_FILES(
[file...])
AC_OUTPUT
The autoscan program can help you create and/or maintain a configure.ac file for a software package. autoscan examines source files in the directory tree rooted at a directory given as a command line argument, or the current directory if none is given. It searches the source files for common portability problems and creates a file configure.scan which is a preliminary configure.ac for that package, and checks a possibly existing configure.ac for completeness.
When using autoscan to create a configure.ac, you
should manually examine configure.scan before renaming it to
configure.ac; it probably needs some adjustments.
Occasionally, autoscan outputs a macro in the wrong order
relative to another macro, so that autoconf produces a warning;
you need to move such macros manually. Also, if you want the package to
use a configuration header file, you must add a call to
AC_CONFIG_HEADERS
(see Configuration Headers). You might
also have to change or add some #if
directives to your program in
order to make it work with Autoconf (see ifnames Invocation, for
information about a program that can help with that job).
When using autoscan to maintain a configure.ac, simply consider adding its suggestions. The file autoscan.log contains detailed information on why a macro is requested.
autoscan uses several data files (installed along with Autoconf) to determine which macros to output when it finds particular symbols in a package's source files. These data files all have the same format: each line consists of a symbol, one or more blanks, and the Autoconf macro to output if that symbol is encountered. Lines starting with ‘#’ are comments.
autoscan accepts the following options:
ifnames can help you write configure.ac for a software package. It prints the identifiers that the package already uses in C preprocessor conditionals. If a package has already been set up to have some portability, ifnames can thus help you figure out what its configure needs to check for. It may help fill in some gaps in a configure.ac generated by autoscan (see autoscan Invocation).
ifnames scans all of the C source files named on the command line
(or the standard input, if none are given) and writes to the standard
output a sorted list of all the identifiers that appear in those files
in #if
, #elif
, #ifdef
, or #ifndef
directives. It prints each identifier on a line, followed by a
space-separated list of the files in which that identifier occurs.
ifnames accepts the following options:
To create configure from configure.ac, run the autoconf program with no arguments. autoconf processes configure.ac with the M4 macro processor, using the Autoconf macros. If you give autoconf an argument, it reads that file instead of configure.ac and writes the configuration script to the standard output instead of to configure. If you give autoconf the argument -, it reads from the standard input instead of configure.ac and writes the configuration script to the standard output.
The Autoconf macros are defined in several files. Some of the files are distributed with Autoconf; autoconf reads them first. Then it looks for the optional file acsite.m4 in the directory that contains the distributed Autoconf macro files, and for the optional file aclocal.m4 in the current directory. Those files can contain your site's or the package's own Autoconf macro definitions (see Writing Autoconf Macros, for more information). If a macro is defined in more than one of the files that autoconf reads, the last definition it reads overrides the earlier ones.
autoconf accepts the following options:
AC_DIAGNOSE
, for a comprehensive list of categories. Special
values include:
Warnings about ‘syntax’ are enabled by default, and the environment variable WARNINGS, a comma separated list of categories, is honored as well. Passing -W category actually behaves as if you had passed --warnings=syntax,$WARNINGS,category. If you want to disable the defaults and WARNINGS, but (for example) enable the warnings about obsolete constructs, you would use -W none,obsolete.
Because autoconf uses autom4te behind the scenes, it
displays a back trace for errors, but not for warnings; if you want
them, just pass -W error. See autom4te Invocation, for some
examples.
The format is a regular string, with newlines if desired, and
several special escape codes. It defaults to ‘$f:$l:$n:$%’; see
autom4te Invocation, for details on the format.
AC_DEFUN
definitions). This
results in a noticeable speedup, but can be disabled by this option.
It is often necessary to check the content of a configure.ac file, but parsing it yourself is extremely fragile and error-prone. It is suggested that you rely upon --trace to scan configure.ac. For instance, to find the list of variables that are substituted, use:
$ autoconf -t AC_SUBST configure.ac:2:AC_SUBST:ECHO_C configure.ac:2:AC_SUBST:ECHO_N configure.ac:2:AC_SUBST:ECHO_T More traces deleted
The example below highlights the difference between ‘$@’, ‘$*’, and ‘$%’.
$ cat configure.ac AC_DEFINE(This, is, [an [example]]) $ autoconf -t 'AC_DEFINE:@: $@ *: $* %: $%' @: [This],[is],[an [example]] *: This,is,an [example] %: This:is:an [example]
The format gives you a lot of freedom:
$ autoconf -t 'AC_SUBST:$$ac_subst{"$1"} = "$f:$l";' $ac_subst{"ECHO_C"} = "configure.ac:2"; $ac_subst{"ECHO_N"} = "configure.ac:2"; $ac_subst{"ECHO_T"} = "configure.ac:2"; More traces deleted
A long separator can be used to improve the readability of complex structures, and to ease their parsing (for instance when no single character is suitable as a separator):
$ autoconf -t 'AM_MISSING_PROG:${|:::::|}*' ACLOCAL|:::::|aclocal|:::::|$missing_dir AUTOCONF|:::::|autoconf|:::::|$missing_dir AUTOMAKE|:::::|automake|:::::|$missing_dir More traces deleted
Installing the various components of the GNU Build System can be tedious: running autopoint for Gettext, automake for Makefile.in etc. in each directory. It may be needed either because some tools such as automake have been updated on your system, or because some of the sources such as configure.ac have been updated, or finally, simply in order to install the GNU Build System in a fresh tree.
autoreconf runs autoconf, autoheader, aclocal, automake, libtoolize, and autopoint (when appropriate) repeatedly to update the GNU Build System in the specified directories and their subdirectories (see Subdirectories). By default, it only remakes those files that are older than their sources.
If you install a new version of some tool, you can make autoreconf remake all of the files by giving it the --force option.
See Automatic Remaking, for Make rules to automatically remake configure scripts when their source files change. That method handles the timestamps of configuration header templates properly, but does not pass --autoconf-dir=dir or --localdir=dir.
Gettext supplies the autopoint command to add translation
infrastructure to a source package. If you use autopoint,
your configure.ac should invoke both AM_GNU_GETTEXT
and
AM_GNU_GETTEXT_VERSION(
gettext-version)
. See Invoking the autopoint
Program, for further details.
autoreconf accepts the following options:
If deemed appropriate, this option triggers calls to
‘automake --add-missing’,
‘libtoolize’, ‘autopoint’, etc.
AC_CONFIG_SUBDIRS
).
Warnings about ‘syntax’ are enabled by default, and the environment variable WARNINGS, a comma separated list of categories, is honored as well. Passing -W category actually behaves as if you had passed --warnings=syntax,$WARNINGS,category. If you want to disable the defaults and WARNINGS, but (for example) enable the warnings about obsolete constructs, you would use -W none,obsolete.
If you want autoreconf to pass flags that are not listed here
on to aclocal, set ACLOCAL_AMFLAGS
in your Makefile.am.
Autoconf-generated configure scripts need some information about how to initialize, such as how to find the package's source files and about the output files to produce. The following sections describe the initialization and the creation of output files.
Every configure script must call AC_INIT
before doing
anything else. The only other required macro is AC_OUTPUT
(see Output).
Process any command-line arguments and perform various initializations and verifications.
Set the name of the package and its version. These are typically used in --version support, including that of configure. The optional argument bug-report should be the email to which users should send bug reports. The package tarname differs from package: the latter designates the full package name (e.g., ‘GNU Autoconf’), while the former is meant for distribution tar ball names (e.g., ‘autoconf’). It defaults to package with ‘GNU ’ stripped, lower-cased, and all characters other than alphanumerics and underscores are changed to ‘-’.
It is preferable that the arguments of
AC_INIT
be static, i.e., there should not be any shell computation, but they can be computed by M4.The following M4 macros (e.g.,
AC_PACKAGE_NAME
), output variables (e.g.,PACKAGE_NAME
), and preprocessor symbols (e.g.,PACKAGE_NAME
) are defined byAC_INIT
:
If your configure script does its own option processing, it
should inspect ‘$@’ or ‘$*’ immediately after calling
AC_INIT
, because other Autoconf macros liberally use the
set command to process strings, and this has the side effect
of updating ‘$@’ and ‘$*’. However, we suggest that you use
standard macros like AC_ARG_ENABLE
instead of attempting to
implement your own option processing. See Site Configuration.
The following macros manage version numbers for configure scripts. Using them is optional.
Ensure that a recent enough version of Autoconf is being used. If the version of Autoconf being used to create configure is earlier than version, print an error message to the standard error output and exit with failure (exit status is 63). For example:
AC_PREREQ([2.60])This macro is the only macro that may be used before
AC_INIT
, but for consistency, you are invited not to do so.
State that, in addition to the Free Software Foundation's copyright on the Autoconf macros, parts of your configure are covered by the copyright-notice.
The copyright-notice shows up in both the head of configure and in ‘configure --version’.
Copy revision stamp revision-info into the configure script, with any dollar signs or double-quotes removed. This macro lets you put a revision stamp from configure.ac into configure without RCS or CVS changing it when you check in configure. That way, you can determine easily which revision of configure.ac a particular configure corresponds to.
For example, this line in configure.ac:
AC_REVISION([$Revision: 1.1 $])produces this in configure:
#!/bin/sh # From configure.ac Revision: 1.30
unique-file-in-source-dir is some file that is in the package's source directory; configure checks for this file's existence to make sure that the directory that it is told contains the source code in fact does. Occasionally people accidentally specify the wrong directory with --srcdir; this is a safety check. See configure Invocation, for more information.
Packages that do manual configuration or use the install program
might need to tell configure where to find some other shell
scripts by calling AC_CONFIG_AUX_DIR
, though the default places
it looks are correct for most cases.
Use the auxiliary build tools (e.g., install-sh, config.sub, config.guess, Cygnus configure, Automake and Libtool scripts, etc.) that are in directory dir. These are auxiliary files used in configuration. dir can be either absolute or relative to srcdir. The default is srcdir or srcdir/.. or srcdir/../.., whichever is the first that contains install-sh. The other files are not checked for, so that using
AC_PROG_INSTALL
does not automatically require distributing the other auxiliary files. It checks for install.sh also, but that name is obsolete because somemake
have a rule that creates install from it if there is no makefile.The auxiliary directory is commonly named build-aux. If you need portability to DOS variants, do not name the auxiliary directory aux. See File System Conventions.
Declares that file is expected in the directory defined above. In Autoconf proper, this macro does nothing: its sole purpose is to be traced by third-party tools to produce a list of expected auxiliary files. For instance it is called by macros like
AC_PROG_INSTALL
(see Particular Programs) orAC_CANONICAL_BUILD
(see Canonicalizing) to register the auxiliary files they need.
Similarly, packages that use aclocal should declare where
local macros can be found using AC_CONFIG_MACRO_DIR
.
Future versions of autopoint, libtoolize, aclocal and autoreconf will use directory dir as the location of additional local Autoconf macros. Be sure to call this macro directly from configure.ac so that tools that install macros for aclocal can find the declaration before --trace can be called safely.
Every Autoconf script, e.g., configure.ac, should finish by
calling AC_OUTPUT
. That is the macro that generates and runs
config.status, which in turn creates the makefiles and any
other files resulting from configuration. This is the only required
macro besides AC_INIT
(see Input).
Generate config.status and launch it. Call this macro once, at the end of configure.ac.
config.status performs all the configuration actions: all the output files (see Configuration Files, macro
AC_CONFIG_FILES
), header files (see Configuration Headers, macroAC_CONFIG_HEADERS
), commands (see Configuration Commands, macroAC_CONFIG_COMMANDS
), links (see Configuration Links, macroAC_CONFIG_LINKS
), subdirectories to configure (see Subdirectories, macroAC_CONFIG_SUBDIRS
) are honored.The location of your
AC_OUTPUT
invocation is the exact point where configuration actions are taken: any code afterwards is executed byconfigure
once config.status was run. If you want to bind actions to config.status itself (independently of whether configure is being run), see Running Arbitrary Configuration Commands.
Historically, the usage of AC_OUTPUT
was somewhat different.
See Obsolete Macros, for a description of the arguments that
AC_OUTPUT
used to support.
If you run make in subdirectories, you should run it using the
make
variable MAKE
. Most versions of make set
MAKE
to the name of the make program plus any options it
was given. (But many do not include in it the values of any variables
set on the command line, so those are not passed on automatically.)
Some old versions of make do not set this variable. The
following macro allows you to use it even with those versions.
If the Make command,
$MAKE
if set or else ‘make’, predefines$(MAKE)
, define output variableSET_MAKE
to be empty. Otherwise, defineSET_MAKE
to a macro definition that sets$(MAKE)
, such as ‘MAKE=make’. CallsAC_SUBST
forSET_MAKE
.
If you use this macro, place a line like this in each Makefile.in
that runs MAKE
on other directories:
@SET_MAKE@
configure is designed so that it appears to do everything itself, but there is actually a hidden slave: config.status. configure is in charge of examining your system, but it is config.status that actually takes the proper actions based on the results of configure. The most typical task of config.status is to instantiate files.
This section describes the common behavior of the four standard
instantiating macros: AC_CONFIG_FILES
, AC_CONFIG_HEADERS
,
AC_CONFIG_COMMANDS
and AC_CONFIG_LINKS
. They all
have this prototype:
AC_CONFIG_FOOS(tag..., [commands], [init-cmds])
where the arguments are:
You are encouraged to use literals as tags. In particular, you should avoid
... && my_foos="$my_foos fooo" ... && my_foos="$my_foos foooo" AC_CONFIG_FOOS([$my_foos])
and use this instead:
... && AC_CONFIG_FOOS([fooo]) ... && AC_CONFIG_FOOS([foooo])
The macros AC_CONFIG_FILES
and AC_CONFIG_HEADERS
use
special tag values: they may have the form ‘output’ or
‘output:inputs’. The file output is instantiated
from its templates, inputs (defaulting to ‘output.in’).
‘AC_CONFIG_FILES([Makefile:boiler/top.mk:boiler/bot.mk)]’, for example, asks for the creation of the file Makefile that contains the expansion of the output variables in the concatenation of boiler/top.mk and boiler/bot.mk.
The special value ‘-’ might be used to denote the standard output when used in output, or the standard input when used in the inputs. You most probably don't need to use this in configure.ac, but it is convenient when using the command line interface of ./config.status, see config.status Invocation, for more details.
The inputs may be absolute or relative file names. In the latter
case they are first looked for in the build tree, and then in the source
tree.
The variables set during the execution of configure are not available here: you first need to set them via the init-cmds. Nonetheless the following variables are precomputed:
srcdir
ac_top_srcdir
ac_top_build_prefix
ac_srcdir
The current directory refers to the directory (or pseudo-directory) containing the input part of tags. For instance, running
AC_CONFIG_COMMANDS([deep/dir/out:in/in.in], [...], [...])
with --srcdir=../package produces the following values:
# Argument of --srcdir srcdir='../package' # Reversing deep/dir ac_top_build_prefix='../../' # Concatenation of $ac_top_build_prefix and srcdir ac_top_srcdir='../../../package' # Concatenation of $ac_top_srcdir and deep/dir ac_srcdir='../../../package/deep/dir'
independently of ‘in/in.in’.
var
. init-cmds
is typically used by configure to give config.status some
variables it needs to run the commands.
You should be extremely cautious in your variable names: all the init-cmds share the same name space and may overwrite each other in unpredictable ways. Sorry...
All these macros can be called multiple times, with different tag values, of course!
Be sure to read the previous section, Configuration Actions.
Make
AC_OUTPUT
create each file by copying an input file (by default file.in), substituting the output variable values. This macro is one of the instantiating macros; see Configuration Actions. See Makefile Substitutions, for more information on using output variables. See Setting Output Variables, for more information on creating them. This macro creates the directory that the file is in if it doesn't exist. Usually, makefiles are created this way, but other files, such as .gdbinit, can be specified as well.Typical calls to
AC_CONFIG_FILES
look like this:AC_CONFIG_FILES([Makefile src/Makefile man/Makefile X/Imakefile]) AC_CONFIG_FILES([autoconf], [chmod +x autoconf])You can override an input file name by appending to file a colon-separated list of input files. Examples:
AC_CONFIG_FILES([Makefile:boiler/top.mk:boiler/bot.mk] [lib/Makefile:boiler/lib.mk])Doing this allows you to keep your file names acceptable to DOS variants, or to prepend and/or append boilerplate to the file.
Each subdirectory in a distribution that contains something to be
compiled or installed should come with a file Makefile.in, from
which configure creates a file Makefile in that directory.
To create Makefile, configure performs a simple variable
substitution, replacing occurrences of ‘@variable@’ in
Makefile.in with the value that configure has determined
for that variable. Variables that are substituted into output files in
this way are called output variables. They are ordinary shell
variables that are set in configure. To make configure
substitute a particular variable into the output files, the macro
AC_SUBST
must be called with that variable name as an argument.
Any occurrences of ‘@variable@’ for other variables are
left unchanged. See Setting Output Variables, for more information
on creating output variables with AC_SUBST
.
A software package that uses a configure script should be distributed with a file Makefile.in, but no makefile; that way, the user has to properly configure the package for the local system before compiling it.
See Makefile Conventions, for more information on what to put in makefiles.
Some output variables are preset by the Autoconf macros. Some of the
Autoconf macros set additional output variables, which are mentioned in
the descriptions for those macros. See Output Variable Index, for a
complete list of output variables. See Installation Directory Variables, for the list of the preset ones related to installation
directories. Below are listed the other preset ones. They all are
precious variables (see Setting Output Variables,
AC_ARG_VAR
).
Debugging and optimization options for the C compiler. If it is not set in the environment when configure runs, the default value is set when you call
AC_PROG_CC
(or empty if you don't). configure uses this variable when compiling programs to test for C features.
A comment saying that the file was generated automatically by configure and giving the name of the input file.
AC_OUTPUT
adds a comment line containing this variable to the top of every makefile it creates. For other files, you should reference this variable in a comment at the top of each input file. For example, an input shell script should begin like this:#!/bin/sh # @configure_input@The presence of that line also reminds people editing the file that it needs to be processed by configure in order to be used.
Header file search directory (-Idir) and any other miscellaneous options for the C and C++ preprocessors and compilers. If it is not set in the environment when configure runs, the default value is empty. configure uses this variable when compiling or preprocessing programs to test for C and C++ features. See Special Chars in Variables, for limitations that
CPPFLAGS
might run into.
Debugging and optimization options for the C++ compiler. If it is not set in the environment when configure runs, the default value is set when you call
AC_PROG_CXX
(or empty if you don't). configure uses this variable when compiling programs to test for C++ features.
-D options to pass to the C compiler. If
AC_CONFIG_HEADERS
is called, configure replaces ‘@DEFS@’ with -DHAVE_CONFIG_H instead (see Configuration Headers). This variable is not defined while configure is performing its tests, only when creating the output files. See Setting Output Variables, for how to check the results of previous tests.
How does one suppress the trailing newline from echo for question-answer message pairs? These variables provide a way:
echo $ECHO_N "And the winner is... $ECHO_C" sleep 100000000000 echo "${ECHO_T}dead."Some old and uncommon echo implementations offer no means to achieve this, in which case
ECHO_T
is set to tab. You might not want to use it.
Debugging and optimization options for the Erlang compiler. If it is not set in the environment when configure runs, the default value is empty. configure uses this variable when compiling programs to test for Erlang features.
Debugging and optimization options for the Fortran compiler. If it is not set in the environment when configure runs, the default value is set when you call
AC_PROG_FC
(or empty if you don't). configure uses this variable when compiling programs to test for Fortran features.
Debugging and optimization options for the Fortran 77 compiler. If it is not set in the environment when configure runs, the default value is set when you call
AC_PROG_F77
(or empty if you don't). configure uses this variable when compiling programs to test for Fortran 77 features.
Stripping (-s), path (-L), and any other miscellaneous options for the linker. Don't use this variable to pass library names (-l) to the linker, use
LIBS
instead. If it is not set in the environment when configure runs, the default value is empty. configure uses this variable when linking programs to test for C, C++, and Fortran features.
-l options to pass to the linker. The default value is empty, but some Autoconf macros may prepend extra libraries to this variable if those libraries are found and provide necessary functions, see Libraries. configure uses this variable when linking programs to test for C, C++, and Fortran features.
Debugging and optimization options for the Objective C compiler. If it is not set in the environment when configure runs, the default value is set when you call
AC_PROG_OBJC
(or empty if you don't). configure uses this variable when compiling programs to test for Objective C features.
The relative name of the top level of the current build tree. In the top-level directory, this is the same as
builddir
.
The relative name of the directory that contains the source code for that makefile.
The relative name of the top-level source code directory for the package. In the top-level directory, this is the same as
srcdir
.
The following variables specify the directories for package installation, see Variables for Installation Directories, for more information. See the end of this section for details on when and how to use these variables.
The directory for installing idiosyncratic read-only architecture-independent data.
The root of the directory tree for read-only architecture-independent data files.
The installation prefix for architecture-dependent files. By default it's the same as prefix. You should avoid installing anything directly to exec_prefix. However, the default value for directories containing architecture-dependent files should be relative to exec_prefix.
The directory for installing locale-dependent but architecture-independent data, such as message catalogs. This directory usually has a subdirectory per locale.
The common installation prefix for all files. If exec_prefix is defined to a different value, prefix is used only for architecture-independent files.
Most of these variables have values that rely on prefix
or
exec_prefix
. It is deliberate that the directory output
variables keep them unexpanded: typically ‘@datarootdir@’ is
replaced by ‘${prefix}/share’, not ‘/usr/local/share’, and
‘@datadir@’ is replaced by ‘${datarootdir}’.
This behavior is mandated by the GNU coding standards, so that when the user runs:
In order to support these features, it is essential that
datarootdir
remains being defined as ‘${prefix}/share’ to
depend upon the current value of prefix
.
A corollary is that you should not use these variables except in
makefiles. For instance, instead of trying to evaluate datadir
in configure and hard-coding it in makefiles using
e.g., ‘AC_DEFINE_UNQUOTED([DATADIR], ["$datadir"], [Data directory.])’,
you should add
-DDATADIR='$(datadir)' to your makefile's definition of
CPPFLAGS
(AM_CPPFLAGS
if you are also using Automake).
Similarly, you should not rely on AC_CONFIG_FILES
to replace
datadir
and friends in your shell scripts and other files; instead,
let make manage their replacement. For instance Autoconf
ships templates of its shell scripts ending with ‘.in’, and uses a
makefile snippet similar to the following to build scripts like
autoheader and autom4te:
edit = sed \ -e 's|@datadir[@]|$(pkgdatadir)|g' \ -e 's|@prefix[@]|$(prefix)|g' autoheader autom4te: Makefile rm -f $@ $@.tmp $(edit) '$(srcdir)/$@.in' >$@.tmp chmod +x $@.tmp chmod a-w $@.tmp mv $@.tmp $@ autoheader: $(srcdir)/autoheader.in autom4te: $(srcdir)/autom4te.in
Some details are noteworthy:
VPATH
should
not contain shell metacharacters or white
space. See Special Chars in Variables.
edit
uses values that depend on the configuration specific
values (prefix
, etc.) and not only on VERSION
and so forth,
the output depends on Makefile, not configure.ac.
autoconf autoheader: Makefile .in: rm -f $@ $@.tmp $(edit) $< >$@.tmp chmod +x $@.tmp mv $@.tmp $@
See Single Suffix Rules, for details.
For the more specific installation of Erlang libraries, the following variables are defined:
The common parent directory of Erlang library installation directories. This variable is set by calling the
AC_ERLANG_SUBST_INSTALL_LIB_DIR
macro in configure.ac.
The installation directory for Erlang library library. This variable is set by calling the ‘AC_ERLANG_SUBST_INSTALL_LIB_SUBDIR(library, version’ macro in configure.ac.
See Erlang Libraries, for details.
In Autoconf 2.60, the set of directory variables has changed, and the defaults of some variables have been adjusted (see Installation Directory Variables) to changes in the GNU Coding Standards. Notably, datadir, infodir, and mandir are now expressed in terms of datarootdir. If you are upgrading from an earlier Autoconf version, you may need to adjust your files to ensure that the directory variables are substituted correctly (see Defining Directories), and that a definition of datarootdir is in place. For example, in a Makefile.in, adding
datarootdir = @datarootdir@
is usually sufficient. If you use Automake to create Makefile.in, it will add this for you.
To help with the transition, Autoconf warns about files that seem to use
datarootdir
without defining it. In some cases, it then expands
the value of $datarootdir
in substitutions of the directory
variables. The following example shows such a warning:
$ cat configure.ac AC_INIT AC_CONFIG_FILES([Makefile]) AC_OUTPUT $ cat Makefile.in prefix = @prefix@ datadir = @datadir@ $ autoconf $ configure configure: creating ./config.status config.status: creating Makefile config.status: WARNING: Makefile.in seems to ignore the --datarootdir setting $ cat Makefile prefix = /usr/local datadir = ${prefix}/share
Usually one can easily change the file to accommodate both older and newer Autoconf releases:
$ cat Makefile.in prefix = @prefix@ datarootdir = @datarootdir@ datadir = @datadir@ $ configure configure: creating ./config.status config.status: creating Makefile $ cat Makefile prefix = /usr/local datarootdir = ${prefix}/share datadir = ${datarootdir}
In some cases, however, the checks may not be able to detect that a suitable
initialization of datarootdir
is in place, or they may fail to detect
that such an initialization is necessary in the output file. If, after
auditing your package, there are still spurious configure warnings about
datarootdir
, you may add the line
AC_DEFUN([AC_DATAROOTDIR_CHECKED])
to your configure.ac to disable the warnings. This is an exception
to the usual rule that you should not define a macro whose name begins with
AC_
(see Macro Names).
You can support compiling a software package for several architectures simultaneously from the same copy of the source code. The object files for each architecture are kept in their own directory.
To support doing this, make uses the VPATH
variable to
find the files that are in the source directory. GNU Make
and most other recent make programs can do this. Older
make programs do not support VPATH
; when using them, the
source code must be in the same directory as the object files.
To support VPATH
, each Makefile.in should contain two
lines that look like:
srcdir = @srcdir@ VPATH = @srcdir@
Do not set VPATH
to the value of another variable, for example
‘VPATH = $(srcdir)’, because some versions of make do not do
variable substitutions on the value of VPATH
.
configure substitutes the correct value for srcdir
when
it produces Makefile.
Do not use the make
variable $<
, which expands to the
file name of the file in the source directory (found with VPATH
),
except in implicit rules. (An implicit rule is one such as ‘.c.o’,
which tells how to create a .o file from a .c file.) Some
versions of make do not set $<
in explicit rules; they
expand it to an empty value.
Instead, Make command lines should always refer to source files by prefixing them with ‘$(srcdir)/’. For example:
time.info: time.texinfo $(MAKEINFO) '$(srcdir)/time.texinfo'
You can put rules like the following in the top-level Makefile.in for a package to automatically update the configuration information when you change the configuration files. This example includes all of the optional files, such as aclocal.m4 and those related to configuration header files. Omit from the Makefile.in rules for any of these files that your package does not use.
The ‘$(srcdir)/’ prefix is included because of limitations in the
VPATH
mechanism.
The stamp- files are necessary because the timestamps of config.h.in and config.h are not changed if remaking them does not change their contents. This feature avoids unnecessary recompilation. You should include the file stamp-h.in your package's distribution, so that make considers config.h.in up to date. Don't use touch (see Limitations of Usual Tools); instead, use echo (using date would cause needless differences, hence CVS conflicts, etc.).
$(srcdir)/configure: configure.ac aclocal.m4 cd '$(srcdir)' && autoconf # autoheader might not change config.h.in, so touch a stamp file. $(srcdir)/config.h.in: stamp-h.in $(srcdir)/stamp-h.in: configure.ac aclocal.m4 cd '$(srcdir)' && autoheader echo timestamp > '$(srcdir)/stamp-h.in' config.h: stamp-h stamp-h: config.h.in config.status ./config.status Makefile: Makefile.in config.status ./config.status config.status: configure ./config.status --recheck
(Be careful if you copy these lines directly into your makefile, as you need to convert the indented lines to start with the tab character.)
In addition, you should use
AC_CONFIG_FILES([stamp-h], [echo timestamp > stamp-h])
so config.status ensures that config.h is considered up to
date. See Output, for more information about AC_OUTPUT
.
See config.status Invocation, for more examples of handling configuration-related dependencies.
When a package contains more than a few tests that define C preprocessor
symbols, the command lines to pass -D options to the compiler
can get quite long. This causes two problems. One is that the
make output is hard to visually scan for errors. More
seriously, the command lines can exceed the length limits of some
operating systems. As an alternative to passing -D options to
the compiler, configure scripts can create a C header file
containing ‘#define’ directives. The AC_CONFIG_HEADERS
macro selects this kind of output. Though it can be called anywhere
between AC_INIT
and AC_OUTPUT
, it is customary to call
it right after AC_INIT
.
The package should ‘#include’ the configuration header file before
any other header files, to prevent inconsistencies in declarations (for
example, if it redefines const
).
To provide for VPATH builds, remember to pass the C compiler a -I. option (or -I..; whichever directory contains config.h). Even if you use ‘#include "config.h"’, the preprocessor searches only the directory of the currently read file, i.e., the source directory, not the build directory.
With the appropriate -I option, you can use ‘#include <config.h>’. Actually, it's a good habit to use it, because in the rare case when the source directory contains another config.h, the build directory should be searched first.
This macro is one of the instantiating macros; see Configuration Actions. Make
AC_OUTPUT
create the file(s) in the blank-or-newline-separated list header containing C preprocessor#define
statements, and replace ‘@DEFS@’ in generated files with -DHAVE_CONFIG_H instead of the value ofDEFS
. The usual name for header is config.h.If header already exists and its contents are identical to what
AC_OUTPUT
would put in it, it is left alone. Doing this allows making some changes in the configuration without needlessly causing object files that depend on the header file to be recompiled.Usually the input file is named header.in; however, you can override the input file name by appending to header a colon-separated list of input files. Examples:
AC_CONFIG_HEADERS([config.h:config.hin]) AC_CONFIG_HEADERS([defines.h:defs.pre:defines.h.in:defs.post])Doing this allows you to keep your file names acceptable to DOS variants, or to prepend and/or append boilerplate to the file.
This macro is defined as the name of the first declared config header and undefined if no config headers have been declared up to this point. A third-party macro may, for example, require use of a config header without invoking AC_CONFIG_HEADERS twice, like this:
AC_CONFIG_COMMANDS_PRE( [m4_ifndef([AH_HEADER], [AC_CONFIG_HEADERS([config.h])])])
See Configuration Actions, for more details on header.
Your distribution should contain a template file that looks as you want
the final header file to look, including comments, with #undef
statements which are used as hooks. For example, suppose your
configure.ac makes these calls:
AC_CONFIG_HEADERS([conf.h]) AC_CHECK_HEADERS([unistd.h])
Then you could have code like the following in conf.h.in. On systems that have unistd.h, configure defines ‘HAVE_UNISTD_H’ to 1. On other systems, the whole line is commented out (in case the system predefines that symbol).
/* Define as 1 if you have unistd.h. */ #undef HAVE_UNISTD_H
Pay attention that ‘#undef’ is in the first column, and there is nothing after ‘HAVE_UNISTD_H’, not even white space. You can then decode the configuration header using the preprocessor directives:
#include <conf.h> #if HAVE_UNISTD_H # include <unistd.h> #else /* We are in trouble. */ #endif
The use of old form templates, with ‘#define’ instead of ‘#undef’ is strongly discouraged. Similarly with old templates with comments on the same line as the ‘#undef’. Anyway, putting comments in preprocessor macros has never been a good idea.
Since it is a tedious task to keep a template header up to date, you may use autoheader to generate it, see autoheader Invocation.
The autoheader program can create a template file of C
‘#define’ statements for configure to use. If
configure.ac invokes AC_CONFIG_HEADERS(
file)
,
autoheader creates file.in; if multiple file
arguments are given, the first one is used. Otherwise,
autoheader creates config.h.in.
In order to do its job, autoheader needs you to document all
of the symbols that you might use. Typically this is done via an
AC_DEFINE
or AC_DEFINE_UNQUOTED
call whose first argument
is a literal symbol and whose third argument describes the symbol
(see Defining Symbols). Alternatively, you can use
AH_TEMPLATE
(see Autoheader Macros), or you can supply a
suitable input file for a subsequent configuration header file.
Symbols defined by Autoconf's builtin tests are already documented properly;
you need to document only those that you
define yourself.
You might wonder why autoheader is needed: after all, why would configure need to “patch” a config.h.in to produce a config.h instead of just creating config.h from scratch? Well, when everything rocks, the answer is just that we are wasting our time maintaining autoheader: generating config.h directly is all that is needed. When things go wrong, however, you'll be thankful for the existence of autoheader.
The fact that the symbols are documented is important in order to check that config.h makes sense. The fact that there is a well-defined list of symbols that should be defined (or not) is also important for people who are porting packages to environments where configure cannot be run: they just have to fill in the blanks.
But let's come back to the point: the invocation of autoheader...
If you give autoheader an argument, it uses that file instead of configure.ac and writes the header file to the standard output instead of to config.h.in. If you give autoheader an argument of -, it reads the standard input instead of configure.ac and writes the header file to the standard output.
autoheader accepts the following options:
autoheader scans configure.ac and figures out which C
preprocessor symbols it might define. It knows how to generate
templates for symbols defined by AC_CHECK_HEADERS
,
AC_CHECK_FUNCS
etc., but if you AC_DEFINE
any additional
symbol, you must define a template for it. If there are missing
templates, autoheader fails with an error message.
The simplest way to create a template for a symbol is to supply the description argument to an ‘AC_DEFINE(symbol)’; see Defining Symbols. You may also use one of the following macros.
Tell autoheader to include the template as-is in the header template file. This template is associated with the key, which is used to sort all the different templates and guarantee their uniqueness. It should be a symbol that can be defined via
AC_DEFINE
.For example:
AH_VERBATIM([_GNU_SOURCE], [/* Enable GNU extensions on systems that have them. */ #ifndef _GNU_SOURCE # define _GNU_SOURCE #endif])
Tell autoheader to generate a template for key. This macro generates standard templates just like
AC_DEFINE
when a description is given.For example:
AH_TEMPLATE([CRAY_STACKSEG_END], [Define to one of _getb67, GETB67, getb67 for Cray-2 and Cray-YMP systems. This function is required for alloca.c support on those systems.])generates the following template, with the description properly justified.
/* Define to one of _getb67, GETB67, getb67 for Cray-2 and Cray-YMP systems. This function is required for alloca.c support on those systems. */ #undef CRAY_STACKSEG_END
You can execute arbitrary commands before, during, and after
config.status is run. The three following macros accumulate the
commands to run when they are called multiple times.
AC_CONFIG_COMMANDS
replaces the obsolete macro
AC_OUTPUT_COMMANDS
; see Obsolete Macros, for details.
Specify additional shell commands to run at the end of config.status, and shell commands to initialize any variables from configure. Associate the commands with tag. Since typically the cmds create a file, tag should naturally be the name of that file. If needed, the directory hosting tag is created. This macro is one of the instantiating macros; see Configuration Actions.
Here is an unrealistic example:
fubar=42 AC_CONFIG_COMMANDS([fubar], [echo this is extra $fubar, and so on.], [fubar=$fubar])Here is a better one:
AC_CONFIG_COMMANDS([timestamp], [date >timestamp])
The following two macros look similar, but in fact they are not of the same breed: they are executed directly by configure, so you cannot use config.status to rerun them.
Execute the cmds right before creating config.status.
This macro presents the last opportunity to call
AC_SUBST
,AC_DEFINE
, orAC_CONFIG_FOOS
macros.
You may find it convenient to create links whose destinations depend upon
results of tests. One can use AC_CONFIG_COMMANDS
but the
creation of relative symbolic links can be delicate when the package is
built in a directory different from the source directory.
Make
AC_OUTPUT
link each of the existing files source to the corresponding link name dest. Makes a symbolic link if possible, otherwise a hard link if possible, otherwise a copy. The dest and source names should be relative to the top level source or build directory. This macro is one of the instantiating macros; see Configuration Actions.For example, this call:
AC_CONFIG_LINKS([host.h:config/$machine.h object.h:config/$obj_format.h])creates in the current directory host.h as a link to srcdir/config/$machine.h, and object.h as a link to srcdir/config/$obj_format.h.
The tempting value ‘.’ for dest is invalid: it makes it impossible for ‘config.status’ to guess the links to establish.
One can then run:
./config.status host.h object.hto create the links.
In most situations, calling AC_OUTPUT
is sufficient to produce
makefiles in subdirectories. However, configure scripts
that control more than one independent package can use
AC_CONFIG_SUBDIRS
to run configure scripts for other
packages in subdirectories.
Make
AC_OUTPUT
run configure in each subdirectory dir in the given blank-or-newline-separated list. Each dir should be a literal, i.e., please do not use:if test "$package_foo_enabled" = yes; then $my_subdirs="$my_subdirs foo" fi AC_CONFIG_SUBDIRS([$my_subdirs])because this prevents ‘./configure --help=recursive’ from displaying the options of the package
foo
. Instead, you should write:if test "$package_foo_enabled" = yes; then AC_CONFIG_SUBDIRS([foo]) fiIf a given dir is not found, an error is reported: if the subdirectory is optional, write:
if test -d "$srcdir/foo"; then AC_CONFIG_SUBDIRS([foo]) fiIf a given dir contains configure.gnu, it is run instead of configure. This is for packages that might use a non-Autoconf script Configure, which can't be called through a wrapper configure since it would be the same file on case-insensitive file systems. Likewise, if a dir contains configure.in but no configure, the Cygnus configure script found by
AC_CONFIG_AUX_DIR
is used.The subdirectory configure scripts are given the same command line options that were given to this configure script, with minor changes if needed, which include:
- adjusting a relative name for the cache file;
- adjusting a relative name for the source directory;
- propagating the current value of
$prefix
, including if it was defaulted, and if the default values of the top level and of the subdirectory configure differ.This macro also sets the output variable
subdirs
to the list of directories ‘dir ...’. Make rules can use this variable to determine which subdirectories to recurse into.This macro may be called multiple times.
By default, configure sets the prefix for files it installs to /usr/local. The user of configure can select a different prefix using the --prefix and --exec-prefix options. There are two ways to change the default: when creating configure, and when running it.
Some software packages might want to install in a directory other than
/usr/local by default. To accomplish that, use the
AC_PREFIX_DEFAULT
macro.
Set the default installation prefix to prefix instead of /usr/local.
It may be convenient for users to have configure guess the
installation prefix from the location of a related program that they
have already installed. If you wish to do that, you can call
AC_PREFIX_PROGRAM
.
If the user did not specify an installation prefix (using the --prefix option), guess a value for it by looking for program in PATH, the way the shell does. If program is found, set the prefix to the parent of the directory containing program, else default the prefix as described above (/usr/local or
AC_PREFIX_DEFAULT
). For example, if program isgcc
and the PATH contains /usr/local/gnu/bin/gcc, set the prefix to /usr/local/gnu.
These macros test for particular system features that packages might need or want to use. If you need to test for a kind of feature that none of these macros check for, you can probably do it by calling primitive test macros with appropriate arguments (see Writing Tests).
These tests print messages telling the user which feature they're checking for, and what they find. They cache their results for future configure runs (see Caching Results).
Some of these macros set output variables. See Makefile Substitutions, for how to get their values. The phrase “define name” is used below as a shorthand to mean “define the C preprocessor symbol name to the value 1”. See Defining Symbols, for how to get those symbol definitions into your program.
Much effort has been expended to make Autoconf easy to learn. The most obvious way to reach this goal is simply to enforce standard interfaces and behaviors, avoiding exceptions as much as possible. Because of history and inertia, unfortunately, there are still too many exceptions in Autoconf; nevertheless, this section describes some of the common rules.
All the generic macros that AC_DEFINE
a symbol as a result of
their test transform their argument values to a standard alphabet.
First, argument is converted to upper case and any asterisks
(‘*’) are each converted to ‘P’. Any remaining characters
that are not alphanumeric are converted to underscores.
For instance,
AC_CHECK_TYPES([struct $Expensive*])
defines the symbol ‘HAVE_STRUCT__EXPENSIVEP’ if the check succeeds.
Several tests depend upon a set of header files. Since these headers are not universally available, tests actually have to provide a set of protected includes, such as:
#if TIME_WITH_SYS_TIME # include <sys/time.h> # include <time.h> #else # if HAVE_SYS_TIME_H # include <sys/time.h> # else # include <time.h> # endif #endif
Unless you know exactly what you are doing, you should avoid using unconditional includes, and check the existence of the headers you include beforehand (see Header Files).
Most generic macros use the following macro to provide the default set of includes:
Expand to include-directives if defined, otherwise to:
#include <stdio.h> #if HAVE_SYS_TYPES_H # include <sys/types.h> #endif #if HAVE_SYS_STAT_H # include <sys/stat.h> #endif #if STDC_HEADERS # include <stdlib.h> # include <stddef.h> #else # if HAVE_STDLIB_H # include <stdlib.h> # endif #endif #if HAVE_STRING_H # if !STDC_HEADERS && HAVE_MEMORY_H # include <memory.h> # endif # include <string.h> #endif #if HAVE_STRINGS_H # include <strings.h> #endif #if HAVE_INTTYPES_H # include <inttypes.h> #endif #if HAVE_STDINT_H # include <stdint.h> #endif #if HAVE_UNISTD_H # include <unistd.h> #endifIf the default includes are used, then check for the presence of these headers and their compatibility, i.e., you don't need to run
AC_HEADER_STDC
, nor check for stdlib.h etc.These headers are checked for in the same order as they are included. For instance, on some systems string.h and strings.h both exist, but conflict. Then
HAVE_STRING_H
is defined, notHAVE_STRINGS_H
.
These macros check for the presence or behavior of particular programs. They are used to choose between several alternative programs and to decide what to do once one has been chosen. If there is no macro specifically defined to check for a program you need, and you don't need to check for any special properties of it, then you can use one of the general program-check macros.
These macros check for particular programs—whether they exist, and in some cases whether they support certain features.
Check for
gawk
,mawk
,nawk
, andawk
, in that order, and set output variableAWK
to the first one that is found. It triesgawk
first because that is reported to be the best implementation.
Look for the best available
grep
orggrep
that accepts the longest input lines possible, and that supports multiple -e options. Set the output variableGREP
to whatever is chosen. See Limitations of Usual Tools, for more information about portability problems with the grep command family.
Check whether
$GREP -E
works, or else look for the best availableegrep
orgegrep
that accepts the longest input lines possible. Set the output variableEGREP
to whatever is chosen.
Check whether
$GREP -F
works, or else look for the best availablefgrep
orgfgrep
that accepts the longest input lines possible. Set the output variableFGREP
to whatever is chosen.
Set output variable
INSTALL
to the name of a BSD-compatible install program, if one is found in the current PATH. Otherwise, setINSTALL
to ‘dir/install-sh -c’, checking the directories specified toAC_CONFIG_AUX_DIR
(or its default directories) to determine dir (see Output). Also set the variablesINSTALL_PROGRAM
andINSTALL_SCRIPT
to ‘${INSTALL}’ andINSTALL_DATA
to ‘${INSTALL} -m 644’.This macro screens out various instances of install known not to work. It prefers to find a C program rather than a shell script, for speed. Instead of install-sh, it can also use install.sh, but that name is obsolete because some make programs have a rule that creates install from it if there is no makefile.
Autoconf comes with a copy of install-sh that you can use. If you use
AC_PROG_INSTALL
, you must include either install-sh or install.sh in your distribution; otherwise configure produces an error message saying it can't find them—even if the system you're on has a good install program. This check is a safety measure to prevent you from accidentally leaving that file out, which would prevent your package from installing on systems that don't have a BSD-compatible install program.If you need to use your own installation program because it has features not found in standard install programs, there is no reason to use
AC_PROG_INSTALL
; just put the file name of your program into your Makefile.in files.
Set output variable
MKDIR_P
to a program that ensures that for each argument, a directory named by this argument exists, creating it and its parent directories if needed, and without race conditions when two instances of the program attempt to make the same directory at nearly the same time.This macro uses the ‘mkdir -p’ command if possible. Otherwise, it falls back on invoking install-sh with the -d option, so your package should contain install-sh as described under
AC_PROG_INSTALL
. An install-sh file that predates Autoconf 2.60 or Automake 1.10 is vulnerable to race conditions, so if you want to support parallel installs from different packages into the same directory you need to make sure you have an up-to-date install-sh. In particular, be careful about using ‘autoreconf -if’ if your Automake predates Automake 1.10.This macro is related to the
AS_MKDIR_P
macro (see Programming in M4sh), but it sets an output variable intended for use in other files, whereasAS_MKDIR_P
is intended for use in scripts like configure. Also,AS_MKDIR_P
does not accept options, butMKDIR_P
supports the -m option, e.g., a makefile might invoke$(MKDIR_P) -m 0 dir
to create an inaccessible directory, and conversely a makefile should use$(MKDIR_P) -- $(FOO)
if FOO might yield a value that begins with ‘-’. Finally,AS_MKDIR_P
does not check for race condition vulnerability, whereasAC_PROG_MKDIR_P
does.
If
flex
is found, set output variableLEX
to ‘flex’ andLEXLIB
to -lfl, if that library is in a standard place. Otherwise setLEX
to ‘lex’ andLEXLIB
to -ll.Define
YYTEXT_POINTER
ifyytext
is a ‘char *’ instead of a ‘char []’. Also set output variableLEX_OUTPUT_ROOT
to the base of the file name that the lexer generates; usually lex.yy, but sometimes something else. These results vary according to whetherlex
orflex
is being used.You are encouraged to use Flex in your sources, since it is both more pleasant to use than plain Lex and the C source it produces is portable. In order to ensure portability, however, you must either provide a function
yywrap
or, if you don't use it (e.g., your scanner has no ‘#include’-like feature), simply include a ‘%noyywrap’ statement in the scanner's source. Once this done, the scanner is portable (unless you felt free to use nonportable constructs) and does not depend on any library. In this case, and in this case only, it is suggested that you use this Autoconf snippet:AC_PROG_LEX if test "$LEX" != flex; then LEX="$SHELL $missing_dir/missing flex" AC_SUBST([LEX_OUTPUT_ROOT], [lex.yy]) AC_SUBST([LEXLIB], ['']) fiThe shell script missing can be found in the Automake distribution.
To ensure backward compatibility, Automake's
AM_PROG_LEX
invokes (indirectly) this macro twice, which causes an annoying but benign “AC_PROG_LEX
invoked multiple times” warning. Future versions of Automake will fix this issue; meanwhile, just ignore this message.As part of running the test, this macro may delete any file in the configuration directory named lex.yy.c or lexyy.c.
If ‘ln -s’ works on the current file system (the operating system and file system support symbolic links), set the output variable
LN_S
to ‘ln -s’; otherwise, if ‘ln’ works, setLN_S
to ‘ln’, and otherwise set it to ‘cp -p’.If you make a link in a directory other than the current directory, its meaning depends on whether ‘ln’ or ‘ln -s’ is used. To safely create links using ‘$(LN_S)’, either find out which form is used and adjust the arguments, or always invoke
ln
in the directory where the link is to be created.In other words, it does not work to do:
$(LN_S) foo /x/barInstead, do:
(cd /x && $(LN_S) foo bar)
Set output variable
RANLIB
to ‘ranlib’ ifranlib
is found, and otherwise to ‘:’ (do nothing).
Set output variable
SED
to a Sed implementation that conforms to Posix and does not have arbitrary length limits. Report an error if no acceptable Sed is found. See Limitations of Usual Tools, for more information about portability problems with Sed.
If
bison
is found, set output variableYACC
to ‘bison -y’. Otherwise, ifbyacc
is found, setYACC
to ‘byacc’. Otherwise setYACC
to ‘yacc’.
These macros are used to find programs not covered by the “particular” test macros. If you need to check the behavior of a program as well as find out whether it is present, you have to write your own test for it (see Writing Tests). By default, these macros use the environment variable PATH. If you need to check for a program that might not be in the user's PATH, you can pass a modified path to use instead, like this:
AC_PATH_PROG([INETD], [inetd], [/usr/libexec/inetd], [$PATH:/usr/libexec:/usr/sbin:/usr/etc:/etc])
You are strongly encouraged to declare the variable passed to
AC_CHECK_PROG
etc. as precious, See Setting Output Variables,
AC_ARG_VAR
, for more details.
Check whether program prog-to-check-for exists in PATH. If it is found, set variable to value-if-found, otherwise to value-if-not-found, if given. Always pass over reject (an absolute file name) even if it is the first found in the search path; in that case, set variable using the absolute file name of the prog-to-check-for found that is not reject. If variable was already set, do nothing. Calls
AC_SUBST
for variable.
Check for each program in the blank-separated list progs-to-check-for existing in the PATH. If one is found, set variable to the name of that program. Otherwise, continue checking the next program in the list. If none of the programs in the list are found, set variable to value-if-not-found; if value-if-not-found is not specified, the value of variable is not changed. Calls
AC_SUBST
for variable.
Like
AC_CHECK_PROG
, but first looks for prog-to-check-for with a prefix of the target type as determined byAC_CANONICAL_TARGET
, followed by a dash (see Canonicalizing). If the tool cannot be found with a prefix, and if the build and target types are equal, then it is also searched for without a prefix.As noted in Specifying the system type, the target is rarely specified, because most of the time it is the same as the host: it is the type of system for which any compiler tool in the package produces code. What this macro looks for is, for example, a tool (assembler, linker, etc.) that the compiler driver (gcc for the GNU C Compiler) uses to produce objects, archives or executables.
Like
AC_CHECK_PROG
, but first looks for prog-to-check-for with a prefix of the host type as determined byAC_CANONICAL_HOST
, followed by a dash (see Canonicalizing). For example, if the user runs ‘configure --host=i386-gnu’, then this call:AC_CHECK_TOOL([RANLIB], [ranlib], [:])sets
RANLIB
to i386-gnu-ranlib if that program exists in PATH, or otherwise to ‘ranlib’ if that program exists in PATH, or to ‘:’ if neither program exists.In the future, when cross-compiling this macro will only accept program names that are prefixed with the host type. For more information, see Specifying the system type.
Like
AC_CHECK_TARGET_TOOL
, each of the tools in the list progs-to-check-for are checked with a prefix of the target type as determined byAC_CANONICAL_TARGET
, followed by a dash (see Canonicalizing). If none of the tools can be found with a prefix, and if the build and target types are equal, then the first one without a prefix is used. If a tool is found, set variable to the name of that program. If none of the tools in the list are found, set variable to value-if-not-found; if value-if-not-found is not specified, the value of variable is not changed. CallsAC_SUBST
for variable.
Like
AC_CHECK_TOOL
, each of the tools in the list progs-to-check-for are checked with a prefix of the host type as determined byAC_CANONICAL_HOST
, followed by a dash (see Canonicalizing). If none of the tools can be found with a prefix, then the first one without a prefix is used. If a tool is found, set variable to the name of that program. If none of the tools in the list are found, set variable to value-if-not-found; if value-if-not-found is not specified, the value of variable is not changed. CallsAC_SUBST
for variable.In the future, when cross-compiling this macro will not accept program names that are not prefixed with the host type.
Like
AC_CHECK_PROG
, but set variable to the absolute name of prog-to-check-for if found.
Like
AC_CHECK_PROGS
, but if any of progs-to-check-for are found, set variable to the absolute name of the program found.
Like
AC_CHECK_TARGET_TOOL
, but set variable to the absolute name of the program if it is found.
Like
AC_CHECK_TOOL
, but set variable to the absolute name of the program if it is found.In the future, when cross-compiling this macro will not accept program names that are not prefixed with the host type.
You might also need to check for the existence of files. Before using these macros, ask yourself whether a runtime test might not be a better solution. Be aware that, like most Autoconf macros, they test a feature of the host machine, and therefore, they die when cross-compiling.
Check whether file file exists on the native system. If it is found, execute action-if-found, otherwise do action-if-not-found, if given.
Executes
AC_CHECK_FILE
once for each file listed in files. Additionally, defines ‘HAVE_file’ (see Standard Symbols) for each file found.
The following macros check for the presence of certain C, C++, or Fortran library archive files.
Test whether the library library is available by trying to link a test program that calls function function with the library. function should be a function provided by the library. Use the base name of the library; e.g., to check for -lmp, use ‘mp’ as the library argument.
action-if-found is a list of shell commands to run if the link with the library succeeds; action-if-not-found is a list of shell commands to run if the link fails. If action-if-found is not specified, the default action prepends -llibrary to
LIBS
and defines ‘HAVE_LIBlibrary’ (in all capitals). This macro is intended to support buildingLIBS
in a right-to-left (least-dependent to most-dependent) fashion such that library dependencies are satisfied as a natural side effect of consecutive tests. Linkers are sensitive to library ordering so the order in whichLIBS
is generated is important to reliable detection of libraries.If linking with library results in unresolved symbols that would be resolved by linking with additional libraries, give those libraries as the other-libraries argument, separated by spaces: e.g., -lXt -lX11. Otherwise, this macro fails to detect that library is present, because linking the test program always fails with unresolved symbols. The other-libraries argument should be limited to cases where it is desirable to test for one library in the presence of another that is not already in
LIBS
.
AC_CHECK_LIB
requires some care in usage, and should be avoided in some common cases. Many standard functions likegethostbyname
appear the standard C library on some hosts, and in special libraries likensl
on other hosts. On some hosts the special libraries contain variant implementations that you may not want to use. These days it is normally better to useAC_SEARCH_LIBS([gethostbyname], [nsl])
instead ofAC_CHECK_LIB([nsl], [gethostbyname])
.
Search for a library defining function if it's not already available. This equates to calling ‘AC_LINK_IFELSE([AC_LANG_CALL([], [function])])’ first with no libraries, then for each library listed in search-libs.
Add -llibrary to
LIBS
for the first library found to contain function, and run action-if-found. If the function is not found, run action-if-not-found.If linking with library results in unresolved symbols that would be resolved by linking with additional libraries, give those libraries as the other-libraries argument, separated by spaces: e.g., -lXt -lX11. Otherwise, this macro fails to detect that function is present, because linking the test program always fails with unresolved symbols.
The following macros check for particular C library functions. If there is no macro specifically defined to check for a function you need, and you don't need to check for any special properties of it, then you can use one of the general function-check macros.
Most usual functions can either be missing, or be buggy, or be limited on some architectures. This section tries to make an inventory of these portability issues. By definition, this list always requires additions. Please help us keeping it as complete as possible.
exit
exit
returned int
.
This is because exit
predates void
, and there was a long
tradition of it returning int
.
On current hosts, the problem more likely is that exit
is not
declared, due to C++ problems of some sort or another. For this reason
we suggest that test programs not invoke exit
, but return from
main
instead.
free
free (NULL)
does nothing, but
some old systems don't support this (e.g., NextStep).
isinf
isnan
isinf
and isnan
are
macros. On some systems just macros are available
(e.g., HP-UX and Solaris 10), on
some systems both macros and functions (e.g., glibc 2.3.2), and on some
systems only functions (e.g., IRIX 6 and Solaris 9). In some cases
these functions are declared in nonstandard headers like
<sunmath.h>
and defined in non-default libraries like
-lm or -lsunmath.
The C99 isinf
and isnan
macros work correctly with
long double
arguments, but pre-C99 systems that use functions
typically assume double
arguments. On such a system,
isinf
incorrectly returns true for a finite long double
argument that is outside the range of double
.
To work around this porting mess, you can use code like the following.
#include <math.h> #ifndef isnan # define isnan(x) \ (sizeof (x) == sizeof (long double) ? isnan_ld (x) \ : sizeof (x) == sizeof (double) ? isnan_d (x) \ : isnan_f (x)) static inline int isnan_f (float x) { return x != x; } static inline int isnan_d (double x) { return x != x; } static inline int isnan_ld (long double x) { return x != x; } #endif #ifndef isinf # define isinf(x) \ (sizeof (x) == sizeof (long double) ? isinf_ld (x) \ : sizeof (x) == sizeof (double) ? isinf_d (x) \ : isinf_f (x)) static inline int isinf_f (float x) { return isnan (x - x); } static inline int isinf_d (double x) { return isnan (x - x); } static inline int isinf_ld (long double x) { return isnan (x - x); } #endif
Use AC_C_INLINE
(see C Compiler) so that this code works on
compilers that lack the inline
keyword. Some optimizing
compilers mishandle these definitions, but systems with that bug
typically have missing or broken isnan
functions anyway, so it's
probably not worth worrying about.
malloc
malloc (0)
is implementation
dependent. It may either return NULL
(e.g., OSF 4) or
non-NULL
(e.g., GNU C Library). AC_FUNC_MALLOC
can be used to insist on non-NULL
(see Particular Functions).
putenv
setenv
to putenv
; among other things,
putenv
is not required of all Posix implementations, but
setenv
is.
Posix specifies that putenv
puts the given string directly in
environ
, but some systems make a copy of it instead (e.g.,
glibc 2.0, or BSD). And when a copy is made, unsetenv
might
not free it, causing a memory leak (e.g., FreeBSD 4).
On some systems putenv ("FOO")
removes ‘FOO’ from the
environment, but this is not standard usage and it dumps core
on some systems (e.g., AIX).
On MinGW, a call putenv ("FOO=")
removes ‘FOO’ from the
environment, rather than inserting it with an empty value.
realloc
realloc (NULL, size)
is equivalent
to malloc (size)
, but some old systems don't support this (e.g.,
NextStep).
signal
handlersignal
takes a handler function with a return type of
void
, but some old systems required int
instead. Any
actual int
value returned is not used; this is only a
difference in the function prototype demanded.
All systems we know of in current use return void
. The
int
was to support K&R C, where of course void
is not
available. AC_TYPE_SIGNAL
(see Particular Types) can be
used to establish the correct type in all cases.
snprintf
snprintf
and vsnprintf
truncate the output and return the number of bytes that ought to have
been produced. Some older systems return the truncated length (e.g.,
GNU C Library 2.0.x or irix 6.5), some a negative value
(e.g., earlier GNU C Library versions), and some the buffer
length without truncation (e.g., 32-bit Solaris 7). Also, some buggy
older systems ignore the length and overrun the buffer (e.g., 64-bit
Solaris 7).
sprintf
sprintf
and vsprintf
return the
number of bytes written. On some ancient systems (SunOS 4 for
instance) they return the buffer pointer instead, but these no
longer need to be worried about.
sscanf
sscanf
requires that its
input string be writable (though it doesn't actually change it). This
can be a problem when using gcc since it normally puts
constant strings in read-only memory (see Incompatibilities of GCC). Apparently in some cases even
having format strings read-only can be a problem.
strerror_r
strerror_r
returns an int
, but many
systems (e.g., GNU C Library version 2.2.4) provide a
different version returning a char *
. AC_FUNC_STRERROR_R
can detect which is in use (see Particular Functions).
strnlen
strnlen ("foobar", 0) = 0 strnlen ("foobar", 1) = 3 strnlen ("foobar", 2) = 2 strnlen ("foobar", 3) = 1 strnlen ("foobar", 4) = 0 strnlen ("foobar", 5) = 6 strnlen ("foobar", 6) = 6 strnlen ("foobar", 7) = 6 strnlen ("foobar", 8) = 6 strnlen ("foobar", 9) = 6
sysconf
_SC_PAGESIZE
is standard, but some older systems (e.g., HP-UX
9) have _SC_PAGE_SIZE
instead. This can be tested with
#ifdef
.
unlink
unlink
causes the given file to be
removed only after there are no more open file handles for it. Some
non-Posix hosts have trouble with this requirement, though,
and some DOS variants even corrupt the file system.
unsetenv
unsetenv
is not available, but a variable ‘FOO’
can be removed with a call putenv ("FOO=")
, as described under
putenv
above.
va_copy
va_copy
for copying
va_list
variables. It may be available in older environments
too, though possibly as __va_copy
(e.g., gcc in strict
pre-C99 mode). These can be tested with #ifdef
. A fallback to
memcpy (&dst, &src, sizeof (va_list))
gives maximum
portability.
va_list
va_list
is not necessarily just a pointer. It can be a
struct
(e.g., gcc on Alpha), which means NULL
is
not portable. Or it can be an array (e.g., gcc in some
PowerPC configurations), which means as a function parameter it can be
effectively call-by-reference and library routines might modify the
value back in the caller (e.g., vsnprintf
in the GNU C Library
2.1).
>>
>>
right shift of a signed type replicates the
high bit, giving a so-called “arithmetic” shift. But care should be
taken since Standard C doesn't require that behavior. On those
few processors without a native arithmetic shift (for instance Cray
vector systems) zero bits may be shifted in, the same as a shift of an
unsigned type.
/
These macros check for particular C functions—whether they exist, and in some cases how they respond when given certain arguments.
Check how to get
alloca
. Tries to get a builtin version by checking for alloca.h or the predefined C preprocessor macros__GNUC__
and_AIX
. If this macro finds alloca.h, it definesHAVE_ALLOCA_H
.If those attempts fail, it looks for the function in the standard C library. If any of those methods succeed, it defines
HAVE_ALLOCA
. Otherwise, it sets the output variableALLOCA
to ‘${LIBOBJDIR}alloca.o’ and definesC_ALLOCA
(so programs can periodically call ‘alloca (0)’ to garbage collect). This variable is separate fromLIBOBJS
so multiple programs can share the value ofALLOCA
without needing to create an actual library, in case only some of them use the code inLIBOBJS
. The ‘${LIBOBJDIR}’ prefix serves the same purpose as inLIBOBJS
(see AC_LIBOBJ vs LIBOBJS).This macro does not try to get
alloca
from the System V R3 libPW or the System V R4 libucb because those libraries contain some incompatible functions that cause trouble. Some versions do not even containalloca
or contain a buggy version. If you still want to use theiralloca
, usear
to extract alloca.o from them instead of compiling alloca.c.Source files that use
alloca
should start with a piece of code like the following, to declare it properly.#if HAVE_ALLOCA_H # include <alloca.h> #elif defined __GNUC__ # define alloca __builtin_alloca #elif defined _AIX # define alloca __alloca #elif defined _MSC_VER # include <malloc.h> # define alloca _alloca #else # include <stddef.h> # ifdef __cplusplus extern "C" # endif void *alloca (size_t); #endif
If the
chown
function is available and works (in particular, it should accept -1 foruid
andgid
), defineHAVE_CHOWN
.
If the
closedir
function does not return a meaningful value, defineCLOSEDIR_VOID
. Otherwise, callers ought to check its return value for an error indicator.Currently this test is implemented by running a test program. When cross compiling the pessimistic assumption that
closedir
does not return a meaningful value is made.This macro is obsolescent, as
closedir
returns a meaningful value on current systems. New programs need not use this macro.
If the
error_at_line
function is not found, require anAC_LIBOBJ
replacement of ‘error’.
If the
fnmatch
function conforms to Posix, defineHAVE_FNMATCH
. Detect common implementation bugs, for example, the bugs in Solaris 2.4.Unlike the other specific
AC_FUNC
macros,AC_FUNC_FNMATCH
does not replace a broken/missingfnmatch
. This is for historical reasons. SeeAC_REPLACE_FNMATCH
below.
Behave like
AC_REPLACE_FNMATCH
(replace) but also test whetherfnmatch
supports GNU extensions. Detect common implementation bugs, for example, the bugs in the GNU C Library 2.1.
This macro checks for the
fork
andvfork
functions. If a workingfork
is found, defineHAVE_WORKING_FORK
. This macro checks whetherfork
is just a stub by trying to run it.If vfork.h is found, define
HAVE_VFORK_H
. If a workingvfork
is found, defineHAVE_WORKING_VFORK
. Otherwise, definevfork
to befork
for backward compatibility with previous versions of autoconf. This macro checks for several known errors in implementations ofvfork
and considers the system to not have a workingvfork
if it detects any of them. It is not considered to be an implementation error if a child's invocation ofsignal
modifies the parent's signal handler, since child processes rarely change their signal handlers.Since this macro defines
vfork
only for backward compatibility with previous versions of autoconf you're encouraged to define it yourself in new code:#if !HAVE_WORKING_VFORK # define vfork fork #endif
If the
fseeko
function is available, defineHAVE_FSEEKO
. Define_LARGEFILE_SOURCE
if necessary to make the prototype visible on some systems (e.g., glibc 2.2). Otherwise linkage problems may occur when compiling withAC_SYS_LARGEFILE
on largefile-sensitive systems whereoff_t
does not default to a 64bit entity.
If the
getgroups
function is available and works (unlike on Ultrix 4.3, where ‘getgroups (0, 0)’ always fails), defineHAVE_GETGROUPS
. SetGETGROUPS_LIBS
to any libraries needed to get that function. This macro runsAC_TYPE_GETGROUPS
.
Check how to get the system load averages. To perform its tests properly, this macro needs the file getloadavg.c; therefore, be sure to set the
AC_LIBOBJ
replacement directory properly (see Generic Functions,AC_CONFIG_LIBOBJ_DIR
).If the system has the
getloadavg
function, defineHAVE_GETLOADAVG
, and setGETLOADAVG_LIBS
to any libraries necessary to get that function. Also addGETLOADAVG_LIBS
toLIBS
. Otherwise, require anAC_LIBOBJ
replacement for ‘getloadavg’ with source code in dir/getloadavg.c, and possibly define several other C preprocessor macros and output variables:
- Define
C_GETLOADAVG
.- Define
SVR4
,DGUX
,UMAX
, orUMAX4_3
if on those systems.- If nlist.h is found, define
HAVE_NLIST_H
.- If ‘struct nlist’ has an ‘n_un.n_name’ member, define
HAVE_STRUCT_NLIST_N_UN_N_NAME
. The obsolete symbolNLIST_NAME_UNION
is still defined, but do not depend upon it.- Programs may need to be installed set-group-ID (or set-user-ID) for
getloadavg
to work. In this case, defineGETLOADAVG_PRIVILEGED
, set the output variableNEED_SETGID
to ‘true’ (and otherwise to ‘false’), and setKMEM_GROUP
to the name of the group that should own the installed program.
Check for
getmntent
in the standard C library, and then in the sun, seq, and gen libraries, for unicos, irix 4, ptx, and UnixWare, respectively. Then, ifgetmntent
is available, defineHAVE_GETMNTENT
.
Define
GETPGRP_VOID
if it is an error to pass 0 togetpgrp
; this is the Posix behavior. On older BSD systems, you must pass 0 togetpgrp
, as it takes an argument and behaves like Posix'sgetpgid
.#if GETPGRP_VOID pid = getpgrp (); #else pid = getpgrp (0); #endifThis macro does not check whether
getpgrp
exists at all; if you need to work in that situation, first callAC_CHECK_FUNC
forgetpgrp
.This macro is obsolescent, as current systems have a
getpgrp
whose signature conforms to Posix. New programs need not use this macro.
If link is a symbolic link, then
lstat
should treat link/ the same as link/.. However, many olderlstat
implementations incorrectly ignore trailing slashes.It is safe to assume that if
lstat
incorrectly ignores trailing slashes, then other symbolic-link-aware functions likeunlink
also incorrectly ignore trailing slashes.If
lstat
behaves properly, defineLSTAT_FOLLOWS_SLASHED_SYMLINK
, otherwise require anAC_LIBOBJ
replacement oflstat
.
If the
malloc
function is compatible with the GNU C librarymalloc
(i.e., ‘malloc (0)’ returns a valid pointer), defineHAVE_MALLOC
to 1. Otherwise defineHAVE_MALLOC
to 0, ask for anAC_LIBOBJ
replacement for ‘malloc’, and definemalloc
torpl_malloc
so that the nativemalloc
is not used in the main project.Typically, the replacement file malloc.c should look like (note the ‘#undef malloc’):
#if HAVE_CONFIG_H # include <config.h> #endif #undef malloc #include <sys/types.h> void *malloc (); /* Allocate an N-byte block of memory from the heap. If N is zero, allocate a 1-byte block. */ void * rpl_malloc (size_t n) { if (n == 0) n = 1; return malloc (n); }
If the
memcmp
function is not available, or does not work on 8-bit data (like the one on SunOS 4.1.3), or fails when comparing 16 bytes or more and with at least one buffer not starting on a 4-byte boundary (such as the one on NeXT x86 OpenStep), require anAC_LIBOBJ
replacement for ‘memcmp’.This macro is obsolescent, as current systems have a working
memcmp
. New programs need not use this macro.
Define
HAVE_MBRTOWC
to 1 if the functionmbrtowc
and the typembstate_t
are properly declared.
If the
mktime
function is not available, or does not work correctly, require anAC_LIBOBJ
replacement for ‘mktime’. For the purposes of this test,mktime
should conform to the Posix standard and should be the inverse oflocaltime
.
If the
mmap
function exists and works correctly, defineHAVE_MMAP
. This checks only private fixed mapping of already-mapped memory.
If the obstacks are found, define
HAVE_OBSTACK
, else require anAC_LIBOBJ
replacement for ‘obstack’.
If the
realloc
function is compatible with the GNU C libraryrealloc
(i.e., ‘realloc (NULL, 0)’ returns a valid pointer), defineHAVE_REALLOC
to 1. Otherwise defineHAVE_REALLOC
to 0, ask for anAC_LIBOBJ
replacement for ‘realloc’, and definerealloc
torpl_realloc
so that the nativerealloc
is not used in the main project. SeeAC_FUNC_MALLOC
for details.
Determines the correct type to be passed for each of the
select
function's arguments, and defines those types inSELECT_TYPE_ARG1
,SELECT_TYPE_ARG234
, andSELECT_TYPE_ARG5
respectively.SELECT_TYPE_ARG1
defaults to ‘int’,SELECT_TYPE_ARG234
defaults to ‘int *’, andSELECT_TYPE_ARG5
defaults to ‘struct timeval *’.This macro is obsolescent, as current systems have a
select
whose signature conforms to Posix. New programs need not use this macro.
If
setpgrp
takes no argument (the Posix version), defineSETPGRP_VOID
. Otherwise, it is the BSD version, which takes two process IDs as arguments. This macro does not check whethersetpgrp
exists at all; if you need to work in that situation, first callAC_CHECK_FUNC
forsetpgrp
.This macro is obsolescent, as current systems have a
setpgrp
whose signature conforms to Posix. New programs need not use this macro.
Determine whether
stat
orlstat
have the bug that it succeeds when given the zero-length file name as argument. Thestat
andlstat
from SunOS 4.1.4 and the Hurd (as of 1998-11-01) do this.If it does, then define
HAVE_STAT_EMPTY_STRING_BUG
(orHAVE_LSTAT_EMPTY_STRING_BUG
) and ask for anAC_LIBOBJ
replacement of it.These macros are obsolescent, as no current systems have the bug. New programs need not use these macros.
If
setvbuf
takes the buffering type as its second argument and the buffer pointer as the third, instead of the other way around, defineSETVBUF_REVERSED
.This macro is obsolescent, as no current systems have the bug. New programs need not use this macro.
If the
strcoll
function exists and works correctly, defineHAVE_STRCOLL
. This does a bit more than ‘AC_CHECK_FUNCS(strcoll)’, because some systems have incorrect definitions ofstrcoll
that should not be used.
If
strerror_r
is available, defineHAVE_STRERROR_R
, and if it is declared, defineHAVE_DECL_STRERROR_R
. If it returns achar *
message, defineSTRERROR_R_CHAR_P
; otherwise it returns anint
error number. The Thread-Safe Functions option of Posix requiresstrerror_r
to returnint
, but many systems (including, for example, version 2.2.4 of the GNU C Library) return achar *
value that is not necessarily equal to the buffer argument.
Check for
strftime
in the intl library, for SCO Unix. Then, ifstrftime
is available, defineHAVE_STRFTIME
.This macro is obsolescent, as no current systems require the intl library for
strftime
. New programs need not use this macro.
If the
strtod
function does not exist or doesn't work correctly, ask for anAC_LIBOBJ
replacement of ‘strtod’. In this case, because strtod.c is likely to need ‘pow’, set the output variablePOW_LIB
to the extra library needed.
If the
strnlen
function is not available, or is buggy (like the one from AIX 4.3), require anAC_LIBOBJ
replacement for it.
If ‘utime (file, NULL)’ sets file's timestamp to the present, define
HAVE_UTIME_NULL
.This macro is obsolescent, as all current systems have a
utime
that behaves this way. New programs need not use this macro.
If
vprintf
is found, defineHAVE_VPRINTF
. Otherwise, if_doprnt
is found, defineHAVE_DOPRNT
. (Ifvprintf
is available, you may assume thatvfprintf
andvsprintf
are also available.)This macro is obsolescent, as all current systems have
vprintf
. New programs need not use this macro.
If the
fnmatch
function does not conform to Posix (seeAC_FUNC_FNMATCH
), ask for itsAC_LIBOBJ
replacement.The files fnmatch.c, fnmatch_loop.c, and fnmatch_.h in the
AC_LIBOBJ
replacement directory are assumed to contain a copy of the source code of GNUfnmatch
. If necessary, this source code is compiled as anAC_LIBOBJ
replacement, and the fnmatch_.h file is linked to fnmatch.h so that it can be included in place of the system<fnmatch.h>
.
These macros are used to find functions not covered by the “particular”
test macros. If the functions might be in libraries other than the
default C library, first call AC_CHECK_LIB
for those libraries.
If you need to check the behavior of a function as well as find out
whether it is present, you have to write your own test for
it (see Writing Tests).
If C function function is available, run shell commands action-if-found, otherwise action-if-not-found. If you just want to define a symbol if the function is available, consider using
AC_CHECK_FUNCS
instead. This macro checks for functions with C linkage even whenAC_LANG(C++)
has been called, since C is more standardized than C++. (see Language Choice, for more information about selecting the language for checks.)
For each function enumerated in the blank-or-newline-separated argument list, define
HAVE_
function (in all capitals) if it is available. If action-if-found is given, it is additional shell code to execute when one of the functions is found. You can give it a value of ‘break’ to break out of the loop on the first match. If action-if-not-found is given, it is executed when one of the functions is not found.
For each function enumerated in the blank-or-newline-separated argument list, define
HAVE_
function (in all capitals) if it is available. This is a once-only variant ofAC_CHECK_FUNCS
. It generates the checking code at most once, so that configure is smaller and faster; but the checks cannot be conditionalized and are always done once, early during the configure run.
Autoconf follows a philosophy that was formed over the years by those who have struggled for portability: isolate the portability issues in specific files, and then program as if you were in a Posix environment. Some functions may be missing or unfixable, and your package must be ready to replace them.
Suitable replacements for many such problem functions are available from Gnulib (see Gnulib).
Specify that ‘function.c’ must be included in the executables to replace a missing or broken implementation of function.
Technically, it adds ‘function.$ac_objext’ to the output variable
LIBOBJS
if it is not already in, and callsAC_LIBSOURCE
for ‘function.c’. You should not directly changeLIBOBJS
, since this is not traceable.
Specify that file might be needed to compile the project. If you need to know what files might be needed by a configure.ac, you should trace
AC_LIBSOURCE
. file must be a literal.This macro is called automatically from
AC_LIBOBJ
, but you must call it explicitly if you pass a shell variable toAC_LIBOBJ
. In that case, since shell variables cannot be traced statically, you must pass toAC_LIBSOURCE
any possible files that the shell variable might causeAC_LIBOBJ
to need. For example, if you want to pass a variable$foo_or_bar
toAC_LIBOBJ
that holds either"foo"
or"bar"
, you should do:AC_LIBSOURCE([foo.c]) AC_LIBSOURCE([bar.c]) AC_LIBOBJ([$foo_or_bar])There is usually a way to avoid this, however, and you are encouraged to simply call
AC_LIBOBJ
with literal arguments.Note that this macro replaces the obsolete
AC_LIBOBJ_DECL
, with slightly different semantics: the old macro took the function name, e.g.,foo
, as its argument rather than the file name.
Like
AC_LIBSOURCE
, but accepts one or more files in a comma-separated M4 list. Thus, the above example might be rewritten:AC_LIBSOURCES([foo.c, bar.c]) AC_LIBOBJ([$foo_or_bar])
Specify that
AC_LIBOBJ
replacement files are to be found in directory, a name relative to the top level of the source tree. The replacement directory defaults to ., the top level directory, and the most typical value is lib, corresponding to ‘AC_CONFIG_LIBOBJ_DIR([lib])’.configure might need to know the replacement directory for the following reasons: (i) some checks use the replacement files, (ii) some macros bypass broken system headers by installing links to the replacement headers (iii) when used in conjunction with Automake, within each makefile, directory is used as a relative path from
$(top_srcdir)
to each object named inLIBOBJS
andLTLIBOBJS
, etc.
It is common to merely check for the existence of a function, and ask for its
AC_LIBOBJ
replacement if missing. The following macro is
a convenient shorthand.
Like
AC_CHECK_FUNCS
, but uses ‘AC_LIBOBJ(function)’ as action-if-not-found. You can declare your replacement function by enclosing the prototype in ‘#if !HAVE_function’. If the system has the function, it probably declares it in a header file you should be including, so you shouldn't redeclare it lest your declaration conflict.
The following macros check for the presence of certain C header files. If there is no macro specifically defined to check for a header file you need, and you don't need to check for any special properties of it, then you can use one of the general header-file check macros.
This section tries to collect knowledge about common headers, and the problems they cause. By definition, this list always requires additions. Please help us keeping it as complete as possible.
LLONG_MIN
,
LLONG_MAX
, and ULLONG_MAX
, but many almost-C99
environments (e.g., default GCC 4.0.2 + glibc 2.4) do not
define them.
AC_CHECK_HEADERS([sys/socket.h]) AC_CHECK_HEADERS([net/if.h], [], [], [#include <stdio.h> #if STDC_HEADERS # include <stdlib.h> # include <stddef.h> #else # if HAVE_STDLIB_H # include <stdlib.h> # endif #endif #if HAVE_SYS_SOCKET_H # include <sys/socket.h> #endif ])
AC_CHECK_HEADERS([sys/socket.h]) AC_CHECK_HEADERS([netinet/if_ether.h], [], [], [#include <stdio.h> #if STDC_HEADERS # include <stdlib.h> # include <stddef.h> #else # if HAVE_STDLIB_H # include <stdlib.h> # endif #endif #if HAVE_SYS_SOCKET_H # include <sys/socket.h> #endif ])
AC_CHECK_HEADERS([X11/extensions/scrnsaver.h], [], [], [[#include <X11/Xlib.h> ]])
These macros check for particular system header files—whether they exist, and in some cases whether they declare certain symbols.
Check whether to enable assertions in the style of assert.h. Assertions are enabled by default, but the user can override this by invoking configure with the --disable-assert option.
Check for the following header files. For the first one that is found and defines ‘DIR’, define the listed C preprocessor macro:
dirent.h HAVE_DIRENT_H
sys/ndir.h HAVE_SYS_NDIR_H
sys/dir.h HAVE_SYS_DIR_H
ndir.h HAVE_NDIR_H
The directory-library declarations in your source code should look something like the following:
#include <sys/types.h> #ifdef HAVE_DIRENT_H # include <dirent.h> # define NAMLEN(dirent) strlen ((dirent)->d_name) #else # define dirent direct # define NAMLEN(dirent) ((dirent)->d_namlen) # if HAVE_SYS_NDIR_H # include <sys/ndir.h> # endif # if HAVE_SYS_DIR_H # include <sys/dir.h> # endif # if HAVE_NDIR_H # include <ndir.h> # endif #endifUsing the above declarations, the program would declare variables to be of type
struct dirent
, notstruct direct
, and would access the length of a directory entry name by passing a pointer to astruct dirent
to theNAMLEN
macro.This macro also checks for the SCO Xenix dir and x libraries.
This macro is obsolescent, as all current systems with directory libraries have
<dirent.h>
. New programs need not use this macro.Also see
AC_STRUCT_DIRENT_D_INO
andAC_STRUCT_DIRENT_D_TYPE
(see Particular Structures).
If sys/types.h does not define
major
,minor
, andmakedev
, but sys/mkdev.h does, defineMAJOR_IN_MKDEV
; otherwise, if sys/sysmacros.h does, defineMAJOR_IN_SYSMACROS
.
Checks for header resolv.h, checking for prerequisites first. To properly use resolv.h, your code should contain something like the following:
#if HAVE_SYS_TYPES_H # include <sys/types.h> #endif #ifdef HAVE_NETINET_IN_H # include <netinet/in.h> /* inet_ functions / structs */ #endif #ifdef HAVE_ARPA_NAMESER_H # include <arpa/nameser.h> /* DNS HEADER struct */ #endif #ifdef HAVE_NETDB_H # include <netdb.h> #endif #include <resolv.h>
If the macros
S_ISDIR
,S_ISREG
, etc. defined in sys/stat.h do not work properly (returning false positives), defineSTAT_MACROS_BROKEN
. This is the case on Tektronix UTekV, Amdahl UTS and Motorola System V/88.This macro is obsolescent, as no current systems have the bug. New programs need not use this macro.
If stdbool.h exists and conforms to C99, define
HAVE_STDBOOL_H
to 1; if the type_Bool
is defined, defineHAVE__BOOL
to 1. To fulfill the C99 requirements, your system.h could contain the following code:#if HAVE_STDBOOL_H # include <stdbool.h> #else # if ! HAVE__BOOL # ifdef __cplusplus typedef bool _Bool; # else # define _Bool signed char # endif # endif # define bool _Bool # define false 0 # define true 1 # define __bool_true_false_are_defined 1 #endifAlternatively you can use the ‘stdbool’ package of Gnulib (see Gnulib); it packages the above code into a replacement header and contains a few other bells and whistles.
Define
STDC_HEADERS
if the system has C header files conforming to ANSI C89 (ISO C90). Specifically, this macro checks for stdlib.h, stdarg.h, string.h, and float.h; if the system has those, it probably has the rest of the C89 header files. This macro also checks whether string.h declaresmemchr
(and thus presumably the othermem
functions), whether stdlib.h declarefree
(and thus presumablymalloc
and other related functions), and whether the ctype.h macros work on characters with the high bit set, as the C standard requires.If you use this macro, your code can refer to
STDC_HEADERS
to determine whether the system has conforming header files (and probably C library functions).This macro is obsolescent, as current systems have conforming header files. New programs need not use this macro.
Nowadays string.h is part of the C standard and declares functions like
strcpy
, and strings.h is standardized by Posix and declares BSD functions likebcopy
; but historically, string functions were a major sticking point in this area. If you still want to worry about portability to ancient systems without standard headers, there is so much variation that it is probably easier to declare the functions you use than to figure out exactly what the system header files declare. Some ancient systems contained a mix of functions from the C standard and from BSD; some were mostly standard but lacked ‘memmove’; some defined the BSD functions as macros in string.h or strings.h; some had only the BSD functions but string.h; some declared the memory functions in memory.h, some in string.h; etc. It is probably sufficient to check for one string function and one memory function; if the library had the standard versions of those then it probably had most of the others. If you put the following in configure.ac:# This example is obsolescent. # Nowadays you can omit these macro calls. AC_HEADER_STDC AC_CHECK_FUNCS([strchr memcpy])then, in your code, you can use declarations like this:
/* This example is obsolescent. Nowadays you can just #include <string.h>. */ #if STDC_HEADERS # include <string.h> #else # if !HAVE_STRCHR # define strchr index # define strrchr rindex # endif char *strchr (), *strrchr (); # if !HAVE_MEMCPY # define memcpy(d, s, n) bcopy ((s), (d), (n)) # define memmove(d, s, n) bcopy ((s), (d), (n)) # endif #endifIf you use a function like
memchr
,memset
,strtok
, orstrspn
, which have no BSD equivalent, then macros don't suffice to port to ancient hosts; you must provide an implementation of each function. An easy way to incorporate your implementations only when needed (since the ones in system C libraries may be hand optimized) is to, takingmemchr
for example, put it in memchr.c and use ‘AC_REPLACE_FUNCS([memchr])’.
If sys/wait.h exists and is compatible with Posix, define
HAVE_SYS_WAIT_H
. Incompatibility can occur if sys/wait.h does not exist, or if it uses the old BSDunion wait
instead ofint
to store a status value. If sys/wait.h is not Posix compatible, then instead of including it, define the Posix macros with their usual interpretations. Here is an example:#include <sys/types.h> #if HAVE_SYS_WAIT_H # include <sys/wait.h> #endif #ifndef WEXITSTATUS # define WEXITSTATUS(stat_val) ((unsigned int) (stat_val) >> 8) #endif #ifndef WIFEXITED # define WIFEXITED(stat_val) (((stat_val) & 255) == 0) #endifThis macro is obsolescent, as current systems are compatible with Posix. New programs need not use this macro.
_POSIX_VERSION
is defined when unistd.h is included on
Posix systems. If there is no unistd.h, it is definitely
not a Posix system. However, some non-Posix systems do
have unistd.h.
The way to check whether the system supports Posix is:
#if HAVE_UNISTD_H # include <sys/types.h> # include <unistd.h> #endif #ifdef _POSIX_VERSION /* Code for Posix systems. */ #endif
If a program may include both time.h and sys/time.h, define
TIME_WITH_SYS_TIME
. On some ancient systems, sys/time.h included time.h, but time.h was not protected against multiple inclusion, so programs could not explicitly include both files. This macro is useful in programs that use, for example,struct timeval
as well asstruct tm
. It is best used in conjunction withHAVE_SYS_TIME_H
, which can be checked for usingAC_CHECK_HEADERS([sys/time.h])
.#if TIME_WITH_SYS_TIME # include <sys/time.h> # include <time.h> #else # if HAVE_SYS_TIME_H # include <sys/time.h> # else # include <time.h> # endif #endifThis macro is obsolescent, as current systems can include both files when they exist. New programs need not use this macro.
If the use of
TIOCGWINSZ
requires <sys/ioctl.h>, then defineGWINSZ_IN_SYS_IOCTL
. OtherwiseTIOCGWINSZ
can be found in <termios.h>.Use:
#if HAVE_TERMIOS_H # include <termios.h> #endif #if GWINSZ_IN_SYS_IOCTL # include <sys/ioctl.h> #endif
These macros are used to find system header files not covered by the “particular” test macros. If you need to check the contents of a header as well as find out whether it is present, you have to write your own test for it (see Writing Tests).
If the system header file header-file is compilable, execute shell commands action-if-found, otherwise execute action-if-not-found. If you just want to define a symbol if the header file is available, consider using
AC_CHECK_HEADERS
instead.For compatibility issues with older versions of Autoconf, please read below.
For each given system header file header-file in the blank-separated argument list that exists, define
HAVE_
header-file (in all capitals). If action-if-found is given, it is additional shell code to execute when one of the header files is found. You can give it a value of ‘break’ to break out of the loop on the first match. If action-if-not-found is given, it is executed when one of the header files is not found.For compatibility issues with older versions of Autoconf, please read below.
Previous versions of Autoconf merely checked whether the header was
accepted by the preprocessor. This was changed because the old test was
inappropriate for typical uses. Headers are typically used to compile,
not merely to preprocess, and the old behavior sometimes accepted
headers that clashed at compile-time. If you need to check whether a
header is preprocessable, you can use AC_PREPROC_IFELSE
(see Running the Preprocessor).
This scheme, which improves the robustness of the test, also requires that you make sure that headers that must be included before the header-file be part of the includes, (see Default Includes). If looking for bar.h, which requires that foo.h be included before if it exists, we suggest the following scheme:
AC_CHECK_HEADERS([foo.h]) AC_CHECK_HEADERS([bar.h], [], [], [#if HAVE_FOO_H # include <foo.h> # endif ])
The following variant generates smaller, faster configure
files if you do not need the full power of AC_CHECK_HEADERS
.
For each given system header file header-file in the blank-separated argument list that exists, define
HAVE_
header-file (in all capitals). This is a once-only variant ofAC_CHECK_HEADERS
. It generates the checking code at most once, so that configure is smaller and faster; but the checks cannot be conditionalized and are always done once, early during the configure run.
The following macros check for the declaration of variables and
functions. If there is no macro specifically defined to check for a
symbol you need, then you can use the general macros (see Generic Declarations) or, for more complex tests, you may use
AC_COMPILE_IFELSE
(see Running the Compiler).
There are no specific macros for declarations.
These macros are used to find declarations not covered by the “particular” test macros.
If symbol (a function or a variable) is not declared in includes and a declaration is needed, run the shell commands action-if-not-found, otherwise action-if-found. If no includes are specified, the default includes are used (see Default Includes).
This macro actually tests whether it is valid to use symbol as an r-value, not if it is really declared, because it is much safer to avoid introducing extra declarations when they are not needed.
For each of the symbols (comma-separated list), define
HAVE_DECL_
symbol (in all capitals) to ‘1’ if symbol is declared, otherwise to ‘0’. If action-if-not-found is given, it is additional shell code to execute when one of the function declarations is needed, otherwise action-if-found is executed.This macro uses an M4 list as first argument:
AC_CHECK_DECLS([strdup]) AC_CHECK_DECLS([strlen]) AC_CHECK_DECLS([malloc, realloc, calloc, free])Unlike the other ‘AC_CHECK_*S’ macros, when a symbol is not declared,
HAVE_DECL_
symbol is defined to ‘0’ instead of leavingHAVE_DECL_
symbol undeclared. When you are sure that the check was performed, useHAVE_DECL_
symbol just like any other result of Autoconf:#if !HAVE_DECL_SYMBOL extern char *symbol; #endifIf the test may have not been performed, however, because it is safer not to declare a symbol than to use a declaration that conflicts with the system's one, you should use:
#if defined HAVE_DECL_MALLOC && !HAVE_DECL_MALLOC void *malloc (size_t *s); #endifYou fall into the second category only in extreme situations: either your files may be used without being configured, or they are used during the configuration. In most cases the traditional approach is enough.
For each of the symbols (comma-separated list), define
HAVE_DECL_
symbol (in all capitals) to ‘1’ if symbol is declared in the default include files, otherwise to ‘0’. This is a once-only variant ofAC_CHECK_DECLS
. It generates the checking code at most once, so that configure is smaller and faster; but the checks cannot be conditionalized and are always done once, early during the configure run.
The following macros check for the presence of certain members in C
structures. If there is no macro specifically defined to check for a
member you need, then you can use the general structure-member macros
(see Generic Structures) or, for more complex tests, you may use
AC_COMPILE_IFELSE
(see Running the Compiler).
The following macros check for certain structures or structure members.
Perform all the actions of
AC_HEADER_DIRENT
(see Particular Headers). Then, ifstruct dirent
contains ad_ino
member, defineHAVE_STRUCT_DIRENT_D_INO
.
HAVE_STRUCT_DIRENT_D_INO
indicates only the presence ofd_ino
, not whether its contents are always reliable. Traditionally, a zerod_ino
indicated a deleted directory entry, though current systems hide this detail from the user and never return zerod_ino
values. Many current systems report an incorrectd_ino
for a directory entry that is a mount point.
Perform all the actions of
AC_HEADER_DIRENT
(see Particular Headers). Then, ifstruct dirent
contains ad_type
member, defineHAVE_STRUCT_DIRENT_D_TYPE
.
If
struct stat
contains anst_blksize
member, defineHAVE_STRUCT_STAT_ST_BLKSIZE
. The former name,HAVE_ST_BLKSIZE
is to be avoided, as its support will cease in the future. This macro is obsoleted, and should be replaced byAC_CHECK_MEMBERS([struct stat.st_blksize])
If
struct stat
contains anst_blocks
member, defineHAVE_STRUCT_STAT_ST_BLOCKS
. Otherwise, require anAC_LIBOBJ
replacement of ‘fileblocks’. The former name,HAVE_ST_BLOCKS
is to be avoided, as its support will cease in the future.
If
struct stat
contains anst_rdev
member, defineHAVE_STRUCT_STAT_ST_RDEV
. The former name for this macro,HAVE_ST_RDEV
, is to be avoided as it will cease to be supported in the future. Actually, even the new macro is obsolete and should be replaced by:AC_CHECK_MEMBERS([struct stat.st_rdev])
If time.h does not define
struct tm
, defineTM_IN_SYS_TIME
, which means that including sys/time.h had better definestruct tm
.This macro is obsolescent, as time.h defines
struct tm
in current systems. New programs need not use this macro.
Figure out how to get the current timezone. If
struct tm
has atm_zone
member, defineHAVE_STRUCT_TM_TM_ZONE
(and the obsoletedHAVE_TM_ZONE
). Otherwise, if the external arraytzname
is found, defineHAVE_TZNAME
; if it is declared, defineHAVE_DECL_TZNAME
.
These macros are used to find structure members not covered by the “particular” test macros.
Check whether member is a member of the aggregate aggregate. If no includes are specified, the default includes are used (see Default Includes).
AC_CHECK_MEMBER([struct passwd.pw_gecos], [], [AC_MSG_ERROR([We need `passwd.pw_gecos'!])], [#include <pwd.h>])You can use this macro for submembers:
AC_CHECK_MEMBER(struct top.middle.bot)
Check for the existence of each ‘aggregate.member’ of members using the previous macro. When member belongs to aggregate, define
HAVE_
aggregate_
member (in all capitals, with spaces and dots replaced by underscores). If action-if-found is given, it is executed for each of the found members. If action-if-not-found is given, it is executed for each of the members that could not be found.This macro uses M4 lists:
AC_CHECK_MEMBERS([struct stat.st_rdev, struct stat.st_blksize])
The following macros check for C types, either builtin or typedefs. If there is no macro specifically defined to check for a type you need, and you don't need to check for any special properties of it, then you can use a general type-check macro.
These macros check for particular C types in sys/types.h, stdlib.h, stdint.h, inttypes.h and others, if they exist.
The Gnulib stdint
module is an alternate way to define many of
these symbols; it is useful if you prefer your code to assume a
C99-or-better environment. See Gnulib.
Define
GETGROUPS_T
to be whichever ofgid_t
orint
is the base type of the array argument togetgroups
.
If stdint.h or inttypes.h defines the type
int8_t
, defineHAVE_INT8_T
. Otherwise, defineint8_t
to a signed integer type that is exactly 8 bits wide and that uses two's complement representation, if such a type exists.
If stdint.h or inttypes.h defines the type
intmax_t
, defineHAVE_INTMAX_T
. Otherwise, defineintmax_t
to the widest signed integer type.
If stdint.h or inttypes.h defines the type
intptr_t
, defineHAVE_INTPTR_T
. Otherwise, defineintptr_t
to a signed integer type wide enough to hold a pointer, if such a type exists.
If the C compiler supports a working
long double
type, defineHAVE_LONG_DOUBLE
. Thelong double
type might have the same range and precision asdouble
.
If the C compiler supports a working
long double
type with more range or precision than thedouble
type, defineHAVE_LONG_DOUBLE_WIDER
.
If the C compiler supports a working
long long int
type, defineHAVE_LONG_LONG_INT
.
Define
HAVE_MBSTATE_T
if<wchar.h>
declares thembstate_t
type. Also, definembstate_t
to be a type if<wchar.h>
does not declare it.
If signal.h declares
signal
as returning a pointer to a function returningvoid
, defineRETSIGTYPE
to bevoid
; otherwise, define it to beint
.Define signal handlers as returning type
RETSIGTYPE
:RETSIGTYPE hup_handler () { ... }
Define
uid_t
andgid_t
to suitable types, if standard headers do not define them.
If stdint.h or inttypes.h defines the type
uint8_t
, defineHAVE_UINT8_T
. Otherwise, defineuint8_t
to an unsigned integer type that is exactly 8 bits wide, if such a type exists.
If stdint.h or inttypes.h defines the type
uintmax_t
, defineHAVE_UINTMAX_T
. Otherwise, defineuintmax_t
to the widest unsigned integer type.
If stdint.h or inttypes.h defines the type
uintptr_t
, defineHAVE_UINTPTR_T
. Otherwise, defineuintptr_t
to an unsigned integer type wide enough to hold a pointer, if such a type exists.
If the C compiler supports a working
unsigned long long int
type, defineHAVE_UNSIGNED_LONG_LONG_INT
.
These macros are used to check for types not covered by the “particular” test macros.
Check whether type is defined. It may be a compiler builtin type or defined by the includes (see Default Includes).
For each type of the types that is defined, define
HAVE_
type (in all capitals). If no includes are specified, the default includes are used (see Default Includes). If action-if-found is given, it is additional shell code to execute when one of the types is found. If action-if-not-found is given, it is executed when one of the types is not found.This macro uses M4 lists:
AC_CHECK_TYPES([ptrdiff_t]) AC_CHECK_TYPES([unsigned long long int, uintmax_t])
Autoconf, up to 2.13, used to provide to another version of
AC_CHECK_TYPE
, broken by design. In order to keep backward
compatibility, a simple heuristics, quite safe but not totally, is
implemented. In case of doubt, read the documentation of the former
AC_CHECK_TYPE
, see Obsolete Macros.
All the tests for compilers (AC_PROG_CC
, AC_PROG_CXX
,
AC_PROG_F77
) define the output variable EXEEXT
based on
the output of the compiler, typically to the empty string if
Posix and ‘.exe’ if a DOS variant.
They also define the output variable OBJEXT
based on the
output of the compiler, after .c files have been excluded, typically
to ‘o’ if Posix, ‘obj’ if a DOS variant.
If the compiler being used does not produce executables, the tests fail. If the executables can't be run, and cross-compilation is not enabled, they fail too. See Manual Configuration, for more on support for cross compiling.
Some compilers exhibit different behaviors.
int main (void) { static int test_array [sizeof (int) == 4 ? 1 : -1]; test_array [0] = 0; return 0; }
To our knowledge, there is a single compiler that does not support this
trick: the HP C compilers (the real one, not only the “bundled”) on
HP-UX 11.00. They incorrectly reject the above program with the diagnostic
“Variable-length arrays cannot have static storage.”
This bug comes from HP compilers' mishandling of sizeof (int)
,
not from the ? 1 : -1
, and
Autoconf works around this problem by casting sizeof (int)
to
long int
before comparing it.
Define
SIZEOF_
type (see Standard Symbols) to be the size in bytes of type. If ‘type’ is unknown, it gets a size of 0. If no includes are specified, the default includes are used (see Default Includes). If you provide include, be sure to include stdio.h which is required for this macro to run.This macro now works even when cross-compiling. The unused argument was used when cross-compiling.
For example, the call
AC_CHECK_SIZEOF([int *])defines
SIZEOF_INT_P
to be 8 on DEC Alpha AXP systems.
Define
ALIGNOF_
type (see Standard Symbols) to be the alignment in bytes of type. If ‘type’ is unknown, it gets a size of 0. If no includes are specified, the default includes are used (see Default Includes). If you provide include, be sure to include stddef.h and stdio.h which are required for this macro to work correctly.
Normally Autoconf ignores warnings generated by the compiler, linker, and preprocessor. If this macro is used, warnings count as fatal errors for the current language. This macro is useful when the results of configuration are used where warnings are unacceptable; for instance, if parts of a program are built with the GCC -Werror option. If the whole program is built using -Werror it is often simpler to put -Werror in the compiler flags (
CFLAGS
, etc.).
The following macros provide ways to find and exercise a C Compiler. There are a few constructs that ought to be avoided, but do not deserve being checked for, since they can easily be worked around.
#ifdef __STDC__ /\ * A comment with backslash-newlines in it. %{ %} *\ \ / char str[] = "\\ " A string with backslash-newlines in it %{ %} \\ ""; char apostrophe = '\\ \ '\ '; #endif
the compiler incorrectly fails with the diagnostics “Non-terminating
comment at end of file” and “Missing ‘#endif’ at end of file.”
Removing the lines with solitary backslashes solves the problem.
$ cc a.c b.c a.c: b.c:
This can cause problems if you observe the output of the compiler to
detect failures. Invoking ‘cc -c a.c && cc -c b.c && cc -o c a.o
b.o’ solves the issue.
#error
failing#error "Unsupported word size"
it is more portable to use an invalid directive like #Unsupported
word size
in Autoconf tests. In ordinary source code, #error
is
OK, since installers with inadequate compilers like irix can simply
examine these compilers' diagnostic output.
#line
support#line
directives whose line
numbers are greater than 32767. Nothing in Posix
makes this invalid. That is why Autoconf stopped issuing
#line
directives.
Determine a C compiler to use. If
CC
is not already set in the environment, check forgcc
andcc
, then for other C compilers. Set output variableCC
to the name of the compiler found.This macro may, however, be invoked with an optional first argument which, if specified, must be a blank-separated list of C compilers to search for. This just gives the user an opportunity to specify an alternative search list for the C compiler. For example, if you didn't like the default order, then you could invoke
AC_PROG_CC
like this:AC_PROG_CC([gcc cl cc])If the C compiler does not handle function prototypes correctly by default, try to add an option to output variable
CC
to make it so. This macro tries various options that select standard-conformance modes on various systems.After calling this macro you can check whether the C compiler has been set to accept ANSI C89 (ISO C90); if not, the shell variable
ac_cv_prog_cc_c89
is set to ‘no’. See alsoAC_C_PROTOTYPES
below.If using the GNU C compiler, set shell variable
GCC
to ‘yes’. If output variableCFLAGS
was not already set, set it to -g -O2 for the GNU C compiler (-O2 on systems where GCC does not accept -g), or -g for other compilers.
If the C compiler does not accept the -c and -o options simultaneously, define
NO_MINUS_C_MINUS_O
. This macro actually tests both the compiler found byAC_PROG_CC
, and, if different, the firstcc
in the path. The test fails if one fails. This macro was created for GNU Make to choose the default C compilation rule.
Set output variable
CPP
to a command that runs the C preprocessor. If ‘$CC -E’ doesn't work, /lib/cpp is used. It is only portable to runCPP
on files with a .c extension.Some preprocessors don't indicate missing include files by the error status. For such preprocessors an internal variable is set that causes other macros to check the standard error from the preprocessor and consider the test failed if any warnings have been reported. For most preprocessors, though, warnings do not cause include-file tests to fail unless
AC_PROG_CPP_WERROR
is also specified.
This acts like
AC_PROG_CPP
, except it treats warnings from the preprocessor as errors even if the preprocessor exit status indicates success. This is useful for avoiding headers that generate mandatory warnings, such as deprecation notices.
The following macros check for C compiler or machine architecture
features. To check for characteristics not listed here, use
AC_COMPILE_IFELSE
(see Running the Compiler) or
AC_RUN_IFELSE
(see Runtime).
If the C compiler cannot compile ISO Standard C (currently C99), try to add an option to output variable
CC
to make it work. If the compiler does not support C99, fall back to supporting ANSI C89 (ISO C90).After calling this macro you can check whether the C compiler has been set to accept Standard C; if not, the shell variable
ac_cv_prog_cc_stdc
is set to ‘no’.
If the C compiler is not in ANSI C89 (ISO C90) mode by default, try to add an option to output variable
CC
to make it so. This macro tries various options that select ANSI C89 on some system or another. It considers the compiler to be in ANSI C89 mode if it handles function prototypes correctly.After calling this macro you can check whether the C compiler has been set to accept ANSI C89; if not, the shell variable
ac_cv_prog_cc_c89
is set to ‘no’.This macro is called automatically by
AC_PROG_CC
.
If the C compiler is not in C99 mode by default, try to add an option to output variable
CC
to make it so. This macro tries various options that select C99 on some system or another. It considers the compiler to be in C99 mode if it handles_Bool
, flexible arrays,inline
,long long int
, mixed code and declarations, named initialization of structs,restrict
, varargs macros, variable declarations infor
loops and variable length arrays.After calling this macro you can check whether the C compiler has been set to accept C99; if not, the shell variable
ac_cv_prog_cc_c99
is set to ‘no’.
Define ‘HAVE_C_BACKSLASH_A’ to 1 if the C compiler understands ‘\a’.
This macro is obsolescent, as current C compilers understand ‘\a’. New programs need not use this macro.
If words are stored with the most significant byte first (like Motorola and SPARC CPUs), execute action-if-true. If words are stored with the least significant byte first (like Intel and VAX CPUs), execute action-if-false.
This macro runs a test-case if endianness cannot be determined from the system header files. When cross-compiling, the test-case is not run but grep'ed for some magic values. action-if-unknown is executed if the latter case fails to determine the byte sex of the host system.
The default for action-if-true is to define ‘WORDS_BIGENDIAN’. The default for action-if-false is to do nothing. And finally, the default for action-if-unknown is to abort configure and tell the installer which variable he should preset to bypass this test.
If the C compiler does not fully support the
const
keyword, defineconst
to be empty. Some C compilers that do not define__STDC__
do supportconst
; some compilers that define__STDC__
do not completely supportconst
. Programs can simply useconst
as if every C compiler supported it; for those that don't, the makefile or configuration header file defines it as empty.Occasionally installers use a C++ compiler to compile C code, typically because they lack a C compiler. This causes problems with
const
, because C and C++ treatconst
differently. For example:const int foo;is valid in C but not in C++. These differences unfortunately cannot be papered over by defining
const
to be empty.If autoconf detects this situation, it leaves
const
alone, as this generally yields better results in practice. However, using a C++ compiler to compile C code is not recommended or supported, and installers who run into trouble in this area should get a C compiler like GCC to compile their C code.This macro is obsolescent, as current C compilers support
const
. New programs need not use this macro.
If the C compiler recognizes the
restrict
keyword, don't do anything. If it recognizes only a variant spelling (__restrict
,__restrict__
, or_Restrict
), then definerestrict
to that. Otherwise, definerestrict
to be empty. Thus, programs may simply userestrict
as if every C compiler supported it; for those that do not, the makefile or configuration header defines it away.Although support in C++ for the
restrict
keyword is not required, several C++ compilers do accept the keyword. This macro works for them, too.
If the C compiler does not understand the keyword
volatile
, definevolatile
to be empty. Programs can simply usevolatile
as if every C compiler supported it; for those that do not, the makefile or configuration header defines it as empty.If the correctness of your program depends on the semantics of
volatile
, simply defining it to be empty does, in a sense, break your code. However, given that the compiler does not supportvolatile
, you are at its mercy anyway. At least your program compiles, when it wouldn't before.In general, the
volatile
keyword is a standard C feature, so you might expect thatvolatile
is available only when__STDC__
is defined. However, Ultrix 4.3's native compiler does support volatile, but does not define__STDC__
.This macro is obsolescent, as current C compilers support
volatile
. New programs need not use this macro.
If the C compiler supports the keyword
inline
, do nothing. Otherwise defineinline
to__inline__
or__inline
if it accepts one of those, otherwise defineinline
to be empty.
If the C type
char
is unsigned, define__CHAR_UNSIGNED__
, unless the C compiler predefines it.
If the C preprocessor supports the stringizing operator, define
HAVE_STRINGIZE
. The stringizing operator is ‘#’ and is found in macros such as this:#define x(y) #yThis macro is obsolescent, as current C compilers support the stringizing operator. New programs need not use this macro.
If the C compiler supports GCC's
typeof
syntax either directly or through a different spelling of the keyword (e.g.,__typeof__
), defineHAVE_TYPEOF
. If the support is available only through a different spelling, definetypeof
to that spelling.
If function prototypes are understood by the compiler (as determined by
AC_PROG_CC
), definePROTOTYPES
and__PROTOTYPES
. Defining__PROTOTYPES
is for the benefit of header files that cannot use macros that infringe on user name space.This macro is obsolescent, as current C compilers support prototypes. New programs need not use this macro.
Add -traditional to output variable
CC
if using the GNU C compiler andioctl
does not work properly without -traditional. That usually happens when the fixed header files have not been installed on an old system.This macro is obsolescent, since current versions of the GNU C compiler fix the header files automatically when installed.
Determine a C++ compiler to use. Check whether the environment variable
CXX
orCCC
(in that order) is set; if so, then set output variableCXX
to its value.Otherwise, if the macro is invoked without an argument, then search for a C++ compiler under the likely names (first
g++
andc++
then other names). If none of those checks succeed, then as a last resort setCXX
tog++
.This macro may, however, be invoked with an optional first argument which, if specified, must be a blank-separated list of C++ compilers to search for. This just gives the user an opportunity to specify an alternative search list for the C++ compiler. For example, if you didn't like the default order, then you could invoke
AC_PROG_CXX
like this:AC_PROG_CXX([gcc cl KCC CC cxx cc++ xlC aCC c++ g++])If using the GNU C++ compiler, set shell variable
GXX
to ‘yes’. If output variableCXXFLAGS
was not already set, set it to -g -O2 for the GNU C++ compiler (-O2 on systems where G++ does not accept -g), or -g for other compilers.
Set output variable
CXXCPP
to a command that runs the C++ preprocessor. If ‘$CXX -E’ doesn't work, /lib/cpp is used. It is portable to runCXXCPP
only on files with a .c, .C, .cc, or .cpp extension.Some preprocessors don't indicate missing include files by the error status. For such preprocessors an internal variable is set that causes other macros to check the standard error from the preprocessor and consider the test failed if any warnings have been reported. However, it is not known whether such broken preprocessors exist for C++.
Test whether the C++ compiler accepts the options -c and -o simultaneously, and define
CXX_NO_MINUS_C_MINUS_O
, if it does not.
Determine an Objective C compiler to use. If
OBJC
is not already set in the environment, check for Objective C compilers. Set output variableOBJC
to the name of the compiler found.This macro may, however, be invoked with an optional first argument which, if specified, must be a blank-separated list of Objective C compilers to search for. This just gives the user an opportunity to specify an alternative search list for the Objective C compiler. For example, if you didn't like the default order, then you could invoke
AC_PROG_OBJC
like this:AC_PROG_OBJC([gcc objcc objc])If using the GNU Objective C compiler, set shell variable
GOBJC
to ‘yes’. If output variableOBJCFLAGS
was not already set, set it to -g -O2 for the GNU Objective C compiler (-O2 on systems where gcc does not accept -g), or -g for other compilers.
Set output variable
OBJCCPP
to a command that runs the Objective C preprocessor. If ‘$OBJC -E’ doesn't work, /lib/cpp is used.
Autoconf defines the following macros for determining paths to the essential Erlang/OTP programs:
Determine an Erlang compiler to use. If
ERLC
is not already set in the environment, check for erlc. Set output variableERLC
to the complete path of the compiler command found. In addition, ifERLCFLAGS
is not set in the environment, set it to an empty value.The two optional arguments have the same meaning as the two last arguments of macro
AC_PROG_PATH
for looking for the erlc program. For example, to look for erlc only in the /usr/lib/erlang/bin directory:AC_ERLANG_PATH_ERLC([not found], [/usr/lib/erlang/bin])
A simplified variant of the
AC_ERLANG_PATH_ERLC
macro, that prints an error message and exits the configure script if the erlc program is not found.
Determine an Erlang interpreter to use. If
ERL
is not already set in the environment, check for erl. Set output variableERL
to the complete path of the interpreter command found.The two optional arguments have the same meaning as the two last arguments of macro
AC_PROG_PATH
for looking for the erl program. For example, to look for erl only in the /usr/lib/erlang/bin directory:AC_ERLANG_PATH_ERL([not found], [/usr/lib/erlang/bin])
A simplified variant of the
AC_ERLANG_PATH_ERL
macro, that prints an error message and exits the configure script if the erl program is not found.
The Autoconf Fortran support is divided into two categories: legacy
Fortran 77 macros (F77
), and modern Fortran macros (FC
).
The former are intended for traditional Fortran 77 code, and have output
variables like F77
, FFLAGS
, and FLIBS
. The latter
are for newer programs that can (or must) compile under the newer
Fortran standards, and have output variables like FC
,
FCFLAGS
, and FCLIBS
.
Except for two new macros AC_FC_SRCEXT
and
AC_FC_FREEFORM
(see below), the FC
and F77
macros
behave almost identically, and so they are documented together in this
section.
Determine a Fortran 77 compiler to use. If
F77
is not already set in the environment, then check forg77
andf77
, and then some other names. Set the output variableF77
to the name of the compiler found.This macro may, however, be invoked with an optional first argument which, if specified, must be a blank-separated list of Fortran 77 compilers to search for. This just gives the user an opportunity to specify an alternative search list for the Fortran 77 compiler. For example, if you didn't like the default order, then you could invoke
AC_PROG_F77
like this:AC_PROG_F77([fl32 f77 fort77 xlf g77 f90 xlf90])If using
g77
(the GNU Fortran 77 compiler), then set the shell variableG77
to ‘yes’. If the output variableFFLAGS
was not already set in the environment, then set it to -g -02 forg77
(or -O2 whereg77
does not accept -g). Otherwise, setFFLAGS
to -g for all other Fortran 77 compilers.
Determine a Fortran compiler to use. If
FC
is not already set in the environment, thendialect
is a hint to indicate what Fortran dialect to search for; the default is to search for the newest available dialect. Set the output variableFC
to the name of the compiler found.By default, newer dialects are preferred over older dialects, but if
dialect
is specified then older dialects are preferred starting with the specified dialect.dialect
can currently be one of Fortran 77, Fortran 90, or Fortran 95. However, this is only a hint of which compiler name to prefer (e.g.,f90
orf95
), and no attempt is made to guarantee that a particular language standard is actually supported. Thus, it is preferable that you avoid thedialect
option, and use AC_PROG_FC only for code compatible with the latest Fortran standard.This macro may, alternatively, be invoked with an optional first argument which, if specified, must be a blank-separated list of Fortran compilers to search for, just as in
AC_PROG_F77
.If the output variable
FCFLAGS
was not already set in the environment, then set it to -g -02 for GNUg77
(or -O2 whereg77
does not accept -g). Otherwise, setFCFLAGS
to -g for all other Fortran compilers.
Test whether the Fortran compiler accepts the options -c and -o simultaneously, and define
F77_NO_MINUS_C_MINUS_O
orFC_NO_MINUS_C_MINUS_O
, respectively, if it does not.
The following macros check for Fortran compiler characteristics.
To check for characteristics not listed here, use
AC_COMPILE_IFELSE
(see Running the Compiler) or
AC_RUN_IFELSE
(see Runtime), making sure to first set the
current language to Fortran 77 or Fortran via AC_LANG([Fortran 77])
or AC_LANG(Fortran)
(see Language Choice).
Determine the linker flags (e.g., -L and -l) for the Fortran intrinsic and runtime libraries that are required to successfully link a Fortran program or shared library. The output variable
FLIBS
orFCLIBS
is set to these flags (which should be included afterLIBS
when linking).This macro is intended to be used in those situations when it is necessary to mix, e.g., C++ and Fortran source code in a single program or shared library (see Mixing Fortran 77 With C and C++).
For example, if object files from a C++ and Fortran compiler must be linked together, then the C++ compiler/linker must be used for linking (since special C++-ish things need to happen at link time like calling global constructors, instantiating templates, enabling exception support, etc.).
However, the Fortran intrinsic and runtime libraries must be linked in as well, but the C++ compiler/linker doesn't know by default how to add these Fortran 77 libraries. Hence, this macro was created to determine these Fortran libraries.
The macros
AC_F77_DUMMY_MAIN
andAC_FC_DUMMY_MAIN
orAC_F77_MAIN
andAC_FC_MAIN
are probably also necessary to link C/C++ with Fortran; see below.
With many compilers, the Fortran libraries detected by
AC_F77_LIBRARY_LDFLAGS
orAC_FC_LIBRARY_LDFLAGS
provide their ownmain
entry function that initializes things like Fortran I/O, and which then calls a user-provided entry function named (say)MAIN__
to run the user's program. TheAC_F77_DUMMY_MAIN
andAC_FC_DUMMY_MAIN
orAC_F77_MAIN
andAC_FC_MAIN
macros figure out how to deal with this interaction.When using Fortran for purely numerical functions (no I/O, etc.) often one prefers to provide one's own
main
and skip the Fortran library initializations. In this case, however, one may still need to provide a dummyMAIN__
routine in order to prevent linking errors on some systems.AC_F77_DUMMY_MAIN
orAC_FC_DUMMY_MAIN
detects whether any such routine is required for linking, and what its name is; the shell variableF77_DUMMY_MAIN
orFC_DUMMY_MAIN
holds this name,unknown
when no solution was found, andnone
when no such dummy main is needed.By default, action-if-found defines
F77_DUMMY_MAIN
orFC_DUMMY_MAIN
to the name of this routine (e.g.,MAIN__
) if it is required. action-if-not-found defaults to exiting with an error.In order to link with Fortran routines, the user's C/C++ program should then include the following code to define the dummy main if it is needed:
#ifdef F77_DUMMY_MAIN # ifdef __cplusplus extern "C" # endif int F77_DUMMY_MAIN() { return 1; } #endif(Replace
F77
withFC
for Fortran instead of Fortran 77.)Note that this macro is called automatically from
AC_F77_WRAPPERS
orAC_FC_WRAPPERS
; there is generally no need to call it explicitly unless one wants to change the default actions.
As discussed above, many Fortran libraries allow you to provide an entry point called (say)
MAIN__
instead of the usualmain
, which is then called by amain
function in the Fortran libraries that initializes things like Fortran I/O. TheAC_F77_MAIN
andAC_FC_MAIN
macros detect whether it is possible to utilize such an alternate main function, and definesF77_MAIN
andFC_MAIN
to the name of the function. (If no alternate main function name is found,F77_MAIN
andFC_MAIN
are simply defined tomain
.)Thus, when calling Fortran routines from C that perform things like I/O, one should use this macro and name the "main" function
F77_MAIN
orFC_MAIN
instead ofmain
.
Defines C macros
F77_FUNC (name, NAME)
,FC_FUNC (name, NAME)
,F77_FUNC_(name, NAME)
, andFC_FUNC_(name, NAME)
to properly mangle the names of C/C++ identifiers, and identifiers with underscores, respectively, so that they match the name-mangling scheme used by the Fortran compiler.Fortran is case-insensitive, and in order to achieve this the Fortran compiler converts all identifiers into a canonical case and format. To call a Fortran subroutine from C or to write a C function that is callable from Fortran, the C program must explicitly use identifiers in the format expected by the Fortran compiler. In order to do this, one simply wraps all C identifiers in one of the macros provided by
AC_F77_WRAPPERS
orAC_FC_WRAPPERS
. For example, suppose you have the following Fortran 77 subroutine:subroutine foobar (x, y) double precision x, y y = 3.14159 * x return endYou would then declare its prototype in C or C++ as:
#define FOOBAR_F77 F77_FUNC (foobar, FOOBAR) #ifdef __cplusplus extern "C" /* prevent C++ name mangling */ #endif void FOOBAR_F77(double *x, double *y);Note that we pass both the lowercase and uppercase versions of the function name to
F77_FUNC
so that it can select the right one. Note also that all parameters to Fortran 77 routines are passed as pointers (see Mixing Fortran 77 With C and C++).(Replace
F77
withFC
for Fortran instead of Fortran 77.)Although Autoconf tries to be intelligent about detecting the name-mangling scheme of the Fortran compiler, there may be Fortran compilers that it doesn't support yet. In this case, the above code generates a compile-time error, but some other behavior (e.g., disabling Fortran-related features) can be induced by checking whether
F77_FUNC
orFC_FUNC
is defined.Now, to call that routine from a C program, we would do something like:
{ double x = 2.7183, y; FOOBAR_F77 (&x, &y); }If the Fortran identifier contains an underscore (e.g.,
foo_bar
), you should useF77_FUNC_
orFC_FUNC_
instead ofF77_FUNC
orFC_FUNC
(with the same arguments). This is because some Fortran compilers mangle names differently if they contain an underscore.
Given an identifier name, set the shell variable shellvar to hold the mangled version name according to the rules of the Fortran linker (see also
AC_F77_WRAPPERS
orAC_FC_WRAPPERS
). shellvar is optional; if it is not supplied, the shell variable is simply name. The purpose of this macro is to give the caller a way to access the name-mangling information other than through the C preprocessor as above, for example, to call Fortran routines from some language other than C/C++.
By default, the
FC
macros perform their tests using a .f extension for source-code files. Some compilers, however, only enable newer language features for appropriately named files, e.g., Fortran 90 features only for .f90 files. On the other hand, some other compilers expect all source files to end in .f and require special flags to support other file name extensions. TheAC_FC_SRCEXT
macro deals with both of these issues.The
AC_FC_SRCEXT
tries to get theFC
compiler to accept files ending with the extension .ext (i.e., ext does not contain the dot). If any special compiler flags are needed for this, it stores them in the output variableFCFLAGS_
ext. This extension and these flags are then used for all subsequentFC
tests (untilAC_FC_SRCEXT
is called again).For example, you would use
AC_FC_SRCEXT(f90)
to employ the .f90 extension in future tests, and it would set aFCFLAGS_f90
output variable with any extra flags that are needed to compile such files.The
FCFLAGS_
ext can not be simply absorbed intoFCFLAGS
, for two reasons based on the limitations of some compilers. First, only oneFCFLAGS_
ext can be used at a time, so files with different extensions must be compiled separately. Second,FCFLAGS_
ext must appear immediately before the source-code file name when compiling. So, continuing the example above, you might compile a foo.f90 file in your makefile with the command:foo.o: foo.f90 $(FC) -c $(FCFLAGS) $(FCFLAGS_f90) '$(srcdir)/foo.f90'If
AC_FC_SRCEXT
succeeds in compiling files with the ext extension, it calls action-if-success (defaults to nothing). If it fails, and cannot find a way to make theFC
compiler accept such files, it calls action-if-failure (defaults to exiting with an error message).
The
AC_FC_FREEFORM
tries to ensure that the Fortran compiler ($FC
) allows free-format source code (as opposed to the older fixed-format style from Fortran 77). If necessary, it may add some additional flags toFCFLAGS
.This macro is most important if you are using the default .f extension, since many compilers interpret this extension as indicating fixed-format source unless an additional flag is supplied. If you specify a different extension with
AC_FC_SRCEXT
, such as .f90 or .f95, thenAC_FC_FREEFORM
ordinarily succeeds without modifyingFCFLAGS
.If
AC_FC_FREEFORM
succeeds in compiling free-form source, it calls action-if-success (defaults to nothing). If it fails, it calls action-if-failure (defaults to exiting with an error message).
The following macros check for operating system services or capabilities.
Try to locate the X Window System include files and libraries. If the user gave the command line options --x-includes=dir and --x-libraries=dir, use those directories.
If either or both were not given, get the missing values by running
xmkmf
(or an executable pointed to by theXMKMF
environment variable) on a trivial Imakefile and examining the makefile that it produces. SettingXMKMF
to ‘false’ disables this method.If this method fails to find the X Window System, configure looks for the files in several directories where they often reside. If either method is successful, set the shell variables
x_includes
andx_libraries
to their locations, unless they are in directories the compiler searches by default.If both methods fail, or the user gave the command line option --without-x, set the shell variable
no_x
to ‘yes’; otherwise set it to the empty string.
An enhanced version of
AC_PATH_X
. It adds the C compiler flags that X needs to output variableX_CFLAGS
, and the X linker flags toX_LIBS
. DefineX_DISPLAY_MISSING
if X is not available.This macro also checks for special libraries that some systems need in order to compile X programs. It adds any that the system needs to output variable
X_EXTRA_LIBS
. And it checks for special X11R6 libraries that need to be linked with before -lX11, and adds any found to the output variableX_PRE_LIBS
.
Check whether the system supports starting scripts with a line of the form ‘#!/bin/sh’ to select the interpreter to use for the script. After running this macro, shell code in configure.ac can check the shell variable
interpval
; it is set to ‘yes’ if the system supports ‘#!’, ‘no’ if not.
Arrange for large-file support. On some hosts, one must use special compiler options to build programs that can access large files. Append any such options to the output variable
CC
. Define_FILE_OFFSET_BITS
and_LARGE_FILES
if necessary.Large-file support can be disabled by configuring with the --disable-largefile option.
If you use this macro, check that your program works even when
off_t
is wider thanlong int
, since this is common when large-file support is enabled. For example, it is not correct to print an arbitraryoff_t
valueX
withprintf ("%ld", (long int) X)
.The LFS introduced the
fseeko
andftello
functions to replace their C counterpartsfseek
andftell
that do not useoff_t
. Take care to useAC_FUNC_FSEEKO
to make their prototypes available when using them and large-file support is enabled.
If the system supports file names longer than 14 characters, define
HAVE_LONG_FILE_NAMES
.
Check to see if the Posix termios headers and functions are available on the system. If so, set the shell variable
ac_cv_sys_posix_termios
to ‘yes’. If not, set the variable to ‘no’.
The following macros check for certain operating systems that need special treatment for some programs, due to exceptional oddities in their header files or libraries. These macros are warts; they will be replaced by a more systematic approach, based on the functions they make available or the environments they provide.
If on AIX, define
_ALL_SOURCE
. Allows the use of some BSD functions. Should be called before any macros that run the C compiler.
If using the GNU C library, define
_GNU_SOURCE
. Allows the use of some GNU functions. Should be called before any macros that run the C compiler.
For interactive Systems Corporation Unix, add -lcposix to output variable
LIBS
if necessary for Posix facilities. Call this afterAC_PROG_CC
and before any other macros that use Posix interfaces. interactive Unix is no longer sold, and Sun says that they will drop support for it on 2006-07-23, so this macro is becoming obsolescent.
If on Minix, define
_MINIX
and_POSIX_SOURCE
and define_POSIX_1_SOURCE
to be 2. This allows the use of Posix facilities. Should be called before any macros that run the C compiler.
If possible, enable extensions to Posix on hosts that normally disable the extensions, typically due to standards-conformance namespace issues. This may involve defining
__EXTENSIONS__
and_POSIX_PTHREAD_SEMANTICS
, which are macros used by Solaris. This macro also has the combined effects ofAC_GNU_SOURCE
,AC_AIX
, andAC_MINIX
.
The following macros check for an installation of Erlang/OTP, and for the presence of certain Erlang libraries. All those macros require the configuration of an Erlang interpreter and an Erlang compiler (see Erlang Compiler and Interpreter).
Set the output variable
ERLANG_ROOT_DIR
to the path to the base directory in which Erlang/OTP is installed (as returned by Erlang'scode:root_dir/0
function). The result of this test is cached if caching is enabled when running configure.
Set the output variable
ERLANG_LIB_DIR
to the path of the library directory of Erlang/OTP (as returned by Erlang'scode:lib_dir/0
function), which subdirectories each contain an installed Erlang/OTP library. The result of this test is cached if caching is enabled when running configure.
Test whether the Erlang/OTP library library is installed by calling Erlang's
code:lib_dir/1
function. The result of this test is cached if caching is enabled when running configure. action-if-found is a list of shell commands to run if the library is installed; action-if-not-found is a list of shell commands to run if it is not. Additionally, if the library is installed, the output variable ‘ERLANG_LIB_DIR_library’ is set to the path to the library installation directory. For example, to check if librarystdlib
is installed:AC_ERLANG_CHECK_LIB([stdlib], [echo "stdlib is installed in $ERLANG_LIB_DIR_stdlib"], [AC_MSG_ERROR([stdlib was not found!])])
In addition to the above macros, which test installed Erlang libraries, the following macros determine the paths to the directories into which newly built Erlang libraries are to be installed:
Set the
ERLANG_INSTALL_LIB_DIR
output variable to the directory into which every built Erlang library should be installed in a separate subdirectory. If this variable is not set in the environment when configure runs, its default value is$ERLANG_LIB_DIR
, which value is set by theAC_ERLANG_SUBST_LIB_DIR
macro.
Set the ‘ERLANG_INSTALL_LIB_DIR_library’ output variable to the directory into which the built Erlang library library version version should be installed. If this variable is not set in the environment when configure runs, its default value is ‘$ERLANG_INSTALL_LIB_DIR/library-version’, the value of the
ERLANG_INSTALL_LIB_DIR
variable being set by theAC_ERLANG_SUBST_INSTALL_LIB_DIR
macro.
If the existing feature tests don't do something you need, you have to write new ones. These macros are the building blocks. They provide ways for other macros to check whether various kinds of features are available and report the results.
This chapter contains some suggestions and some of the reasons why the existing tests are written the way they are. You can also learn a lot about how to write Autoconf tests by looking at the existing ones. If something goes wrong in one or more of the Autoconf tests, this information can help you understand the assumptions behind them, which might help you figure out how to best solve the problem.
These macros check the output of the compiler system of the current language (see Language Choice). They do not cache the results of their tests for future use (see Caching Results), because they don't know enough about the information they are checking for to generate a cache variable name. They also do not print any messages, for the same reason. The checks for particular kinds of features call these macros and do cache their results and print messages about what they're checking for.
When you write a feature test that could be applicable to more than one software package, the best thing to do is encapsulate it in a new macro. See Writing Autoconf Macros, for how to do that.
Autoconf-generated configure scripts check for the C compiler and its features by default. Packages that use other programming languages (maybe more than one, e.g., C and C++) need to test features of the compilers for the respective languages. The following macros determine which programming language is used in the subsequent tests in configure.ac.
Do compilation tests using the compiler, preprocessor, and file extensions for the specified language.
Supported languages are:
- ‘C’
- Do compilation tests using
CC
andCPP
and use extension .c for test programs. Use compilation flags:CPPFLAGS
withCPP
, and bothCPPFLAGS
andCFLAGS
withCC
.- ‘C++’
- Do compilation tests using
CXX
andCXXCPP
and use extension .C for test programs. Use compilation flags:CPPFLAGS
withCXXPP
, and bothCPPFLAGS
andCXXFLAGS
withCXX
.- ‘Fortran 77’
- Do compilation tests using
F77
and use extension .f for test programs. Use compilation flags:FFLAGS
.- ‘Fortran’
- Do compilation tests using
FC
and use extension .f (or whatever has been set byAC_FC_SRCEXT
) for test programs. Use compilation flags:FCFLAGS
.- ‘Erlang’
- Compile and execute tests using
ERLC
andERL
and use extension .erl for test Erlang modules. Use compilation flags:ERLCFLAGS
.- ‘Objective C’
- Do compilation tests using
OBJC
andOBJCCPP
and use extension .m for test programs. Use compilation flags:CPPFLAGS
withOBJCPP
, and bothCPPFLAGS
andOBJCFLAGS
withOBJC
.
Remember the current language (as set by
AC_LANG
) on a stack, and then select the language. Use this macro andAC_LANG_POP
in macros that need to temporarily switch to a particular language.
Select the language that is saved on the top of the stack, as set by
AC_LANG_PUSH
, and remove it from the stack.If given, language specifies the language we just quit. It is a good idea to specify it when it's known (which should be the case...), since Autoconf detects inconsistencies.
AC_LANG_PUSH([Fortran 77]) # Perform some tests on Fortran 77. # ... AC_LANG_POP([Fortran 77])
Check statically that the current language is language. You should use this in your language specific macros to avoid that they be called with an inappropriate language.
This macro runs only at autoconf time, and incurs no cost at configure time. Sadly enough and because Autoconf is a two layer language 2, the macros
AC_LANG_PUSH
andAC_LANG_POP
cannot be “optimizing”, therefore as much as possible you ought to avoid using them to wrap your code, rather, require from the user to run the macro with a correct current language, and check it withAC_LANG_ASSERT
. And anyway, that may help the user understand she is running a Fortran macro while expecting a result about her Fortran 77 compiler...
Ensure that whichever preprocessor would currently be used for tests has been found. Calls
AC_REQUIRE
(see Prerequisite Macros) with an argument of eitherAC_PROG_CPP
orAC_PROG_CXXCPP
, depending on which language is current.
Autoconf tests follow a common scheme: feed some program with some input, and most of the time, feed a compiler with some source file. This section is dedicated to these source samples.
The most important rule to follow when writing testing samples is:
This motto means that testing samples must be written with the same strictness as real programs are written. In particular, you should avoid “shortcuts” and simplifications.
Don't just play with the preprocessor if you want to prepare a compilation. For instance, using cpp to check whether a header is functional might let your configure accept a header which causes some compiler error. Do not hesitate to check a header with other headers included before, especially required headers.
Make sure the symbols you use are properly defined, i.e., refrain for simply declaring a function yourself instead of including the proper header.
Test programs should not write to standard output. They
should exit with status 0 if the test succeeds, and with status 1
otherwise, so that success
can be distinguished easily from a core dump or other failure;
segmentation violations and other failures produce a nonzero exit
status. Unless you arrange for exit
to be declared, test
programs should return
, not exit
, from main
,
because on many systems exit
is not declared by default.
Test programs can use #if
or #ifdef
to check the values of
preprocessor macros defined by tests that have already run. For
example, if you call AC_HEADER_STDBOOL
, then later on in
configure.ac you can have a test program that includes
stdbool.h conditionally:
#if HAVE_STDBOOL_H # include <stdbool.h> #endif
If a test program needs to use or create a data file, give it a name that starts with conftest, such as conftest.data. The configure script cleans up by running ‘rm -f -r conftest*’ after running test programs and if the script is interrupted.
These days it's safe to assume support for function prototypes (introduced in C89).
Functions that test programs declare should also be conditionalized for C++, which requires ‘extern "C"’ prototypes. Make sure to not include any header files containing clashing prototypes.
#ifdef __cplusplus extern "C" #endif void *valloc (size_t);
If a test program calls a function with invalid parameters (just to see
whether it exists), organize the program to ensure that it never invokes
that function. You can do this by calling it in another function that is
never invoked. You can't do it by putting it after a call to
exit
, because GCC version 2 knows that exit
never returns
and optimizes out any code that follows it in the same block.
If you include any header files, be sure to call the functions
relevant to them with the correct number of arguments, even if they are
just 0, to avoid compilation errors due to prototypes. GCC
version 2
has internal prototypes for several functions that it automatically
inlines; for example, memcpy
. To avoid errors when checking for
them, either pass them the correct number of arguments or redeclare them
with a different return type (such as char
).
Autoconf provides a set of macros that can be used to generate test source files. They are written to be language generic, i.e., they actually depend on the current language (see Language Choice) to “format” the output properly.
Save the source text in the current test source file: conftest.extension where the extension depends on the current language.
Note that the source is evaluated exactly once, like regular Autoconf macro arguments, and therefore (i) you may pass a macro invocation, (ii) if not, be sure to double quote if needed.
Expands into the source, with the definition of all the
AC_DEFINE
performed so far.
For instance executing (observe the double quotation!):
AC_INIT([Hello], [1.0], [bug-hello@example.org]) AC_DEFINE([HELLO_WORLD], ["Hello, World\n"], [Greetings string.]) AC_LANG(C) AC_LANG_CONFTEST( [AC_LANG_SOURCE([[const char hw[] = "Hello, World\n";]])]) gcc -E -dD -o - conftest.c
results in:
... # 1 "conftest.c" #define PACKAGE_NAME "Hello" #define PACKAGE_TARNAME "hello" #define PACKAGE_VERSION "1.0" #define PACKAGE_STRING "Hello 1.0" #define PACKAGE_BUGREPORT "bug-hello@example.org" #define HELLO_WORLD "Hello, World\n" const char hw[] = "Hello, World\n";
When the test language is Fortran or Erlang, the AC_DEFINE
definitions
are not automatically translated into constants in the source code by this
macro.
Expands into a source file which consists of the prologue, and then body as body of the main function (e.g.,
main
in C). Since it usesAC_LANG_SOURCE
, the features of the latter are available.
For instance:
AC_INIT([Hello], [1.0], [bug-hello@example.org]) AC_DEFINE([HELLO_WORLD], ["Hello, World\n"], [Greetings string.]) AC_LANG_CONFTEST( [AC_LANG_PROGRAM([[const char hw[] = "Hello, World\n";]], [[fputs (hw, stdout);]])]) gcc -E -dD -o - conftest.c
results in:
... # 1 "conftest.c" #define PACKAGE_NAME "Hello" #define PACKAGE_TARNAME "hello" #define PACKAGE_VERSION "1.0" #define PACKAGE_STRING "Hello 1.0" #define PACKAGE_BUGREPORT "bug-hello@example.org" #define HELLO_WORLD "Hello, World\n" const char hw[] = "Hello, World\n"; int main () { fputs (hw, stdout); ; return 0; }
In Erlang tests, the created source file is that of an Erlang module called
conftest
(conftest.erl). This module defines and exports at least
one start/0
function, which is called to perform the test. The
prologue is optional code that is inserted between the module header and
the start/0
function definition. body is the body of the
start/0
function without the final period (see Runtime, about
constraints on this function's behaviour).
For instance:
AC_INIT([Hello], [1.0], [bug-hello@example.org]) AC_LANG(Erlang) AC_LANG_CONFTEST( [AC_LANG_PROGRAM([[-define(HELLO_WORLD, "Hello, world!").]], [[io:format("~s~n", [?HELLO_WORLD])]])]) cat conftest.erl
results in:
-module(conftest). -export([start/0]). -define(HELLO_WORLD, "Hello, world!"). start() -> io:format("~s~n", [?HELLO_WORLD]) .
Expands into a source file which consists of the prologue, and then a call to the function as body of the main function (e.g.,
main
in C). Since it usesAC_LANG_PROGRAM
, the feature of the latter are available.This function will probably be replaced in the future by a version which would enable specifying the arguments. The use of this macro is not encouraged, as it violates strongly the typing system.
This macro cannot be used for Erlang tests.
Expands into a source file which uses the function in the body of the main function (e.g.,
main
in C). Since it usesAC_LANG_PROGRAM
, the features of the latter are available.As
AC_LANG_CALL
, this macro is documented only for completeness. It is considered to be severely broken, and in the future will be removed in favor of actual function calls (with properly typed arguments).This macro cannot be used for Erlang tests.
Sometimes one might need to run the preprocessor on some source file. Usually it is a bad idea, as you typically need to compile your project, not merely run the preprocessor on it; therefore you certainly want to run the compiler, not the preprocessor. Resist the temptation of following the easiest path.
Nevertheless, if you need to run the preprocessor, then use
AC_PREPROC_IFELSE
.
The macros described in this section cannot be used for tests in Erlang or Fortran, since those languages require no preprocessor.
Run the preprocessor of the current language (see Language Choice) on the input, run the shell commands action-if-true on success, action-if-false otherwise. The input can be made by
AC_LANG_PROGRAM
and friends.This macro uses
CPPFLAGS
, but notCFLAGS
, because -g, -O, etc. are not valid options to many C preprocessors.It is customary to report unexpected failures with
AC_MSG_FAILURE
.
For instance:
AC_INIT([Hello], [1.0], [bug-hello@example.org]) AC_DEFINE([HELLO_WORLD], ["Hello, World\n"], [Greetings string.]) AC_PREPROC_IFELSE( [AC_LANG_PROGRAM([[const char hw[] = "Hello, World\n";]], [[fputs (hw, stdout);]])], [AC_MSG_RESULT([OK])], [AC_MSG_FAILURE([unexpected preprocessor failure])])
results in:
checking for gcc... gcc checking for C compiler default output file name... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ISO C89... none needed checking how to run the C preprocessor... gcc -E OK
The macro
AC_TRY_CPP
(see Obsolete Macros) used to play the
role of AC_PREPROC_IFELSE
, but double quotes its argument, making
it impossible to use it to elaborate sources. You are encouraged to
get rid of your old use of the macro AC_TRY_CPP
in favor of
AC_PREPROC_IFELSE
, but, in the first place, are you sure you need
to run the preprocessor and not the compiler?
If the output of running the preprocessor on the system header file header-file matches the extended regular expression pattern, execute shell commands action-if-found, otherwise execute action-if-not-found.
program is the text of a C or C++ program, on which shell variable, back quote, and backslash substitutions are performed. If the output of running the preprocessor on program matches the extended regular expression pattern, execute shell commands action-if-found, otherwise execute action-if-not-found.
To check for a syntax feature of the current language's (see Language Choice) compiler, such as whether it recognizes a certain keyword, or
simply to try some library feature, use AC_COMPILE_IFELSE
to try
to compile a small program that uses that feature.
Run the compiler and compilation flags of the current language (see Language Choice) on the input, run the shell commands action-if-true on success, action-if-false otherwise. The input can be made by
AC_LANG_PROGRAM
and friends.It is customary to report unexpected failures with
AC_MSG_FAILURE
. This macro does not try to link; useAC_LINK_IFELSE
if you need to do that (see Running the Linker).
For tests in Erlang, the input must be the source code of a module named
conftest
. AC_COMPILE_IFELSE
generates a conftest.beam
file that can be interpreted by the Erlang virtual machine (ERL
). It is
recommended to use AC_LANG_PROGRAM
to specify the test program, to ensure
that the Erlang module has the right name.
To check for a library, a function, or a global variable, Autoconf
configure scripts try to compile and link a small program that
uses it. This is unlike Metaconfig, which by default uses nm
or
ar
on the C library to try to figure out which functions are
available. Trying to link with the function is usually a more reliable
approach because it avoids dealing with the variations in the options
and output formats of nm
and ar
and in the location of the
standard libraries. It also allows configuring for cross-compilation or
checking a function's runtime behavior if needed. On the other hand,
it can be slower than scanning the libraries once, but accuracy is more
important than speed.
AC_LINK_IFELSE
is used to compile test programs to test for
functions and global variables. It is also used by AC_CHECK_LIB
to check for libraries (see Libraries), by adding the library being
checked for to LIBS
temporarily and trying to link a small
program.
Run the compiler (and compilation flags) and the linker of the current language (see Language Choice) on the input, run the shell commands action-if-true on success, action-if-false otherwise. The input can be made by
AC_LANG_PROGRAM
and friends.
LDFLAGS
andLIBS
are used for linking, in addition to the current compilation flags.It is customary to report unexpected failures with
AC_MSG_FAILURE
. This macro does not try to execute the program; useAC_RUN_IFELSE
if you need to do that (see Runtime).
The AC_LINK_IFELSE
macro cannot be used for Erlang tests, since Erlang
programs are interpreted and do not require linking.
Sometimes you need to find out how a system performs at runtime, such as whether a given function has a certain capability or bug. If you can, make such checks when your program runs instead of when it is configured. You can check for things like the machine's endianness when your program initializes itself.
If you really need to test for a runtime behavior while configuring,
you can write a test program to determine the result, and compile and
run it using AC_RUN_IFELSE
. Avoid running test programs if
possible, because this prevents people from configuring your package for
cross-compiling.
If program compiles and links successfully and returns an exit status of 0 when executed, run shell commands action-if-true. Otherwise, run shell commands action-if-false.
The input can be made by
AC_LANG_PROGRAM
and friends.LDFLAGS
andLIBS
are used for linking, in addition to the compilation flags of the current language (see Language Choice).If the compiler being used does not produce executables that run on the system where configure is being run, then the test program is not run. If the optional shell commands action-if-cross-compiling are given, they are run instead. Otherwise, configure prints an error message and exits.
In the action-if-false section, the failing exit status is available in the shell variable ‘$?’. This exit status might be that of a failed compilation, or it might be that of a failed program execution.
It is customary to report unexpected failures with
AC_MSG_FAILURE
.
Try to provide a pessimistic default value to use when cross-compiling
makes runtime tests impossible. You do this by passing the optional
last argument to AC_RUN_IFELSE
. autoconf prints a
warning message when creating configure each time it
encounters a call to AC_RUN_IFELSE
with no
action-if-cross-compiling argument given. You may ignore the
warning, though users cannot configure your package for
cross-compiling. A few of the macros distributed with Autoconf produce
this warning message.
To configure for cross-compiling you can also choose a value for those parameters based on the canonical system name (see Manual Configuration). Alternatively, set up a test results cache file with the correct values for the host system (see Caching Results).
To provide a default for calls of AC_RUN_IFELSE
that are embedded
in other macros, including a few of the ones that come with Autoconf,
you can test whether the shell variable cross_compiling
is set to
‘yes’, and then use an alternate method to get the results instead
of calling the macros.
A C or C++ runtime test should be portable. See Portable C and C++.
Erlang tests must exit themselves the Erlang VM by calling the halt/1
function: the given status code is used to determine the success of the test
(status is 0
) or its failure (status is different than 0
), as
explained above. It must be noted that data output through the standard output
(e.g. using io:format/2
) may be truncated when halting the VM.
Therefore, if a test must output configuration information, it is recommended
to create and to output data into the temporary file named conftest.out,
using the functions of module file
. The conftest.out
file is
automatically deleted by the AC_RUN_IFELSE
macro. For instance, a
simplified implementation of Autoconf's AC_ERLANG_SUBST_LIB_DIR
macro is:
AC_INIT([LibdirTest], [1.0], [bug-libdirtest@example.org]) AC_ERLANG_NEED_ERL AC_LANG(Erlang) AC_RUN_IFELSE( [AC_LANG_PROGRAM([], [dnl file:write_file("conftest.out", code:lib_dir()), halt(0)])], [echo "code:lib_dir() returned: `cat conftest.out`"], [AC_MSG_FAILURE([test Erlang program execution failed])])
This section aims at presenting some systems and pointers to documentation. It may help you addressing particular problems reported by users.
Posix-conforming systems are derived from the Unix operating system.
The Rosetta Stone for Unix contains a table correlating the features of various Posix-conforming systems. Unix History is a simplified diagram of how many Unix systems were derived from each other.
The Heirloom Project provides some variants of traditional implementations of Unix utilities.
That's all dependent on whether the file system is a UFS (case
sensitive) or HFS+ (case preserving). By default Apple wants you to
install the OS on HFS+. Unfortunately, there are some pieces of
software which really need to be built on UFS. We may want to rebuild
Darwin to have both UFS and HFS+ available (and put the /local/build
tree on the UFS).
Some operations are accomplished in several possible ways, depending on the OS variant. Checking for them essentially requires a “case statement”. Autoconf does not directly provide one; however, it is easy to simulate by using a shell variable to keep track of whether a way to perform the operation has been found yet.
Here is an example that uses the shell variable fstype
to keep
track of whether the remaining cases need to be checked.
AC_MSG_CHECKING([how to get file system type]) fstype=no # The order of these tests is important. AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <sys/statvfs.h> #include <sys/fstyp.h>]])], [AC_DEFINE([FSTYPE_STATVFS], [1], [Define if statvfs exists.]) fstype=SVR4]) if test $fstype = no; then AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <sys/statfs.h> #include <sys/fstyp.h>]])], [AC_DEFINE([FSTYPE_USG_STATFS], [1], [Define if USG statfs.]) fstype=SVR3]) fi if test $fstype = no; then AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <sys/statfs.h> #include <sys/vmount.h>]])]), [AC_DEFINE([FSTYPE_AIX_STATFS], [1], [Define if AIX statfs.]) fstype=AIX]) fi # (more cases omitted here) AC_MSG_RESULT([$fstype])
Once configure has determined whether a feature exists, what can it do to record that information? There are four sorts of things it can do: define a C preprocessor symbol, set a variable in the output files, save the result in a cache file for future configure runs, and print a message letting the user know the result of the test.
A common action to take in response to a feature test is to define a C
preprocessor symbol indicating the results of the test. That is done by
calling AC_DEFINE
or AC_DEFINE_UNQUOTED
.
By default, AC_OUTPUT
places the symbols defined by these macros
into the output variable DEFS
, which contains an option
-Dsymbol=value for each symbol defined. Unlike in
Autoconf version 1, there is no variable DEFS
defined while
configure is running. To check whether Autoconf macros have
already defined a certain C preprocessor symbol, test the value of the
appropriate cache variable, as in this example:
AC_CHECK_FUNC([vprintf], [AC_DEFINE([HAVE_VPRINTF], [1], [Define if vprintf exists.])]) if test "$ac_cv_func_vprintf" != yes; then AC_CHECK_FUNC([_doprnt], [AC_DEFINE([HAVE_DOPRNT], [1], [Define if _doprnt exists.])]) fi
If AC_CONFIG_HEADERS
has been called, then instead of creating
DEFS
, AC_OUTPUT
creates a header file by substituting the
correct values into #define
statements in a template file.
See Configuration Headers, for more information about this kind of
output.
Define the C preprocessor variable variable to value (verbatim). value should not contain literal newlines, and if you are not using
AC_CONFIG_HEADERS
it should not contain any ‘#’ characters, as make tends to eat them. To use a shell variable, useAC_DEFINE_UNQUOTED
instead. description is only useful if you are usingAC_CONFIG_HEADERS
. In this case, description is put into the generated config.h.in as the comment before the macro define. The following example defines the C preprocessor variableEQUATION
to be the string constant ‘"$a > $b"’:AC_DEFINE([EQUATION], ["$a > $b"], [Equation string.])If neither value nor description are given, then value defaults to 1 instead of to the empty string. This is for backwards compatibility with older versions of Autoconf, but this usage is obsolescent and may be withdrawn in future versions of Autoconf.
If the variable is a literal string, it is passed to
m4_pattern_allow
(see Forbidden Patterns).
Like
AC_DEFINE
, but three shell expansions are performed—once—on variable and value: variable expansion (‘$’), command substitution (‘`’), and backslash escaping (‘\’). Single and double quote characters in the value have no special meaning. Use this macro instead ofAC_DEFINE
when variable or value is a shell variable. Examples:AC_DEFINE_UNQUOTED([config_machfile], ["$machfile"], [Configuration machine file.]) AC_DEFINE_UNQUOTED([GETGROUPS_T], [$ac_cv_type_getgroups], [getgroups return type.]) AC_DEFINE_UNQUOTED([$ac_tr_hdr], [1], [Translated header name.])
Due to a syntactical bizarreness of the Bourne shell, do not use
semicolons to separate AC_DEFINE
or AC_DEFINE_UNQUOTED
calls from other macro calls or shell code; that can cause syntax errors
in the resulting configure script. Use either blanks or
newlines. That is, do this:
AC_CHECK_HEADER([elf.h], [AC_DEFINE([SVR4], [1], [System V Release 4]) LIBS="-lelf $LIBS"])
or this:
AC_CHECK_HEADER([elf.h], [AC_DEFINE([SVR4], [1], [System V Release 4]) LIBS="-lelf $LIBS"])
instead of this:
AC_CHECK_HEADER([elf.h], [AC_DEFINE([SVR4], [1], [System V Release 4]); LIBS="-lelf $LIBS"])
Another way to record the results of tests is to set output variables, which are shell variables whose values are substituted into files that configure outputs. The two macros below create new output variables. See Preset Output Variables, for a list of output variables that are always available.
Create an output variable from a shell variable. Make
AC_OUTPUT
substitute the variable variable into output files (typically one or more makefiles). This means thatAC_OUTPUT
replaces instances of ‘@variable@’ in input files with the value that the shell variable variable has whenAC_OUTPUT
is called. The value can contain newlines. The substituted value is not rescanned for more output variables; occurrences of ‘@variable@’ in the value are inserted literally into the output file. (The algorithm uses the special marker|#_!!_#|
internally, so the substituted value cannot contain|#_!!_#|
.)If value is given, in addition assign it to variable.
The string variable is passed to
m4_pattern_allow
(see Forbidden Patterns).
Another way to create an output variable from a shell variable. Make
AC_OUTPUT
insert (without substitutions) the contents of the file named by shell variable variable into output files. This means thatAC_OUTPUT
replaces instances of ‘@variable@’ in output files (such as Makefile.in) with the contents of the file that the shell variable variable names whenAC_OUTPUT
is called. Set the variable to /dev/null for cases that do not have a file to insert. This substitution occurs only when the ‘@variable@’ is on a line by itself, optionally surrounded by spaces and tabs. The substitution replaces the whole line, including the spaces, tabs, and the terminating newline.This macro is useful for inserting makefile fragments containing special dependencies or other
make
directives for particular host or target types into makefiles. For example, configure.ac could contain:AC_SUBST_FILE([host_frag]) host_frag=$srcdir/conf/sun4.mhand then a Makefile.in could contain:
@host_frag@The string variable is passed to
m4_pattern_allow
(see Forbidden Patterns).
Running configure in varying environments can be extremely dangerous. If for instance the user runs ‘CC=bizarre-cc ./configure’, then the cache, config.h, and many other output files depend upon bizarre-cc being the C compiler. If for some reason the user runs ./configure again, or if it is run via ‘./config.status --recheck’, (See Automatic Remaking, and see config.status Invocation), then the configuration can be inconsistent, composed of results depending upon two different compilers.
Environment variables that affect this situation, such as ‘CC’
above, are called precious variables, and can be declared as such
by AC_ARG_VAR
.
Declare variable is a precious variable, and include its description in the variable section of ‘./configure --help’.
Being precious means that
- variable is substituted via
AC_SUBST
.- The value of variable when configure was launched is saved in the cache, including if it was not specified on the command line but via the environment. Indeed, while configure can notice the definition of
CC
in ‘./configure CC=bizarre-cc’, it is impossible to notice it in ‘CC=bizarre-cc ./configure’, which, unfortunately, is what most users do.We emphasize that it is the initial value of variable which is saved, not that found during the execution of configure. Indeed, specifying ‘./configure FOO=foo’ and letting ‘./configure’ guess that
FOO
isfoo
can be two different things.- variable is checked for consistency between two configure runs. For instance:
$ ./configure --silent --config-cache $ CC=cc ./configure --silent --config-cache configure: error: `CC' was not set in the previous run configure: error: changes in the environment can compromise \ the build configure: error: run `make distclean' and/or \ `rm config.cache' and start overand similarly if the variable is unset, or if its content is changed.
- variable is kept during automatic reconfiguration (see config.status Invocation) as if it had been passed as a command line argument, including when no cache is used:
$ CC=/usr/bin/cc ./configure undeclared_var=raboof --silent $ ./config.status --recheck running /bin/sh ./configure undeclared_var=raboof --silent \ CC=/usr/bin/cc --no-create --no-recursion
Many output variables are intended to be evaluated both by make and by the shell. Some characters are expanded differently in these two contexts, so to avoid confusion these variables' values should not contain any of the following characters:
" # $ & ' ( ) * ; < > ? [ \ ^ ` |
Also, these variables' values should neither contain newlines, nor start with ‘~’, nor contain white space or ‘:’ immediately followed by ‘~’. The values can contain nonempty sequences of white space characters like tabs and spaces, but each such sequence might arbitrarily be replaced by a single space during substitution.
These restrictions apply both to the values that configure
computes, and to the values set directly by the user. For example, the
following invocations of configure are problematic, since they
attempt to use special characters within CPPFLAGS
and white space
within $(srcdir)
:
CPPFLAGS='-DOUCH="&\"#$*?"' '../My Source/ouch-1.0/configure' '../My Source/ouch-1.0/configure' CPPFLAGS='-DOUCH="&\"#$*?"'
To avoid checking for the same features repeatedly in various configure scripts (or in repeated runs of one script), configure can optionally save the results of many checks in a cache file (see Cache Files). If a configure script runs with caching enabled and finds a cache file, it reads the results of previous runs from the cache and avoids rerunning those checks. As a result, configure can then run much faster than if it had to perform all of the checks every time.
Ensure that the results of the check identified by cache-id are available. If the results of the check were in the cache file that was read, and configure was not given the --quiet or --silent option, print a message saying that the result was cached; otherwise, run the shell commands commands-to-set-it. If the shell commands are run to determine the value, the value is saved in the cache file just before configure creates its output files. See Cache Variable Names, for how to choose the name of the cache-id variable.
The commands-to-set-it must have no side effects except for setting the variable cache-id, see below.
A wrapper for
AC_CACHE_VAL
that takes care of printing the messages. This macro provides a convenient shorthand for the most common way to use these macros. It callsAC_MSG_CHECKING
for message, thenAC_CACHE_VAL
with the cache-id and commands arguments, andAC_MSG_RESULT
with cache-id.The commands-to-set-it must have no side effects except for setting the variable cache-id, see below.
It is common to find buggy macros using AC_CACHE_VAL
or
AC_CACHE_CHECK
, because people are tempted to call
AC_DEFINE
in the commands-to-set-it. Instead, the code that
follows the call to AC_CACHE_VAL
should call
AC_DEFINE
, by examining the value of the cache variable. For
instance, the following macro is broken:
AC_DEFUN([AC_SHELL_TRUE], [AC_CACHE_CHECK([whether true(1) works], [ac_cv_shell_true_works], [ac_cv_shell_true_works=no (true) 2>/dev/null && ac_cv_shell_true_works=yes if test "$ac_cv_shell_true_works" = yes; then AC_DEFINE([TRUE_WORKS], [1], [Define if `true(1)' works properly.]) fi]) ])
This fails if the cache is enabled: the second time this macro is run,
TRUE_WORKS
will not be defined. The proper implementation
is:
AC_DEFUN([AC_SHELL_TRUE], [AC_CACHE_CHECK([whether true(1) works], [ac_cv_shell_true_works], [ac_cv_shell_true_works=no (true) 2>/dev/null && ac_cv_shell_true_works=yes]) if test "$ac_cv_shell_true_works" = yes; then AC_DEFINE([TRUE_WORKS], [1], [Define if `true(1)' works properly.]) fi ])
Also, commands-to-set-it should not print any messages, for
example with AC_MSG_CHECKING
; do that before calling
AC_CACHE_VAL
, so the messages are printed regardless of whether
the results of the check are retrieved from the cache or determined by
running the shell commands.
The names of cache variables should have the following format:
package-prefix_cv_value-type_specific-value_[additional-options]
for example, ‘ac_cv_header_stat_broken’ or ‘ac_cv_prog_gcc_traditional’. The parts of the variable name are:
_cv_
The values assigned to cache variables may not contain newlines. Usually, their values are Boolean (‘yes’ or ‘no’) or the names of files or functions; so this is not an important restriction.
A cache file is a shell script that caches the results of configure tests run on one system so they can be shared between configure scripts and configure runs. It is not useful on other systems. If its contents are invalid for some reason, the user may delete or edit it.
By default, configure uses no cache file, to avoid problems caused by accidental use of stale cache files.
To enable caching, configure accepts --config-cache (or
-C) to cache results in the file config.cache.
Alternatively, --cache-file=file specifies that
file be the cache file. The cache file is created if it does not
exist already. When configure calls configure scripts in
subdirectories, it uses the --cache-file argument so that they
share the same cache. See Subdirectories, for information on
configuring subdirectories with the AC_CONFIG_SUBDIRS
macro.
config.status only pays attention to the cache file if it is given the --recheck option, which makes it rerun configure.
It is wrong to try to distribute cache files for particular system types. There is too much room for error in doing that, and too much administrative overhead in maintaining them. For any features that can't be guessed automatically, use the standard method of the canonical system type and linking files (see Manual Configuration).
The site initialization script can specify a site-wide cache file to use, instead of the usual per-program cache. In this case, the cache file gradually accumulates information whenever someone runs a new configure script. (Running configure merges the new cache results with the existing cache file.) This may cause problems, however, if the system configuration (e.g., the installed libraries or compilers) changes and the stale cache file is not deleted.
If your configure script, or a macro called from configure.ac, happens
to abort the configure process, it may be useful to checkpoint the cache
a few times at key points using AC_CACHE_SAVE
. Doing so
reduces the amount of time it takes to rerun the configure script with
(hopefully) the error that caused the previous abort corrected.
Loads values from existing cache file, or creates a new cache file if a cache file is not found. Called automatically from
AC_INIT
.
Flushes all cached values to the cache file. Called automatically from
AC_OUTPUT
, but it can be quite useful to callAC_CACHE_SAVE
at key points in configure.ac.
For instance:
... AC_INIT, etc. ... # Checks for programs. AC_PROG_CC AC_PROG_AWK ... more program checks ... AC_CACHE_SAVE # Checks for libraries. AC_CHECK_LIB([nsl], [gethostbyname]) AC_CHECK_LIB([socket], [connect]) ... more lib checks ... AC_CACHE_SAVE # Might abort... AM_PATH_GTK([1.0.2], [], [AC_MSG_ERROR([GTK not in path])]) AM_PATH_GTKMM([0.9.5], [], [AC_MSG_ERROR([GTK not in path])]) ... AC_OUTPUT, etc. ...
configure scripts need to give users running them several kinds of information. The following macros print messages in ways appropriate for each kind. The arguments to all of them get enclosed in shell double quotes, so the shell performs variable and back-quote substitution on them.
These macros are all wrappers around the echo shell command. They direct output to the appropriate file descriptor (see File Descriptor Macros). configure scripts should rarely need to run echo directly to print messages for the user. Using these macros makes it easy to change how and when each kind of message is printed; such changes need only be made to the macro definitions and all the callers change automatically.
To diagnose static issues, i.e., when autoconf is run, see Reporting Messages.
Notify the user that configure is checking for a particular feature. This macro prints a message that starts with ‘checking ’ and ends with ‘...’ and no newline. It must be followed by a call to
AC_MSG_RESULT
to print the result of the check and the newline. The feature-description should be something like ‘whether the Fortran compiler accepts C++ comments’ or ‘for c89’.This macro prints nothing if configure is run with the --quiet or --silent option.
Notify the user of the results of a check. result-description is almost always the value of the cache variable for the check, typically ‘yes’, ‘no’, or a file name. This macro should follow a call to
AC_MSG_CHECKING
, and the result-description should be the completion of the message printed by the call toAC_MSG_CHECKING
.This macro prints nothing if configure is run with the --quiet or --silent option.
Deliver the message to the user. It is useful mainly to print a general description of the overall purpose of a group of feature checks, e.g.,
AC_MSG_NOTICE([checking if stack overflow is detectable])This macro prints nothing if configure is run with the --quiet or --silent option.
Notify the user of an error that prevents configure from completing. This macro prints an error message to the standard error output and exits configure with exit-status (1 by default). error-description should be something like ‘invalid value $HOME for \$HOME’.
The error-description should start with a lower-case letter, and “cannot” is preferred to “can't”.
This
AC_MSG_ERROR
wrapper notifies the user of an error that prevents configure from completing and that additional details are provided in config.log. This is typically used when abnormal results are found during a compilation.
Notify the configure user of a possible problem. This macro prints the message to the standard error output; configure continues running afterward, so macros that call
AC_MSG_WARN
should provide a default (back-up) behavior for the situations they warn about. problem-description should be something like ‘ln -s seems to make hard links’.
Autoconf is written on top of two layers: M4sugar, which provides convenient macros for pure M4 programming, and M4sh, which provides macros dedicated to shell script generation.
As of this version of Autoconf, these two layers are still experimental, and their interface might change in the future. As a matter of fact, anything that is not documented must not be used.
The most common problem with existing macros is an improper quotation. This section, which users of Autoconf can skip, but which macro writers must read, first justifies the quotation scheme that was chosen for Autoconf and then ends with a rule of thumb. Understanding the former helps one to follow the latter.
To fully understand where proper quotation is important, you first need to know what the special characters are in Autoconf: ‘#’ introduces a comment inside which no macro expansion is performed, ‘,’ separates arguments, ‘[’ and ‘]’ are the quotes themselves, and finally ‘(’ and ‘)’ (which M4 tries to match by pairs).
In order to understand the delicate case of macro calls, we first have to present some obvious failures. Below they are “obvious-ified”, but when you find them in real life, they are usually in disguise.
Comments, introduced by a hash and running up to the newline, are opaque tokens to the top level: active characters are turned off, and there is no macro expansion:
# define([def], ine) =># define([def], ine)
Each time there can be a macro expansion, there is a quotation expansion, i.e., one level of quotes is stripped:
int tab[10]; =>int tab10; [int tab[10];] =>int tab[10];
Without this in mind, the reader might try hopelessly to use her macro
array
:
define([array], [int tab[10];]) array =>int tab10; [array] =>array
How can you correctly output the intended results3?
Let's proceed on the interaction between active characters and macros with this small macro, which just returns its first argument:
define([car], [$1])
The two pairs of quotes above are not part of the arguments of
define
; rather, they are understood by the top level when it
tries to find the arguments of define
. Therefore, assuming
car
is not already defined, it is equivalent to write:
define(car, $1)
But, while it is acceptable for a configure.ac to avoid unnecessary quotes, it is bad practice for Autoconf macros which must both be more robust and also advocate perfect style.
At the top level, there are only two possibilities: either you quote or you don't:
car(foo, bar, baz) =>foo [car(foo, bar, baz)] =>car(foo, bar, baz)
Let's pay attention to the special characters:
car(#) error-->EOF in argument list
The closing parenthesis is hidden in the comment; with a hypothetical quoting, the top level understood it this way:
car([#)]
Proper quotation, of course, fixes the problem:
car([#]) =>#
Here are more examples:
car(foo, bar) =>foo car([foo, bar]) =>foo, bar car((foo, bar)) =>(foo, bar) car([(foo], [bar)]) =>(foo define([a], [b]) => car(a) =>b car([a]) =>b car([[a]]) =>a car([[[a]]]) =>[a]
With this in mind, we can explore the cases where macros invoke macros...
The examples below use the following macros:
define([car], [$1]) define([active], [ACT, IVE]) define([array], [int tab[10]])
Each additional embedded macro call introduces other possible interesting quotations:
car(active) =>ACT car([active]) =>ACT, IVE car([[active]]) =>active
In the first case, the top level looks for the arguments of car
,
and finds ‘active’. Because M4 evaluates its arguments
before applying the macro, ‘active’ is expanded, which results in:
car(ACT, IVE) =>ACT
In the second case, the top level gives ‘active’ as first and only
argument of car
, which results in:
active =>ACT, IVE
i.e., the argument is evaluated after the macro that invokes it.
In the third case, car
receives ‘[active]’, which results in:
[active] =>active
exactly as we already saw above.
The example above, applied to a more realistic example, gives:
car(int tab[10];) =>int tab10; car([int tab[10];]) =>int tab10; car([[int tab[10];]]) =>int tab[10];
Huh? The first case is easily understood, but why is the second wrong,
and the third right? To understand that, you must know that after
M4 expands a macro, the resulting text is immediately subjected
to macro expansion and quote removal. This means that the quote removal
occurs twice—first before the argument is passed to the car
macro, and second after the car
macro expands to the first
argument.
As the author of the Autoconf macro car
, you then consider it to
be incorrect that your users have to double-quote the arguments of
car
, so you “fix” your macro. Let's call it qar
for
quoted car:
define([qar], [[$1]])
and check that qar
is properly fixed:
qar([int tab[10];]) =>int tab[10];
Ahhh! That's much better.
But note what you've done: now that the arguments are literal strings, if the user wants to use the results of expansions as arguments, she has to use an unquoted macro call:
qar(active) =>ACT
where she wanted to reproduce what she used to do with car
:
car([active]) =>ACT, IVE
Worse yet: she wants to use a macro that produces a set of cpp
macros:
define([my_includes], [#include <stdio.h>]) car([my_includes]) =>#include <stdio.h> qar(my_includes) error-->EOF in argument list
This macro, qar
, because it double quotes its arguments, forces
its users to leave their macro calls unquoted, which is dangerous.
Commas and other active symbols are interpreted by M4 before
they are given to the macro, often not in the way the users expect.
Also, because qar
behaves differently from the other macros,
it's an exception that should be avoided in Autoconf.
changequote
is Evil
The temptation is often high to bypass proper quotation, in particular
when it's late at night. Then, many experienced Autoconf hackers
finally surrender to the dark side of the force and use the ultimate
weapon: changequote
.
The M4 builtin changequote
belongs to a set of primitives that
allow one to adjust the syntax of the language to adjust it to one's
needs. For instance, by default M4 uses ‘`’ and ‘'’ as
quotes, but in the context of shell programming (and actually of most
programming languages), that's about the worst choice one can make:
because of strings and back-quoted expressions in shell code (such as
‘'this'’ and ‘`that`’), because of literal characters in usual
programming languages (as in ‘'0'’), there are many unbalanced
‘`’ and ‘'’. Proper M4 quotation then becomes a nightmare, if
not impossible. In order to make M4 useful in such a context, its
designers have equipped it with changequote
, which makes it
possible to choose another pair of quotes. M4sugar, M4sh, Autoconf, and
Autotest all have chosen to use ‘[’ and ‘]’. Not especially
because they are unlikely characters, but because they are
characters unlikely to be unbalanced.
There are other magic primitives, such as changecom
to specify
what syntactic forms are comments (it is common to see
‘changecom(<!--, -->)’ when M4 is used to produce HTML pages),
changeword
and changesyntax
to change other syntactic
details (such as the character to denote the nth argument, ‘$’ by
default, the parenthesis around arguments, etc.).
These primitives are really meant to make M4 more useful for specific domains: they should be considered like command line options: --quotes, --comments, --words, and --syntax. Nevertheless, they are implemented as M4 builtins, as it makes M4 libraries self contained (no need for additional options).
There lies the problem...
The problem is that it is then tempting to use them in the middle of an M4 script, as opposed to its initialization. This, if not carefully thought out, can lead to disastrous effects: you are changing the language in the middle of the execution. Changing and restoring the syntax is often not enough: if you happened to invoke macros in between, these macros are lost, as the current syntax is probably not the one they were implemented with.
When writing an Autoconf macro you may occasionally need to generate special characters that are difficult to express with the standard Autoconf quoting rules. For example, you may need to output the regular expression ‘[^[]’, which matches any character other than ‘[’. This expression contains unbalanced brackets so it cannot be put easily into an M4 macro.
You can work around this problem by using one of the following quadrigraphs:
Quadrigraphs are replaced at a late stage of the translation process, after m4 is run, so they do not get in the way of M4 quoting. For example, the string ‘^@<:@’, independently of its quotation, appears as ‘^[’ in the output.
The empty quadrigraph can be used:
Trailing spaces are smashed by autom4te. This is a feature.
For instance ‘@<@&t@:@’ produces ‘@<:@’.
For instance you might want to mention AC_FOO
in a comment, while
still being sure that autom4te still catches unexpanded
‘AC_*’. Then write ‘AC@&t@_FOO’.
The name ‘@&t@’ was suggested by Paul Eggert:
I should give some credit to the ‘@&t@’ pun. The ‘&’ is my own invention, but the ‘t’ came from the source code of the algol68c compiler, written by Steve Bourne (of Bourne shell fame), and which used ‘mt’ to denote the empty string. In C, it would have looked like something like:char const mt[] = "";but of course the source code was written in Algol 68.
I don't know where he got ‘mt’ from: it could have been his own invention, and I suppose it could have been a common pun around the Cambridge University computer lab at the time.
To conclude, the quotation rule of thumb is:
Never over-quote, never under-quote, in particular in the definition of macros. In the few places where the macros need to use brackets (usually in C program text or regular expressions), properly quote the arguments!
It is common to read Autoconf programs with snippets like:
AC_TRY_LINK( changequote(<<, >>)dnl <<#include <time.h> #ifndef tzname /* For SGI. */ extern char *tzname[]; /* RS6000 and others reject char **tzname. */ #endif>>, changequote([, ])dnl [atoi (*tzname);], ac_cv_var_tzname=yes, ac_cv_var_tzname=no)
which is incredibly useless since AC_TRY_LINK
is already
double quoting, so you just need:
AC_TRY_LINK( [#include <time.h> #ifndef tzname /* For SGI. */ extern char *tzname[]; /* RS6000 and others reject char **tzname. */ #endif], [atoi (*tzname);], [ac_cv_var_tzname=yes], [ac_cv_var_tzname=no])
The M4-fluent reader might note that these two examples are rigorously equivalent, since M4 swallows both the ‘changequote(<<, >>)’ and ‘<<’ ‘>>’ when it collects the arguments: these quotes are not part of the arguments!
Simplified, the example above is just doing this:
changequote(<<, >>)dnl <<[]>> changequote([, ])dnl
instead of simply:
[[]]
With macros that do not double quote their arguments (which is the rule), double-quote the (risky) literals:
AC_LINK_IFELSE([AC_LANG_PROGRAM( [[#include <time.h> #ifndef tzname /* For SGI. */ extern char *tzname[]; /* RS6000 and others reject char **tzname. */ #endif]], [atoi (*tzname);])], [ac_cv_var_tzname=yes], [ac_cv_var_tzname=no])
Please note that the macro AC_TRY_LINK
is obsolete, so you really
should be using AC_LINK_IFELSE
instead.
See Quadrigraphs, for what to do if you run into a hopeless case where quoting does not suffice.
When you create a configure script using newly written macros, examine it carefully to check whether you need to add more quotes in your macros. If one or more words have disappeared in the M4 output, you need more quotes. When in doubt, quote.
However, it's also possible to put on too many layers of quotes. If
this happens, the resulting configure script may contain
unexpanded macros. The autoconf program checks for this problem
by looking for the string ‘AC_’ in configure. However, this
heuristic does not work in general: for example, it does not catch
overquoting in AC_DEFINE
descriptions.
The Autoconf suite, including M4sugar, M4sh, and Autotest, in addition to Autoconf per se, heavily rely on M4. All these different uses revealed common needs factored into a layer over M4: autom4te4.
autom4te is a preprocessor that is like m4. It supports M4 extensions designed for use in tools like Autoconf.
The command line arguments are modeled after M4's:
autom4te options files
where the files are directly passed to m4. By default, GNU M4 is found during configuration, but the environment variable M4 can be set to tell autom4te where to look. In addition to the regular expansion, it handles the replacement of the quadrigraphs (see Quadrigraphs), and of ‘__oline__’, the current line in the output. It supports an extended syntax for the files:
Of course, it supports the Autoconf common subset of options:
As an extension of m4, it includes the following options:
AC_DIAGNOSE
, for a comprehensive list of categories. Special
values include:
Warnings about ‘syntax’ are enabled by default, and the environment variable WARNINGS, a comma separated list of categories, is honored. ‘autom4te -W category’ actually behaves as if you had run:
autom4te --warnings=syntax,$WARNINGS,category
For example, if you want to disable defaults and WARNINGS of autom4te, but enable the warnings about obsolete constructs, you would use -W none,obsolete.
autom4te displays a back trace for errors, but not for
warnings; if you want them, just pass -W error.
.m4f
is
replaced by file.m4
. This helps tracing the macros which
are executed only when the files are frozen, typically
m4_define
. For instance, running:
autom4te --melt 1.m4 2.m4f 3.m4 4.m4f input.m4
is roughly equivalent to running:
m4 1.m4 2.m4 3.m4 4.m4 input.m4
while
autom4te 1.m4 2.m4f 3.m4 4.m4f input.m4
is equivalent to:
m4 --reload-state=4.m4f input.m4
autom4te 1.m4 2.m4 3.m4 --freeze --output=3.m4f
corresponds to
m4 1.m4 2.m4 3.m4 --freeze-state=3.m4f
As another additional feature over m4, autom4te caches its results. GNU M4 is able to produce a regular output and traces at the same time. Traces are heavily used in the GNU Build System: autoheader uses them to build config.h.in, autoreconf to determine what GNU Build System components are used, automake to “parse” configure.ac etc. To avoid recomputation, traces are cached while performing regular expansion, and conversely. This cache is (actually, the caches are) stored in the directory autom4te.cache. It can safely be removed at any moment (especially if for some reason autom4te considers it is trashed).
Because traces are so important to the GNU Build System, autom4te provides high level tracing features as compared to M4, and helps exploiting the cache:
The format is a regular string, with newlines if desired, and several special escape codes. It defaults to ‘$f:$l:$n:$%’. It can use the following special escapes:
The escape ‘$%’ produces single-line trace outputs (unless you put newlines in the ‘separator’), while ‘$@’ and ‘$*’ do not.
See autoconf Invocation, for examples of trace uses.
Finally, autom4te introduces the concept of Autom4te libraries. They consists in a powerful yet extremely simple feature: sets of combined command line arguments:
M4sugar
M4sh
Autotest
Autoconf-without-aclocal-m4
Autoconf
Autoconf-without-aclocal-m4
and
additionally reads aclocal.m4.
As an example, if Autoconf is installed in its default location, /usr/local, the command ‘autom4te -l m4sugar foo.m4’ is strictly equivalent to the command:
autom4te --prepend-include /usr/local/share/autoconf \ m4sugar/m4sugar.m4f --warnings syntax foo.m4
Recursive expansion applies here: the command ‘autom4te -l m4sh foo.m4’ is the same as ‘autom4te --language M4sugar m4sugar/m4sh.m4f foo.m4’, i.e.:
autom4te --prepend-include /usr/local/share/autoconf \ m4sugar/m4sugar.m4f m4sugar/m4sh.m4f --mode 777 foo.m4
The definition of the languages is stored in autom4te.cfg.
One can customize autom4te via ~/.autom4te.cfg (i.e., as found in the user home directory), and ./.autom4te.cfg (i.e., as found in the directory from which autom4te is run). The order is first reading autom4te.cfg, then ~/.autom4te.cfg, then ./.autom4te.cfg, and finally the command line arguments.
In these text files, comments are introduced with #
, and empty
lines are ignored. Customization is performed on a per-language basis,
wrapped in between a ‘begin-language: "language"’,
‘end-language: "language"’ pair.
Customizing a language stands for appending options (see autom4te Invocation) to the current definition of the language. Options, and more generally arguments, are introduced by ‘args: arguments’. You may use the traditional shell syntax to quote the arguments.
As an example, to disable Autoconf caches (autom4te.cache) globally, include the following lines in ~/.autom4te.cfg:
## ------------------ ## ## User Preferences. ## ## ------------------ ## begin-language: "Autoconf-without-aclocal-m4" args: --no-cache end-language: "Autoconf-without-aclocal-m4"
M4 by itself provides only a small, but sufficient, set of all-purpose macros. M4sugar introduces additional generic macros. Its name was coined by Lars J. Aas: “Readability And Greater Understanding Stands 4 M4sugar”.
With a few exceptions, all the M4 native macros are moved in the
‘m4_’ pseudo-namespace, e.g., M4sugar renames define
as
m4_define
etc.
Some M4 macros are redefined, and are slightly incompatible with their native equivalent.
Unlike the M4 builtin, this macro fails if macro is not defined. See
m4_undefine
.
Like the M4 builtins, but warn against multiple inclusions of file.
This macro corresponds to
patsubst
. The namem4_patsubst
is kept for future versions of M4sh, on top of GNU M4 which will provide extended regular expression syntax viaepatsubst
.
Unlike the M4 builtin, this macro fails if macro is not defined. See
m4_undefine
.
This macro corresponds to
regexp
. The namem4_regexp
is kept for future versions of M4sh, on top of GNU M4 which will provide extended regular expression syntax viaeregexp
.
This macro corresponds to
m4wrap
.Posix requires arguments of multiple
m4wrap
calls to be reprocessed at EOF in the same order as the original calls. GNU M4 versions through 1.4.x, however, reprocess them in reverse order. Your code should not depend on the order.Also, Posix requires
m4wrap
to ignore its second and succeeding arguments, but GNU M4 versions through 1.4.x concatenate the arguments with intervening spaces. Your code should not pass more than one argument.You are encouraged to end text with ‘[]’, to avoid unexpected token pasting between consecutive invocations of
m4_wrap
, as in:m4_define([foo], [bar]) m4_define([foofoo], [OUCH]) m4_wrap([foo]) m4_wrap([foo]) =>OUCH
Unlike the M4 builtin, this macro fails if macro is not defined. Use
m4_ifdef([macro], [m4_undefine([macro])])to recover the behavior of the builtin.
The following macros implement loops in M4.
Loop over the numeric values between first and last including bounds by increments of step. For each iteration, expand expression with the numeric value assigned to var. If step is omitted, it defaults to ‘1’ or ‘-1’ depending on the order of the limits. If given, step has to match this order.
Loop over the comma-separated M4 list list, assigning each value to var, and expand expression. The following example outputs two lines:
m4_foreach([myvar], [[foo], [bar, baz]], [echo myvar ])
Loop over the whitespace-separated list list, assigning each value to var, and expand expression.
The deprecated macro
AC_FOREACH
is an alias ofm4_foreach_w
.
The following macros give some control over the order of the evaluation by adding or removing levels of quotes. They are meant for hard-core M4 programmers.
Return the arguments as a single entity, i.e., wrap them into a pair of quotes.
The following example aims at emphasizing the difference between (i), not
using these macros, (ii), using m4_quote
, and (iii), using
m4_dquote
.
$ cat example.m4 # Overquote, so that quotes are visible. m4_define([show], [$[]1 = [$1], $[]@ = [$@]]) m4_divert(0)dnl show(a, b) show(m4_quote(a, b)) show(m4_dquote(a, b)) $ autom4te -l m4sugar example.m4 $1 = a, $@ = [a],[b] $1 = a,b, $@ = [a,b] $1 = [a],[b], $@ = [[a],[b]]
The following macros may be used to manipulate strings in M4. They are not intended for casual use.
Backslash-escape all characters in string that are active in regexps.
Return string with letters converted to upper or lower case, respectively.
Split string into an M4 list of elements quoted by ‘[’ and ‘]’, while keeping white space at the beginning and at the end. If regexp is given, use it instead of ‘[\t ]+’ for splitting. If string is empty, the result is an empty list.
Remove leading and trailing spaces and tabs, sequences of backslash-then-newline, and replace multiple spaces and tabs with a single space.
Redefine macro-name to its former contents with separator and string added at the end. If macro-name was undefined before (but not if it was defined but empty), then no separator is added.
m4_append
can be used to grow strings, andm4_append_uniq
to grow strings without duplicating substrings.
M4sugar provides a means to define suspicious patterns, patterns describing tokens which should not be found in the output. For instance, if an Autoconf configure script includes tokens such as ‘AC_DEFINE’, or ‘dnl’, then most probably something went wrong (typically a macro was not evaluated because of overquotation).
M4sugar forbids all the tokens matching ‘^m4_’ and ‘^dnl$’.
Declare that no token matching pattern must be found in the output. Comments are not checked; this can be a problem if, for instance, you have some macro left unexpanded after an ‘#include’. No consensus is currently found in the Autoconf community, as some people consider it should be valid to name macros in comments (which doesn't make sense to the author of this documentation, as ‘#’-comments should document the output, not the input, documented by ‘dnl’ comments).
Of course, you might encounter exceptions to these generic rules, for instance you might have to refer to ‘$m4_flags’.
Any token matching pattern is allowed, including if it matches an
m4_pattern_forbid
pattern.
M4sh, pronounced “mash”, is aiming at producing portable Bourne shell scripts. This name was coined by Lars J. Aas, who notes that, according to the Webster's Revised Unabridged Dictionary (1913):
Mash \Mash\, n. [Akin to G. meisch, maisch, meische, maische, mash, wash, and prob. to AS. miscian to mix. See “Mix”.]
- A mass of mixed ingredients reduced to a soft pulpy state by beating or pressure...
- A mixture of meal or bran and water fed to animals.
- A mess; trouble. [Obs.] –Beau. & Fl.
For the time being, it is not mature enough to be widely used.
M4sh provides portable alternatives for some common shell constructs that unfortunately are not portable in practice.
Set up the shell to be more compatible with the Bourne shell as standardized by Posix, if possible. This may involve setting environment variables, or setting options, or similar implementation-specific actions.
Expand into a shell ‘case’ statement, where word is matched against one or more patterns. if-matched is run if the corresponding pattern matched word, else default is run.
Output the directory portion of file-name. For example, if
$file
is ‘/one/two/three’, the commanddir=`AS_DIRNAME(["$file"])`
setsdir
to ‘/one/two’.
Run shell code test1. If test1 exits with a zero status then run shell code run-if-true1, else examine further tests. If no test exits with a zero status, run shell code run-if-false, with simplifications if either run-if-true1 or run-if-false1 is empty. For example,
AS_IF([test "$foo" = yes], [HANDLE_FOO([yes])], [test "$foo" != no], [HANDLE_FOO([maybe])], [echo foo not specified])ensures any required macros of
HANDLE_FOO
are expanded before the first test.
Make the directory file-name, including intervening directories as necessary. This is equivalent to ‘mkdir -p file-name’, except that it is portable to older versions of mkdir that lack support for the -p option. Also,
AS_MKDIR_P
succeeds if file-name is a symbolic link to an existing directory, even though Posix is unclear whether ‘mkdir -p’ should succeed in that case. If creation of file-name fails, exit the script.Also see the
AC_PROG_MKDIR_P
macro (see Particular Programs).
Initialize the shell suitably for
configure
scripts. This has the effect ofAS_BOURNE_COMPATIBLE
, and sets some other environment variables for predictable results from configuration tests. For example, it sets LC_ALL to change to the default C locale. See Special Shell Variables.
Transform expression into a valid right-hand side for a C
#define
. For example:# This outputs "#define HAVE_CHAR_P 1". type="char *" echo "#define AS_TR_CPP([HAVE_$type]) 1"
Transform expression into a valid shell variable name. For example:
# This outputs "Have it!". header="sys/some file.h" AS_TR_SH([HAVE_$header])=yes if test "$HAVE_sys_some_file_h" = yes; then echo "Have it!"; fi
Set the shell variable var to dir/file, but optimizing the common cases (dir or file is ‘.’, file is absolute, etc.).
The following macros define file descriptors used to output messages (or input values) from configure scripts. For example:
echo "$wombats found" >&AS_MESSAGE_LOG_FD echo 'Enter desired kangaroo count:' >&AS_MESSAGE_FD read kangaroos <&AS_ORIGINAL_STDIN_FD`
However doing so is seldom needed, because Autoconf provides higher level macros as described below.
The file descriptor for ‘checking for...’ messages and results. Normally this directs messages to the standard output, however when configure is run with the -q option, messages sent to
AS_MESSAGE_FD
are discarded.If you want to display some messages, consider using one of the printing macros (see Printing Messages) instead. Copies of messages output via these macros are also recorded in config.log.
The file descriptor for messages logged to config.log. Macros that run tools, like
AC_COMPILE_IFELSE
(see Running the Compiler), redirect all output to this descriptor. You may want to do so if you develop such a low-level macro.
The file descriptor for the original standard input.
When configure runs, it may accidentally execute an interactive command that has the same name as the non-interactive meant to be used or checked. If the standard input was the terminal, such interactive programs would cause configure to stop, pending some user input. Therefore configure redirects its standard input from /dev/null during its initialization. This is not normally a problem, since configure normally does not need user input.
In the extreme case where your configure script really needs to obtain some values from the original standard input, you can read them explicitly from
AS_ORIGINAL_STDIN_FD
.
When you write a feature test that could be applicable to more than one software package, the best thing to do is encapsulate it in a new macro. Here are some instructions and guidelines for writing Autoconf macros.
Autoconf macros are defined using the AC_DEFUN
macro, which is
similar to the M4 builtin m4_define
macro. In addition to
defining a macro, AC_DEFUN
adds to it some code that is used to
constrain the order in which macros are called (see Prerequisite Macros).
An Autoconf macro definition looks like this:
AC_DEFUN(macro-name, macro-body)
You can refer to any arguments passed to the macro as ‘$1’, ‘$2’, etc. See How to define new macros, for more complete information on writing M4 macros.
Be sure to properly quote both the macro-body and the macro-name to avoid any problems if the macro happens to have been previously defined.
Each macro should have a header comment that gives its prototype, and a brief description. When arguments have default values, display them in the prototype. For example:
# AC_MSG_ERROR(ERROR, [EXIT-STATUS = 1]) # -------------------------------------- m4_define([AC_MSG_ERROR], [{ AS_MESSAGE([error: $1], [2]) exit m4_default([$2], [1]); }])
Comments about the macro should be left in the header comment. Most other comments make their way into configure, so just keep using ‘#’ to introduce comments.
If you have some special comments about pure M4 code, comments
that make no sense in configure and in the header comment, then
use the builtin dnl
: it causes M4 to discard the text
through the next newline.
Keep in mind that dnl
is rarely needed to introduce comments;
dnl
is more useful to get rid of the newlines following macros
that produce no output, such as AC_REQUIRE
.
All of the Autoconf macros have all-uppercase names starting with ‘AC_’ to prevent them from accidentally conflicting with other text. All shell variables that they use for internal purposes have mostly-lowercase names starting with ‘ac_’. To ensure that your macros don't conflict with present or future Autoconf macros, you should prefix your own macro names and any shell variables they use with some other sequence. Possibilities include your initials, or an abbreviation for the name of your organization or software package.
Most of the Autoconf macros' names follow a structured naming convention that indicates the kind of feature check by the name. The macro names consist of several words, separated by underscores, going from most general to most specific. The names of their cache variables use the same convention (see Cache Variable Names, for more information on them).
The first word of the name after ‘AC_’ usually tells the category of the feature being tested. Here are the categories used in Autoconf for specific test macros, the kind of macro that you are more likely to write. They are also used for cache variables, in all-lowercase. Use them where applicable; where they're not, invent your own categories.
C
DECL
FUNC
GROUP
HEADER
LIB
PATH
PROG
MEMBER
SYS
TYPE
VAR
After the category comes the name of the particular feature being
tested. Any further words in the macro name indicate particular aspects
of the feature. For example, AC_FUNC_FNMATCH_GNU
checks whether
the fnmatch
function supports GNU extensions.
An internal macro should have a name that starts with an underscore;
Autoconf internals should therefore start with ‘_AC_’.
Additionally, a macro that is an internal subroutine of another macro
should have a name that starts with an underscore and the name of that
other macro, followed by one or more words saying what the internal
macro does. For example, AC_PATH_X
has internal macros
_AC_PATH_X_XMKMF
and _AC_PATH_X_DIRECT
.
When macros statically diagnose abnormal situations, benign or fatal, they should report them using these macros. For dynamic issues, i.e., when configure is run, see Printing Messages.
Report message as a warning (or as an error if requested by the user) if warnings of the category are turned on. You are encouraged to use standard categories, which currently include:
- ‘all’
- messages that don't fall into one of the following categories. Use of an empty category is equivalent.
- ‘cross’
- related to cross compilation issues.
- ‘obsolete’
- use of an obsolete construct.
- ‘syntax’
- dubious syntactic constructs, incorrectly ordered macro calls.
Equivalent to ‘AC_DIAGNOSE([syntax], message)’, but you are strongly encouraged to use a finer grained category.
When the user runs ‘autoconf -W error’, warnings from
AC_DIAGNOSE
and AC_WARNING
are reported as error, see
autoconf Invocation.
Some Autoconf macros depend on other macros having been called first in order to work correctly. Autoconf provides a way to ensure that certain macros are called if needed and a way to warn the user if macros are called in an order that might cause incorrect operation.
A macro that you write might need to use values that have previously
been computed by other macros. For example, AC_DECL_YYTEXT
examines the output of flex
or lex
, so it depends on
AC_PROG_LEX
having been called first to set the shell variable
LEX
.
Rather than forcing the user of the macros to keep track of the
dependencies between them, you can use the AC_REQUIRE
macro to do
it automatically. AC_REQUIRE
can ensure that a macro is only
called if it is needed, and only called once.
If the M4 macro macro-name has not already been called, call it (without any arguments). Make sure to quote macro-name with square brackets. macro-name must have been defined using
AC_DEFUN
or else contain a call toAC_PROVIDE
to indicate that it has been called.
AC_REQUIRE
must be used inside a macro defined byAC_DEFUN
; it must not be called from the top level.
AC_REQUIRE
is often misunderstood. It really implements
dependencies between macros in the sense that if one macro depends upon
another, the latter is expanded before the body of the
former. To be more precise, the required macro is expanded before
the outermost defined macro in the current expansion stack.
In particular, ‘AC_REQUIRE([FOO])’ is not replaced with the body of
FOO
. For instance, this definition of macros:
AC_DEFUN([TRAVOLTA], [test "$body_temperature_in_celsius" -gt "38" && dance_floor=occupied]) AC_DEFUN([NEWTON_JOHN], [test "$hair_style" = "curly" && dance_floor=occupied]) AC_DEFUN([RESERVE_DANCE_FLOOR], [if date | grep '^Sat.*pm' >/dev/null 2>&1; then AC_REQUIRE([TRAVOLTA]) AC_REQUIRE([NEWTON_JOHN]) fi])
with this configure.ac
AC_INIT([Dance Manager], [1.0], [bug-dance@example.org]) RESERVE_DANCE_FLOOR if test "$dance_floor" = occupied; then AC_MSG_ERROR([cannot pick up here, let's move]) fi
does not leave you with a better chance to meet a kindred soul at other times than Saturday night since it expands into:
test "$body_temperature_in_Celsius" -gt "38" && dance_floor=occupied test "$hair_style" = "curly" && dance_floor=occupied fi if date | grep '^Sat.*pm' >/dev/null 2>&1; then fi
This behavior was chosen on purpose: (i) it prevents messages in required macros from interrupting the messages in the requiring macros; (ii) it avoids bad surprises when shell conditionals are used, as in:
if ...; then AC_REQUIRE([SOME_CHECK]) fi ... SOME_CHECK
The helper macros AS_IF
and AS_CASE
may be used to
enforce expansion of required macros outside of shell conditional
constructs. You are furthermore encouraged to put all AC_REQUIRE
calls
at the beginning of a macro. You can use dnl
to avoid the empty
lines they leave.
Some macros should be run before another macro if both are called, but neither requires that the other be called. For example, a macro that changes the behavior of the C compiler should be called before any macros that run the C compiler. Many of these dependencies are noted in the documentation.
Autoconf provides the AC_BEFORE
macro to warn users when macros
with this kind of dependency appear out of order in a
configure.ac file. The warning occurs when creating
configure from configure.ac, not when running
configure.
For example, AC_PROG_CPP
checks whether the C compiler
can run the C preprocessor when given the -E option. It should
therefore be called after any macros that change which C compiler is
being used, such as AC_PROG_CC
. So AC_PROG_CC
contains:
AC_BEFORE([$0], [AC_PROG_CPP])dnl
This warns the user if a call to AC_PROG_CPP
has already occurred
when AC_PROG_CC
is called.
Make M4 print a warning message to the standard error output if called-macro-name has already been called. this-macro-name should be the name of the macro that is calling
AC_BEFORE
. The macro called-macro-name must have been defined usingAC_DEFUN
or else contain a call toAC_PROVIDE
to indicate that it has been called.
Some macros should be called only once, either because calling them
multiple time is unsafe, or because it is bad style. For instance
Autoconf ensures that AC_CANONICAL_BUILD
and cousins
(see Canonicalizing) are evaluated only once, because it makes no
sense to run these expensive checks more than once. Such one-shot
macros can be defined using AC_DEFUN_ONCE
.
Declare macro macro-name like
AC_DEFUN
would (see Macro Definitions), and emit a warning any time the macro is called more than once.
Obviously it is not sensible to evaluate a macro defined by
AC_DEFUN_ONCE
in a macro defined by AC_DEFUN
.
Most of the time you want to use AC_REQUIRE
(see Prerequisite Macros).
Configuration and portability technology has evolved over the years. Often better ways of solving a particular problem are developed, or ad-hoc approaches are systematized. This process has occurred in many parts of Autoconf. One result is that some of the macros are now considered obsolete; they still work, but are no longer considered the best thing to do, hence they should be replaced with more modern macros. Ideally, autoupdate should replace the old macro calls with their modern implementation.
Autoconf provides a simple means to obsolete a macro.
Define old-macro as implementation. The only difference with
AC_DEFUN
is that the user is warned that old-macro is now obsolete.If she then uses autoupdate, the call to old-macro is replaced by the modern implementation. message should include information on what to do after running autoupdate; autoupdate prints it as a warning, and includes it in the updated configure.ac file.
The details of this macro are hairy: if autoconf encounters an
AU_DEFUN
ed macro, all macros inside its second argument are expanded as usual. However, when autoupdate is run, only M4 and M4sugar macros are expanded here, while all other macros are disabled and appear literally in the updated configure.ac.
Used if the old-name is to be replaced by a call to new-macro with the same parameters. This happens for example if the macro was renamed.
The Autoconf macros follow a strict coding style. You are encouraged to follow this style, especially if you intend to distribute your macro, either by contributing it to Autoconf itself, or via other means.
The first requirement is to pay great attention to the quotation. For more details, see Autoconf Language, and M4 Quotation.
Do not try to invent new interfaces. It is likely that there is a macro in Autoconf that resembles the macro you are defining: try to stick to this existing interface (order of arguments, default values, etc.). We are conscious that some of these interfaces are not perfect; nevertheless, when harmless, homogeneity should be preferred over creativity.
Be careful about clashes both between M4 symbols and between shell variables.
If you stick to the suggested M4 naming scheme (see Macro Names),
you are unlikely to generate conflicts. Nevertheless, when you need to
set a special value, avoid using a regular macro name; rather,
use an “impossible” name. For instance, up to version 2.13, the macro
AC_SUBST
used to remember what symbol macros were already defined
by setting AC_SUBST_
symbol, which is a regular macro name.
But since there is a macro named AC_SUBST_FILE
, it was just
impossible to ‘AC_SUBST(FILE)’! In this case,
AC_SUBST(
symbol)
or _AC_SUBST(
symbol)
should
have been used (yes, with the parentheses).
No Autoconf macro should ever enter the user-variable name space; i.e.,
except for the variables that are the actual result of running the
macro, all shell variables should start with ac_
. In
addition, small macros or any macro that is likely to be embedded in
other macros should be careful not to use obvious names.
Do not use dnl
to introduce comments: most of the comments you
are likely to write are either header comments which are not output
anyway, or comments that should make their way into configure.
There are exceptional cases where you do want to comment special M4
constructs, in which case dnl
is right, but keep in mind that it
is unlikely.
M4 ignores the leading blanks and newlines before each argument. Use this feature to indent in such a way that arguments are (more or less) aligned with the opening parenthesis of the macro being called. For instance, instead of
AC_CACHE_CHECK(for EMX OS/2 environment, ac_cv_emxos2, [AC_COMPILE_IFELSE([AC_LANG_PROGRAM(, [return __EMX__;])], [ac_cv_emxos2=yes], [ac_cv_emxos2=no])])
write
AC_CACHE_CHECK([for EMX OS/2 environment], [ac_cv_emxos2], [AC_COMPILE_IFELSE([AC_LANG_PROGRAM([], [return __EMX__;])], [ac_cv_emxos2=yes], [ac_cv_emxos2=no])])
or even
AC_CACHE_CHECK([for EMX OS/2 environment], [ac_cv_emxos2], [AC_COMPILE_IFELSE([AC_LANG_PROGRAM([], [return __EMX__;])], [ac_cv_emxos2=yes], [ac_cv_emxos2=no])])
When using AC_RUN_IFELSE
or any macro that cannot work when
cross-compiling, provide a pessimistic value (typically ‘no’).
Feel free to use various tricks to prevent auxiliary tools, such as syntax-highlighting editors, from behaving improperly. For instance, instead of:
m4_bpatsubst([$1], [$"])
use
m4_bpatsubst([$1], [$""])
so that Emacsen do not open an endless “string” at the first quote. For the same reasons, avoid:
test $[#] != 0
and use:
test $[@%:@] != 0
Otherwise, the closing bracket would be hidden inside a ‘#’-comment,
breaking the bracket-matching highlighting from Emacsen. Note the
preferred style to escape from M4: ‘$[1]’, ‘$[@]’, etc. Do
not escape when it is unnecessary. Common examples of useless quotation
are ‘[$]$1’ (write ‘$$1’), ‘[$]var’ (use ‘$var’),
etc. If you add portability issues to the picture, you'll prefer
‘${1+"$[@]"}’ to ‘"[$]@"’, and you'll prefer do something
better than hacking Autoconf :-)
.
When using sed, don't use -e except for indenting
purposes. With the s
and y
commands, the preferred
separator is ‘/’ unless ‘/’ itself might appear in the pattern
or replacement, in which case you should use ‘|’, or optionally
‘,’ if you know the pattern and replacement cannot contain a file
name. If none of these characters will do, choose a printable character
that cannot appear in the pattern or replacement. Characters from the
set ‘"#$&'()*;<=>?`|~’ are good choices if the pattern or
replacement might contain a file name, since they have special meaning
to the shell and are less likely to occur in file names.
See Macro Definitions, for details on how to define a macro. If a
macro doesn't use AC_REQUIRE
, is expected to never be the object
of an AC_REQUIRE
directive, and macros required by other macros
inside arguments do not need to be expanded before this macro, then
use m4_define
. In case of doubt, use AC_DEFUN
.
All the AC_REQUIRE
statements should be at the beginning of the
macro, and each statement should be followed by dnl
.
You should not rely on the number of arguments: instead of checking whether an argument is missing, test that it is not empty. It provides both a simpler and a more predictable interface to the user, and saves room for further arguments.
Unless the macro is short, try to leave the closing ‘])’ at the beginning of a line, followed by a comment that repeats the name of the macro being defined. This introduces an additional newline in configure; normally, that is not a problem, but if you want to remove it you can use ‘[]dnl’ on the last line. You can similarly use ‘[]dnl’ after a macro call to remove its newline. ‘[]dnl’ is recommended instead of ‘dnl’ to ensure that M4 does not interpret the ‘dnl’ as being attached to the preceding text or macro output. For example, instead of:
AC_DEFUN([AC_PATH_X],
[AC_MSG_CHECKING([for X])
AC_REQUIRE_CPP()
# ...omitted...
AC_MSG_RESULT([libraries $x_libraries, headers $x_includes])
fi])
you would write:
AC_DEFUN([AC_PATH_X],
[AC_REQUIRE_CPP()[]dnl
AC_MSG_CHECKING([for X])
# ...omitted...
AC_MSG_RESULT([libraries $x_libraries, headers $x_includes])
fi[]dnl
])# AC_PATH_X
If the macro is long, try to split it into logical chunks. Typically,
macros that check for a bug in a function and prepare its
AC_LIBOBJ
replacement should have an auxiliary macro to perform
this setup. Do not hesitate to introduce auxiliary macros to factor
your code.
In order to highlight the recommended coding style, here is a macro written the old way:
dnl Check for EMX on OS/2. dnl _AC_EMXOS2 AC_DEFUN(_AC_EMXOS2, [AC_CACHE_CHECK(for EMX OS/2 environment, ac_cv_emxos2, [AC_COMPILE_IFELSE([AC_LANG_PROGRAM(, return __EMX__;)], ac_cv_emxos2=yes, ac_cv_emxos2=no)]) test "$ac_cv_emxos2" = yes && EMXOS2=yes])
and the new way:
# _AC_EMXOS2 # ---------- # Check for EMX on OS/2. m4_define([_AC_EMXOS2], [AC_CACHE_CHECK([for EMX OS/2 environment], [ac_cv_emxos2], [AC_COMPILE_IFELSE([AC_LANG_PROGRAM([], [return __EMX__;])], [ac_cv_emxos2=yes], [ac_cv_emxos2=no])]) test "$ac_cv_emxos2" = yes && EMXOS2=yes[]dnl ])# _AC_EMXOS2
When writing your own checks, there are some shell-script programming techniques you should avoid in order to make your code portable. The Bourne shell and upward-compatible shells like the Korn shell and Bash have evolved over the years, but to prevent trouble, do not take advantage of features that were added after Unix version 7, circa 1977 (see Systemology).
You should not use shell functions, aliases, negated character
classes, or other features that are not found in all Bourne-compatible
shells; restrict yourself to the lowest common denominator. Even
unset
is not supported by all shells!
Some ancient systems have quite small limits on the length of the ‘#!’ line; for instance, 32 bytes (not including the newline) on SunOS 4. A few ancient 4.2BSD based systems (such as Dynix circa 1984) required a single space between the ‘#!’ and the ‘/’. However, these ancient systems are no longer of practical concern.
The set of external programs you should run in a configure script is fairly small. See Utilities in Makefiles, for the list. This restriction allows users to start out with a fairly small set of programs and build the rest, avoiding too many interdependencies between packages.
Some of these external utilities have a portable subset of features; see Limitations of Usual Tools.
There are other sources of documentation about shells. The specification for the Posix Shell Command Language, though more generous than the restrictive shell subset described above, is fairly portable nowadays. Also please see the Shell FAQs.
There are several families of shells, most prominently the Bourne family and the C shell family which are deeply incompatible. If you want to write portable shell scripts, avoid members of the C shell family. The the Shell difference FAQ includes a small history of Posix shells, and a comparison between several of them.
Below we describe some of the members of the Bourne shell family.
To be compatible with Ash 0.2:
foo= false $foo echo "Do not use it: $?" false eval 'echo "Do not use it: $?"'
cat ${FOO=`bar`}
BASH_VERSION
is set. To require
Posix compatibility, run ‘set -o posix’. See Bash Posix Mode, for details.
Solaris systems have three variants: /usr/bin/ksh is ‘ksh88’; it is standard on Solaris 2.0 and later. /usr/xpg4/bin/sh is a Posix-compliant variant of ‘ksh88’; it is standard on Solaris 9 and later. /usr/dt/bin/dtksh is ‘ksh93’. Variants that are not standard may be parts of optional packages. There is no extra charge for these packages, but they are not part of a minimal OS install and therefore some installations may not have it.
Starting with Tru64 Version 4.0, the Korn shell /usr/bin/ksh
is also available as /usr/bin/posix/sh. If the environment
variable BIN_SH is set to xpg4
, subsidiary invocations of
the standard shell conform to Posix.
KSH_VERSION
, except if invoked as
/bin/sh on OpenBSD, and similarly to Bash you can require
Posix compatibility by running ‘set -o posix’. Unfortunately, with
pdksh 5.2.14 (the latest stable version as of February 2006)
Posix mode is buggy and causes pdksh to depart from Posix in
at least one respect:
$ echo "`echo \"hello\"`" hello $ set -o posix $ echo "`echo \"hello\"`" "hello"
The last line of output contains spurious quotes. This is yet another
reason why portable shell code should not contain
"`...\"...\"...`"
constructs (see Shell Substitutions).
ZSH_VERSION
is set. By default zsh is not
compatible with the Bourne shell: you must execute ‘emulate sh’,
and for zsh versions before 3.1.6-dev-18 you must also
set NULLCMD
to ‘:’. See Compatibility, for details.
The default Mac OS X sh was originally Zsh; it was changed to Bash in Mac OS X 10.2.
The following discussion between Russ Allbery and Robert Lipe is worth reading:
Russ Allbery:
The GNU assumption that /bin/sh is the one and only shell leads to a permanent deadlock. Vendors don't want to break users' existing shell scripts, and there are some corner cases in the Bourne shell that are not completely compatible with a Posix shell. Thus, vendors who have taken this route will never (OK...“never say never”) replace the Bourne shell (as /bin/sh) with a Posix shell.
Robert Lipe:
This is exactly the problem. While most (at least most System V's) do have a Bourne shell that accepts shell functions most vendor /bin/sh programs are not the Posix shell.So while most modern systems do have a shell somewhere that meets the Posix standard, the challenge is to find it.
Don't rely on ‘\’ being preserved just because it has no special meaning together with the next symbol. In the native sh on OpenBSD 2.7 ‘\"’ expands to ‘"’ in here-documents with unquoted delimiter. As a general rule, if ‘\\’ expands to ‘\’ use ‘\\’ to get ‘\’.
With OpenBSD 2.7's sh
$ cat <<EOF > \" \\ > EOF " \
and with Bash:
bash-2.04$ cat <<EOF > \" \\ > EOF \" \
Some shells mishandle large here-documents: for example,
Solaris 10 dtksh and the UnixWare 7.1.1 Posix shell, which are
derived from Korn shell version M-12/28/93d, mishandle braced variable
expansion that crosses a 1024- or 4096-byte buffer boundary
within a here-document. Only the part of the variable name after the boundary
is used. For example, ${variable}
could be replaced by the expansion
of ${ble}
. If the end of the variable name is aligned with the block
boundary, the shell reports an error, as if you used ${}
.
Instead of ${variable-default}
, the shell may expand
${riable-default}
, or even ${fault}
. This bug can often
be worked around by omitting the braces: $variable
. The bug was fixed in
‘ksh93g’ (1998-04-30) but as of 2006 many operating systems were
still shipping older versions with the bug.
Many older shells (including the Bourne shell) implement here-documents inefficiently. In particular, some shells can be extremely inefficient when a single statement contains many here-documents. For instance if your configure.ac includes something like:
if <cross_compiling>; then assume this and that else check this check that check something else ... on and on forever ... fi
A shell parses the whole if
/fi
construct, creating
temporary files for each here-document in it. Some shells create links
for such here-documents on every fork
, so that the clean-up code
they had installed correctly removes them. It is creating the links
that can take the shell forever.
Moving the tests out of the if
/fi
, or creating multiple
if
/fi
constructs, would improve the performance
significantly. Anyway, this kind of construct is not exactly the
typical use of Autoconf. In fact, it's even not recommended, because M4
macros can't look into shell conditionals, so we may fail to expand a
macro when it was expanded before in a conditional path, and the
condition turned out to be false at runtime, and we end up not
executing the macro at all.
Don't redirect the same file descriptor several times, as you are doomed to failure under Ultrix.
ULTRIX V4.4 (Rev. 69) System #31: Thu Aug 10 19:42:23 GMT 1995 UWS V4.4 (Rev. 11) $ eval 'echo matter >fullness' >void illegal io $ eval '(echo matter >fullness)' >void illegal io $ (eval '(echo matter >fullness)') >void Ambiguous output redirect.
In each case the expected result is of course fullness containing ‘matter’ and void being empty.
Don't try to redirect the standard error of a command substitution: it must be done inside the command substitution: when running ‘: `cd /zorglub` 2>/dev/null’ expect the error message to escape, while ‘: `cd /zorglub 2>/dev/null`’ works properly.
It is worth noting that Zsh (but not Ash nor Bash) makes it possible in assignments though: ‘foo=`cd /zorglub` 2>/dev/null’.
Most shells, if not all (including Bash, Zsh, Ash), output traces on stderr, even for subshells. This might result in undesirable content if you meant to capture the standard-error output of the inner command:
$ ash -x -c '(eval "echo foo >&2") 2>stderr' $ cat stderr + eval echo foo >&2 + echo foo foo $ bash -x -c '(eval "echo foo >&2") 2>stderr' $ cat stderr + eval 'echo foo >&2' ++ echo foo foo $ zsh -x -c '(eval "echo foo >&2") 2>stderr' # Traces on startup files deleted here. $ cat stderr +zsh:1> eval echo foo >&2 +zsh:1> echo foo foo
You'll appreciate the various levels of detail...
One workaround is to grep out uninteresting lines, hoping not to remove good ones...
Don't try to move/delete open files, such as in ‘exec >foo; mv foo bar’; see Limitations of Builtins, mv for more details.
Don't rely on file descriptors 0, 1, and 2 remaining closed in a subsidiary program. If any of these descriptors is closed, the operating system may open an unspecified file for the descriptor in the new process image. Posix says this may be done only if the subsidiary program is set-user-ID or set-group-ID, but HP-UX 11.23 does it even for ordinary programs.
Don't rely on open file descriptors being open in child processes. In ksh, file descriptors above 2 which are opened using ‘exec n>file’ are closed by a subsequent ‘exec’ (such as that involved in the fork-and-exec which runs a program or script). Thus, using sh, we have:
$ cat ./descrips #!/bin/sh - echo hello >&5 $ exec 5>t $ ./descrips $ cat t $ hello
But using ksh:
$ exec 5>t $ ./descrips hello $ cat t $
Within the process which runs the ‘descrips’ script, file descriptor 5 is closed.
A few ancient systems reserved some file descriptors. By convention, file descriptor 3 was opened to /dev/tty when you logged into Eighth Edition (1985) through Tenth Edition Unix (1989). File descriptor 4 had a special use on the Stardent/Kubota Titan (circa 1990), though we don't now remember what it was. Both these systems are obsolete, so it's now safe to treat file descriptors 3 and 4 like any other file descriptors.
Autoconf uses shell-script processing extensively, so the file names that it processes should not contain characters that are special to the shell. Special characters include space, tab, newline, nul, and the following:
" # $ & ' ( ) * ; < = > ? [ \ ` |
Also, file names should not begin with ‘~’ or ‘-’, and should contain neither ‘-’ immediately after ‘/’ nor ‘~’ immediately after ‘:’. On Posix-like platforms, directory names should not contain ‘:’, as this runs afoul of ‘:’ used as the path separator.
These restrictions apply not only to the files that you distribute, but also to the absolute file names of your source, build, and destination directories.
On some Posix-like platforms, ‘!’ and ‘^’ are special too, so they should be avoided.
Posix lets implementations treat leading // specially, but requires leading /// and beyond to be equivalent to /. Most Unix variants treat // like /. However, some treat // as a “super-root” that can provide access to files that are not otherwise reachable from /. The super-root tradition began with Apollo Domain/OS, which died out long ago, but unfortunately Cygwin has revived it.
While autoconf and friends are usually run on some Posix variety, they can be used on other systems, most notably DOS variants. This impacts several assumptions regarding file names.
For example, the following code:
case $foo_dir in /*) # Absolute ;; *) foo_dir=$dots$foo_dir ;; esac
fails to properly detect absolute file names on those systems, because they can use a drivespec, and usually use a backslash as directory separator. If you want to be portable to DOS variants (at the price of rejecting valid but oddball Posix file names like a:\b), you can check for absolute file names like this:
case $foo_dir in [\\/]* | ?:[\\/]* ) # Absolute ;; *) foo_dir=$dots$foo_dir ;; esac
Make sure you quote the brackets if appropriate and keep the backslash as first character (see Limitations of Builtins).
Also, because the colon is used as part of a drivespec, these systems don't
use it as path separator. When creating or accessing paths, you can use the
PATH_SEPARATOR
output variable instead. configure sets this
to the appropriate value (‘:’ or ‘;’) when it starts up.
File names need extra care as well. While DOS variants that are Posixy enough to run autoconf (such as DJGPP) are usually able to handle long file names properly, there are still limitations that can seriously break packages. Several of these issues can be easily detected by the doschk package.
A short overview follows; problems are marked with sfn/lfn to indicate where they apply: sfn means the issues are only relevant to plain DOS, not to DOS under Microsoft Windows variants, while lfn identifies problems that exist even under Microsoft Windows variants.
This is perfectly OK on Posix variants:
AC_CONFIG_HEADERS([config.h]) AC_CONFIG_FILES([source.c foo.bar]) AC_OUTPUT
but it causes problems on DOS, as it requires ‘config.h.in’, ‘source.c.in’ and ‘foo.bar.in’. To make your package more portable to DOS-based environments, you should use this instead:
AC_CONFIG_HEADERS([config.h:config.hin]) AC_CONFIG_FILES([source.c:source.cin foo.bar:foobar.in]) AC_OUTPUT
The 8+3 limit is not usually a problem under Microsoft Windows, as it
uses numeric
tails in the short version of file names to make them unique. However, a
registry setting can turn this behavior off. While this makes it
possible to share file trees containing long file names between sfn
and lfn environments, it also means the above problem applies there
as well.
Contrary to a persistent urban legend, the Bourne shell does not
systematically split variables and back-quoted expressions, in particular
on the right-hand side of assignments and in the argument of case
.
For instance, the following code:
case "$given_srcdir" in .) top_srcdir="`echo "$dots" | sed 's,/$,,'`" ;; *) top_srcdir="$dots$given_srcdir" ;; esac
is more readable when written as:
case $given_srcdir in .) top_srcdir=`echo "$dots" | sed 's,/$,,'` ;; *) top_srcdir=$dots$given_srcdir ;; esac
and in fact it is even more portable: in the first case of the
first attempt, the computation of top_srcdir
is not portable,
since not all shells properly understand "`..."..."...`"
.
Worse yet, not all shells understand "`...\"...\"...`"
the same way. There is just no portable way to use double-quoted
strings inside double-quoted back-quoted expressions (pfew!).
$@
The traditional way to work around this portability problem is to use ‘${1+"$@"}’. Unfortunately this method does not work with Zsh (3.x and 4.x), which is used on Mac OS X. When emulating the Bourne shell, Zsh performs word splitting on ‘${1+"$@"}’:
zsh $ emulate sh zsh $ for i in "$@"; do echo $i; done Hello World ! zsh $ for i in ${1+"$@"}; do echo $i; done Hello World !
Zsh handles plain ‘"$@"’ properly, but we can't use plain ‘"$@"’ because of the portability problems mentioned above. One workaround relies on Zsh's “global aliases” to convert ‘${1+"$@"}’ into ‘"$@"’ by itself:
test "${ZSH_VERSION+set}" = set && alias -g '${1+"$@"}'='"$@"'
A more conservative workaround is to avoid ‘"$@"’ if it is possible that there may be no positional arguments. For example, instead of:
cat conftest.c "$@"
you can use this instead:
case $# in 0) cat conftest.c;; *) cat conftest.c "$@";; esac
Autoconf macros often use the set command to update
‘$@’, so if you are writing shell code intended for
configure you should not assume that the value of ‘$@’
persists for any length of time.
${10}
shift
. The 7th Edition shell reported an error if given
${10}
, and
Solaris 10 /bin/sh still acts that way:
$ set 1 2 3 4 5 6 7 8 9 10 $ echo ${10} bad substitution
${
var:-
value}
sh
, don't accept the
colon for any shell substitution, and complain and die.
${
var=
literal}
: ${var='Some words'}
otherwise some shells, such as on Digital Unix V 5.0, die because of a “bad substitution”.
Solaris /bin/sh has a frightening bug in its interpretation of this. Imagine you need set a variable to a string containing ‘}’. This ‘}’ character confuses Solaris /bin/sh when the affected variable was already set. This bug can be exercised by running:
$ unset foo $ foo=${foo='}'} $ echo $foo } $ foo=${foo='}' # no error; this hints to what the bug is $ echo $foo } $ foo=${foo='}'} $ echo $foo }} ^ ugh!
It seems that ‘}’ is interpreted as matching ‘${’, even
though it is enclosed in single quotes. The problem doesn't happen
using double quotes.
${
var=
expanded-value}
default="yu,yaa" : ${var="$default"}
sets var to ‘M-yM-uM-,M-yM-aM-a’, i.e., the 8th bit of each char is set. You don't observe the phenomenon using a simple ‘echo $var’ since apparently the shell resets the 8th bit when it expands $var. Here are two means to make this shell confess its sins:
$ cat -v <<EOF $var EOF
and
$ set | grep '^var=' | cat -v
One classic incarnation of this bug is:
default="a b c" : ${list="$default"} for c in $list; do echo $c done
You'll get ‘a b c’ on a single line. Why? Because there are no spaces in ‘$list’: there are ‘M- ’, i.e., spaces with the 8th bit set, hence no IFS splitting is performed!!!
One piece of good news is that Ultrix works fine with ‘: ${list=$default}’; i.e., if you don't quote. The bad news is then that QNX 4.25 then sets list to the last item of default!
The portable way out consists in using a double assignment, to switch the 8th bit twice on Ultrix:
list=${list="$default"}
...but beware of the ‘}’ bug from Solaris (see above). For safety, use:
test "${var+set}" = set || var={value}
`
commands`
While in general it makes no sense, do not substitute a single builtin with side effects, because Ash 0.2, trying to optimize, does not fork a subshell to perform the command.
For instance, if you wanted to check that cd is silent, do not use ‘test -z "`cd /`"’ because the following can happen:
$ pwd /tmp $ test -z "`cd /`" && pwd /
The result of ‘foo=`exit 1`’ is left as an exercise to the reader.
The MSYS shell leaves a stray byte in the expansion of a double-quoted command substitution of a native program, if the end of the substution is not aligned with the end of the double quote. This may be worked around by inserting another pair of quotes:
$ echo "`printf 'foo\r\n'` bar" > broken $ echo "`printf 'foo\r\n'`"" bar" | cmp - broken - broken differ: char 4, line 1
$(
commands)
`
commands`
.
This construct can be nested while this is impossible to do portably with back quotes. Unfortunately it is not yet universally supported. Most notably, even recent releases of Solaris don't support it:
$ showrev -c /bin/sh | grep version Command version: SunOS 5.10 Generic 121004-01 Oct 2005 $ echo $(echo blah) syntax error: `(' unexpected
nor does irix 6.5's Bourne shell:
$ uname -a IRIX firebird-image 6.5 07151432 IP22 $ echo $(echo blah) $(echo blah)
If you do use ‘$(commands)’, make sure that the commands do not start with a parenthesis, as that would cause confusion with a different notation ‘$((expression))’ that in modern shells is an arithmetic expression not a command. To avoid the confusion, insert a space between the two opening parentheses.
Avoid commands that contain unbalanced parentheses in here-documents, comments, or case statement patterns, as many shells mishandle them. For example, Bash 3.1, ‘ksh88’, pdksh 5.2.14, and Zsh 4.2.6 all mishandle the following valid command:
echo $(case x in x) echo hello;; esac)
^
When setting several variables in a row, be aware that the order of the evaluation is undefined. For instance ‘foo=1 foo=2; echo $foo’ gives ‘1’ with Solaris /bin/sh, but ‘2’ with Bash. You must use ‘;’ to enforce the order: ‘foo=1; foo=2; echo $foo’.
Don't rely on the following to find subdir/program:
PATH=subdir$PATH_SEPARATOR$PATH program
as this does not work with Zsh 3.0.6. Use something like this instead:
(PATH=subdir$PATH_SEPARATOR$PATH; export PATH; exec program)
Don't rely on the exit status of an assignment: Ash 0.2 does not change the status and propagates that of the last statement:
$ false || foo=bar; echo $? 1 $ false || foo=`:`; echo $? 0
and to make things even worse, QNX 4.25 just sets the exit status to 0 in any case:
$ foo=`exit 1`; echo $? 0
To assign default values, follow this algorithm:
: ${var='my literal'}
: ${var="$default"}
var=${var="$default"}
test "${var+set}" = set || var="has a '}'"
In most cases ‘var=${var="$default"}’ is fine, but in case of doubt, just use the last form. See Shell Substitutions, items ‘${var:-value}’ and ‘${var=value}’ for the rationale.
Beware of two opening parentheses in a row, as some shell implementations mishandle them. For example, ‘pdksh’ 5.2.14 misparses the following code:
if ((true) || false); then echo ok fi
To work around this problem, insert a space between the two opening parentheses. There is a similar problem and workaround with ‘$((’; see Shell Substitutions.
Posix requires support for case
patterns with opening
parentheses like this:
case $file_name in (*.c) echo "C source code";; esac
but the (
in this example is not portable to many older Bourne
shell implementations. It can be omitted safely.
Unpatched Tru64 5.1 sh omits the last slash of command-line arguments that contain two trailing slashes:
$ echo / // /// //// .// //. / / // /// ./ //. $ x=// $ eval "echo \$x" / $ set -x $ echo abc | tr -t ab // + echo abc + tr -t ab / /bc
However, our understanding is that patches are available, so perhaps it's not worth worrying about working around this horrendous bug.
Some shell variables should not be used, since they can have a deep influence on the behavior of the shell. In order to recover a sane behavior from the shell, some variables should be unset, but unset is not portable (see Limitations of Builtins) and a fallback value is needed.
As a general rule, shell variable names containing a lower-case letter
are safe; you can define and use these variables without worrying about
their effect on the underlying system, and without worrying about
whether the shell changes them unexpectedly. (The exception is the
shell variable status
, as described below.)
Here is a list of names that are known to cause trouble. This list is
not exhaustive, but you should be safe if you avoid the name
status
and names containing only upper-case letters and
underscores.
_
BIN_SH
xpg4
, subsidiary invocations of
the standard shell conform to Posix.
Autoconf-generated scripts export this variable when they start up.
CDPATH
cd
with a relative file name that did not start
with ‘./’ or ‘../’. Posix
1003.1-2001 says that if a nonempty directory name from CDPATH
is used successfully, cd
prints the resulting absolute
file name. Unfortunately this output can break idioms like
‘abs=`cd src && pwd`’ because abs
receives the name twice.
Also, many shells do not conform to this part of Posix; for
example, zsh prints the result only if a directory name
other than . was chosen from CDPATH.
In practice the shells that have this problem also support unset, so you can work around the problem as follows:
(unset CDPATH) >/dev/null 2>&1 && unset CDPATH
You can also avoid output by ensuring that your directory name is absolute or anchored at ‘./’, as in ‘abs=`cd ./src && pwd`’.
Autoconf-generated scripts automatically unset CDPATH if
possible, so you need not worry about this problem in those scripts.
DUALCASE
ENV
MAIL
MAILPATH
PS1
PS2
PS4
(unset ENV) >/dev/null 2>&1 && unset ENV MAIL MAILPATH PS1='$ ' PS2='> ' PS4='+ '
IFS
Don't set the first character of IFS
to backslash. Indeed,
Bourne shells use the first character (backslash) when joining the
components in ‘"$@"’ and some shells then reinterpret (!) the
backslash escapes, so you can end up with backspace and other strange
characters.
The proper value for IFS
(in regular code, not when performing
splits) is ‘<SPC><TAB><RET>’. The first character is
especially important, as it is used to join the arguments in ‘$*’;
however, note that traditional shells, but also bash-2.04, fail to adhere
to this and join with a space anyway.
LANG
LC_ALL
LC_COLLATE
LC_CTYPE
LC_MESSAGES
LC_MONETARY
LC_NUMERIC
LC_TIME
LANGUAGE
LC_ADDRESS
LC_IDENTIFICATION
LC_MEASUREMENT
LC_NAME
LC_PAPER
LC_TELEPHONE
LINENO
LINENO
.
Its value is the line number of the beginning of the current command.
Autoconf attempts to execute configure with a shell that
supports LINENO
.
If no such shell is available, it attempts to implement LINENO
with a Sed prepass that replaces each instance of the string
$LINENO
(not followed by an alphanumeric character) with the
line's number.
You should not rely on LINENO
within eval, as the
behavior differs in practice. Also, the possibility of the Sed
prepass means that you should not rely on $LINENO
when quoted,
when in here-documents, or when in long commands that cross line
boundaries. Subshells should be OK, though. In the following
example, lines 1, 6, and 9 are portable, but the other instances of
LINENO
are not:
$ cat lineno echo 1. $LINENO cat <<EOF 3. $LINENO 4. $LINENO EOF ( echo 6. $LINENO ) eval 'echo 7. $LINENO' echo 8. '$LINENO' echo 9. $LINENO ' 10.' $LINENO $ bash-2.05 lineno 1. 1 3. 2 4. 2 6. 6 7. 1 8. $LINENO 9. 9 10. 9 $ zsh-3.0.6 lineno 1. 1 3. 2 4. 2 6. 6 7. 7 8. $LINENO 9. 9 10. 9 $ pdksh-5.2.14 lineno 1. 1 3. 2 4. 2 6. 6 7. 0 8. $LINENO 9. 9 10. 9 $ sed '=' <lineno | > sed ' > N > s,$,-, > t loop > :loop > s,^\([0-9]*\)\(.*\)[$]LINENO\([^a-zA-Z0-9_]\),\1\2\1\3, > t loop > s,-$,, > s,^[0-9]*\n,, > ' | > sh 1. 1 3. 3 4. 4 6. 6 7. 7 8. 8 9. 9 10. 10
NULLCMD
PATH_SEPARATOR
PATH_SEPARATOR
.
PWD
RANDOM
RANDOM
, a variable that returns a different
integer each time it is used. Most of the time, its value does not
change when it is not used, but on irix 6.5 the value changes all
the time. This can be observed by using set. It is common
practice to use $RANDOM
as part of a file name, but code
shouldn't rely on $RANDOM
expanding to a nonempty string.
status
zsh
(at least 3.1.6),
hence read-only. Do not use it.
No, no, we are serious: some shells do have limitations! :)
You should always keep in mind that any builtin or command may support
options, and therefore differ in behavior with arguments
starting with a dash. For instance, the innocent ‘echo "$word"’
can give unexpected results when word
starts with a dash. It is
often possible to avoid this problem using ‘echo "x$word"’, taking
the ‘x’ into account later in the pipe.
if ! cmp file1 file2 >/dev/null 2>&1; then echo files differ or trouble fi
is therefore not portable in practice. Typically it is easy to rewrite such code, e.g.:
cmp file1 file2 >/dev/null 2>&1 || echo files differ or trouble
More generally, one can always rewrite ‘! command’ as:
if command; then (exit 1); else :; fi
You don't need the final ‘;;’, but you should use it.
Because of a bug in its fnmatch
, Bash fails to properly
handle backslashes in character classes:
bash-2.02$ case /tmp in [/\\]*) echo OK;; esac bash-2.02$
This is extremely unfortunate, since you are likely to use this code to handle Posix or ms-dos absolute file names. To work around this bug, always put the backslash first:
bash-2.02$ case '\TMP' in [\\/]*) echo OK;; esac OK bash-2.02$ case /tmp in [\\/]*) echo OK;; esac OK
Many Bourne shells cannot handle closing brackets in character classes correctly.
Some shells also have problems with backslash escaping in case you do not want to match the backslash: both a backslash and the escaped character match this pattern. To work around this, specify the character class in a variable, so that quote removal does not apply afterwards, and the special characters don't have to be backslash-escaped:
$ case '\' in [\<]) echo OK;; esac OK $ scanset='[<]'; case '\' in $scanset) echo OK;; esac $
Even with this, Solaris ksh matches a backslash if the set contains any of the characters ‘|’, ‘&’, ‘(’, or ‘)’.
Conversely, Tru64 ksh (circa 2003) erroneously always matches a closing parenthesis if not specified in a character class:
$ case foo in *\)*) echo fail ;; esac fail $ case foo in *')'*) echo fail ;; esac fail
Some shells, such as Ash 0.3.8, are confused by an empty
case
/esac
:
ash-0.3.8 $ case foo in esac; error-->Syntax error: ";" unexpected (expecting ")")
Many shells still do not support parenthesized cases, which is a pity for those of us using tools that rely on balanced parentheses. For instance, Solaris /bin/sh:
$ case foo in (foo) echo foo;; esac error-->syntax error: `(' unexpected
Portable scripts should assume neither option is supported, and should
assume neither behavior is the default. This can be a bit tricky,
since the Posix default behavior means that, for example,
‘ls ..’ and ‘cd ..’ may refer to different directories if
the current logical directory is a symbolic link. It is safe to use
cd dir if dir contains no .. components.
Also, Autoconf-generated scripts check for this problem when computing
variables like ac_top_srcdir
(see Configuration Actions),
so it is safe to cd to these variables.
See See Special Shell Variables, for portability problems involving
cd and the CDPATH environment variable.
Also please see the discussion of the pwd command.
Don't expect any option. See Preset Output Variables, ECHO_N
etc. for a means to simulate -n.
Do not use backslashes in the arguments, as there is no consensus on their handling. For ‘echo '\n' | wc -l’, the sh of Solaris outputs 2, but Bash and Zsh (in sh emulation mode) output 1. The problem is truly echo: all the shells understand ‘'\n'’ as the string composed of a backslash and an ‘n’.
Because of these problems, do not pass a string containing arbitrary characters to echo. For example, ‘echo "$foo"’ is safe if you know that foo's value cannot contain backslashes and cannot start with ‘-’, but otherwise you should use a here-document like this:
cat <<EOF $foo EOF
It is obviously unwise to use ‘eval $cmd’ if the string value of ‘cmd’ was derived from an untrustworthy source. But even if the string value is valid, ‘eval $cmd’ might not work as intended, since it causes field splitting and file name expansion to occur twice, once for the eval and once for the command itself. It is therefore safer to use ‘eval "$cmd"’. For example, if cmd has the value ‘cat test?.c’, ‘eval $cmd’ might expand to the equivalent of ‘cat test;.c’ if there happens to be a file named test;.c in the current directory; and this in turn mistakenly attempts to invoke cat on the file test and then execute the command .c. To avoid this problem, use ‘eval "$cmd"’ rather than ‘eval $cmd’.
However, suppose that you want to output the text of the evaluated command just before executing it. Assuming the previous example, ‘echo "Executing: $cmd"’ outputs ‘Executing: cat test?.c’, but this output doesn't show the user that ‘test;.c’ is the actual name of the copied file. Conversely, ‘eval "echo Executing: $cmd"’ works on this example, but it fails with ‘cmd='cat foo >bar'’, since it mistakenly replaces the contents of bar by the string ‘cat foo’. No simple, general, and portable solution to this problem is known.
You should also be wary of common bugs in eval implementations. In some shell implementations (e.g., older ash, OpenBSD 3.8 sh, pdksh v5.2.14 99/07/13.2, and zsh 4.2.5), the arguments of ‘eval’ are evaluated in a context where ‘$?’ is 0, so they exhibit behavior like this:
$ false; eval 'echo $?' 0
The correct behavior here is to output a nonzero value, but portable scripts should not rely on this.
You should not rely on LINENO
within eval.
See Special Shell Variables.
$?
;
unfortunately, some shells, such as the DJGPP port of Bash 2.04, just
perform ‘exit 0’.
bash-2.04$ foo=`exit 1` || echo fail fail bash-2.04$ foo=`(exit 1)` || echo fail fail bash-2.04$ foo=`(exit 1); exit` || echo fail bash-2.04$
Using ‘exit $?’ restores the expected behavior.
Some shell scripts, such as those generated by autoconf, use a trap to clean up before exiting. If the last shell command exited with nonzero status, the trap also exits with nonzero status so that the invoker can tell that an error occurred.
Unfortunately, in some shells, such as Solaris /bin/sh, an exit
trap ignores the exit
command's argument. In these shells, a trap
cannot determine whether it was invoked by plain exit
or by
exit 1
. Instead of calling exit
directly, use the
AC_MSG_ERROR
macro that has a workaround for this problem.
Alas, many shells, such as Solaris /bin/sh, irix 6.3, irix 5.2, AIX 4.1.5, and Digital Unix 4.0, forget to export the environment variables they receive. As a result, two variables coexist: the environment variable and the shell variable. The following code demonstrates this failure:
#!/bin/sh echo $FOO FOO=bar echo $FOO exec /bin/sh $0
when run with ‘FOO=foo’ in the environment, these shells print alternately ‘foo’ and ‘bar’, although they should print only ‘foo’ and then a sequence of ‘bar’s.
Therefore you should export again each environment variable
that you update.
for arg do echo "$arg" done
You may not leave the do
on the same line as for
,
since some shells improperly grok:
for arg; do echo "$arg" done
If you want to explicitly refer to the positional arguments, given the ‘$@’ bug (see Shell Substitutions), use:
for arg in ${1+"$@"}; do echo "$arg" done
But keep in mind that Zsh, even in Bourne shell emulation mode, performs
word splitting on ‘${1+"$@"}’; see Shell Substitutions,
item ‘$@’, for more.
if ! cmp -s file file.new; then mv file.new file fi
use:
if cmp -s file file.new; then :; else mv file.new file fi
There are shells that do not reset the exit status from an if:
$ if (exit 42); then true; fi; echo $? 42
whereas a proper shell should have printed ‘0’. This is especially bad in makefiles since it produces false failures. This is why properly written makefiles, such as Automake's, have such hairy constructs:
if test -f "$file"; then install "$file" "$dest" else : fi
printf %s -foo
Posix 1003.1-2001 requires that pwd must support the -L (“logical”) and -P (“physical”) options, with -L being the default. However, traditional shells do not support these options, and their pwd command has the -P behavior.
Portable scripts should assume neither option is supported, and should assume neither behavior is the default. Also, on many hosts ‘/bin/pwd’ is equivalent to ‘pwd -P’, but Posix does not require this behavior and portable scripts should not rely on it.
Typically it's best to use plain pwd. On modern hosts this outputs logical directory names, which have the following advantages:
Also please see the discussion of the cd command.
The set builtin faces the usual problem with arguments starting with a dash. Modern shells such as Bash or Zsh understand -- to specify the end of the options (any argument after -- is a parameter, even ‘-x’ for instance), but many traditional shells (e.g., Solaris 10 /bin/sh) simply stop option processing as soon as a non-option argument is found. Therefore, use ‘dummy’ or simply ‘x’ to end the option processing, and use shift to pop it out:
set x $my_list; shift
Avoid ‘set -’, e.g., ‘set - $my_list’. Posix no longer requires support for this command, and in traditional shells ‘set - $my_list’ resets the -v and -x options, which makes scripts harder to debug.
Some nonstandard shells do not recognize more than one option (e.g., ‘set -e -x’ assigns ‘-x’ to the command line). It is better to combine them:
set -ex
The BSD shell has had several problems with the -e option, partly because BSD make traditionally used -e even though this was incompatible with Posix (see Failure in Make Rules). Older versions of the BSD shell (circa 1990) mishandled ‘&&’, ‘||’, ‘if’, and ‘case’ when -e was in effect, causing the shell to exit unexpectedly in some cases. This was particularly a problem with makefiles, and led to circumlocutions like ‘sh -c 'test -f file || touch file'’, where the seemingly-unnecessary ‘sh -c '...'’ wrapper works around the bug.
Even relatively-recent versions of the BSD shell (e.g., OpenBSD 3.4) wrongly exit with -e if a command within ‘&&’ fails inside a compound statement. For example:
#! /bin/sh set -e foo='' test -n "$foo" && exit 1 echo one if :; then test -n "$foo" && exit 1 fi echo two
does not print ‘two’. One workaround is to use ‘if test -n
"$foo"; then exit 1; fi’ rather than ‘test -n "$foo" && exit 1’.
Another possibility is to warn BSD users not to use ‘sh -e’.
Don't use ‘shift 2’ etc.; it was not in the 7th Edition Bourne shell,
and it is also absent in many pre-Posix shells.
test
program is the way to perform many file and string
tests. It is often invoked by the alternate name ‘[’, but using
that name in Autoconf code is asking for trouble since it is an M4 quote
character.
If you need to make multiple checks using test
, combine them with
the shell operators ‘&&’ and ‘||’ instead of using the
test
operators -a and -o. On System V, the
precedence of -a and -o is wrong relative to the unary
operators; consequently, Posix does not specify them, so using them
is nonportable. If you combine ‘&&’ and ‘||’ in the same
statement, keep in mind that they have equal precedence.
It is safe to use ‘!’ as a test operator. For example,
‘if test ! -d foo; ...’ is portable even though ‘if ! test
-d foo; ...’ is not.
/bin/sh
support only -h.
test
might interpret its argument as an
option (e.g., ‘string = "-n"’).
Contrary to a common belief, ‘test -n string’ and ‘test -z string’ are portable. Nevertheless many shells (such as Solaris, AIX 3.2, unicos 10.0.0.6, Digital Unix 4, etc.) have bizarre precedence and may be confused if string looks like an operator:
$ test -n = test: argument expected
If there are risks, use ‘test "xstring" = x’ or ‘test "xstring" != x’ instead.
It is common to find variations of the following idiom:
test -n "`echo $ac_feature | sed 's/[-a-zA-Z0-9_]//g'`" && action
to take an action when a token matches a given pattern. Such constructs should always be avoided by using:
echo "$ac_feature" | grep '[^-a-zA-Z0-9_]' >/dev/null 2>&1 && action
Use case
where possible since it is faster, being a shell builtin:
case $ac_feature in *[!-a-zA-Z0-9_]*) action;; esac
Alas, negated character classes are probably not portable, although no shell is known to not support the Posix syntax ‘[!...]’ (when in interactive mode, zsh is confused by the ‘[!...]’ syntax and looks for an event in its history because of ‘!’). Many shells do not support the alternative syntax ‘[^...]’ (Solaris, Digital Unix, etc.).
One solution can be:
expr "$ac_feature" : '.*[^-a-zA-Z0-9_]' >/dev/null && action
or better yet
expr "X$ac_feature" : '.*[^-a-zA-Z0-9_]' >/dev/null && action
‘expr "Xfoo" : "Xbar"’ is more robust than ‘echo
"Xfoo" | grep "^Xbar"’, because it avoids problems when
‘foo’ contains backslashes.
Posix says that ‘trap - 1 2 13 15’ resets the traps for the specified signals to their default values, but many common shells (e.g., Solaris /bin/sh) misinterpret this and attempt to execute a “command” named - when the specified conditions arise. There is no portable workaround, except for ‘trap - 0’, for which ‘trap '' 0’ is a portable substitute.
Although Posix is not absolutely clear on this point, it is widely admitted that when entering the trap ‘$?’ should be set to the exit status of the last command run before the trap. The ambiguity can be summarized as: “when the trap is launched by an exit, what is the last command run: that before exit, or exit itself?”
Bash considers exit to be the last command, while Zsh and Solaris /bin/sh consider that when the trap is run it is still in the exit, hence it is the previous exit status that the trap receives:
$ cat trap.sh trap 'echo $?' 0 (exit 42); exit 0 $ zsh trap.sh 42 $ bash trap.sh 0
The portable solution is then simple: when you want to ‘exit 42’, run ‘(exit 42); exit 42’, the first exit being used to set the exit status to 42 for Zsh, and the second to trigger the trap and pass 42 as exit status for Bash.
The shell in FreeBSD 4.0 has the following bug: ‘$?’ is reset to 0 by empty lines if the code is inside trap.
$ trap 'false echo $?' 0 $ exit 0
Fortunately, this bug only affects trap.
In a sense, yes, because if it doesn't exist, the shell will produce an exit status of failure, which is correct for false, but not for true.
PS1
, you can test for its existence and use
it provided you give a neutralizing value when unset is
not supported:
if (unset FOO) >/dev/null 2>&1; then unset=unset else unset=false fi $unset PS1 || PS1='$ '
See Special Shell Variables, for some neutralizing values. Also, see Limitations of Builtins, documentation of export, for the case of environment variables.
The small set of tools you can expect to find on any machine can still include some limitations you should be aware of.
$ gawk 'function die () { print "Aaaaarg!" } BEGIN { die () }' gawk: cmd. line:2: BEGIN { die () } gawk: cmd. line:2: ^ parse error $ gawk 'function die () { print "Aaaaarg!" } BEGIN { die() }' Aaaaarg!
If you want your program to be deterministic, don't depend on for
on arrays:
$ cat for.awk END { arr["foo"] = 1 arr["bar"] = 1 for (i in arr) print i } $ gawk -f for.awk </dev/null foo bar $ nawk -f for.awk </dev/null bar foo
Some Awk implementations, such as HP-UX 11.0's native one, mishandle anchors:
$ echo xfoo | $AWK '/foo|^bar/ { print }' $ echo bar | $AWK '/foo|^bar/ { print }' bar $ echo xfoo | $AWK '/^bar|foo/ { print }' xfoo $ echo bar | $AWK '/^bar|foo/ { print }' bar
Either do not depend on such patterns (i.e., use ‘/^(.*foo|bar)/’, or use a simple test to reject such implementations.
AIX version 5.2 has an arbitrary limit of 399 on the the length of regular expressions and literal strings in an Awk program.
Traditional Awk implementations derived from Unix version 7, such as
Solaris /bin/awk, have many limitations and do not
conform to Posix. Nowadays AC_PROG_AWK
(see Particular Programs) finds you an Awk that doesn't have these problems, but if
for some reason you prefer not to use AC_PROG_AWK
you may need to
address them.
Traditional Awk does not support multidimensional arrays or user-defined functions.
Traditional Awk does not support the -v option. You can use
assignments after the program instead, e.g., $AWK '{print v
$1}' v=x; however, don't forget that such assignments are not
evaluated until they are encountered (e.g., after any BEGIN
action).
Traditional Awk does not support the keywords delete
or do
.
Traditional Awk does not support the expressions
a?
b:
c, !
a, a^
b,
or a^=
b.
Traditional Awk does not support the predefined CONVFMT
variable.
Traditional Awk supports only the predefined functions exp
,
int
, length
, log
, split
, sprintf
,
sqrt
, and substr
.
Traditional Awk getline
is not at all compatible with Posix;
avoid it.
Traditional Awk split
supports only two arguments.
Traditional Awk has a limit of 99
fields in a record. You may be able to circumvent this problem by using
split
.
AC_PROG_CC_C_O
.
When a compilation such as ‘cc -o foo foo.c’ fails, some compilers (such as cds on Reliant Unix) leave a foo.o.
HP-UX cc doesn't accept .S files to preprocess and assemble. ‘cc -c foo.S’ appears to succeed, but in fact does nothing.
The default executable, produced by ‘cc foo.c’, can be
The C compiler's traditional name is cc, but other names like
gcc are common. Posix 1003.1-2001 specifies the
name c99, but older Posix editions specified
c89 and anyway these standard names are rarely used in
practice. Typically the C compiler is invoked from makefiles that use
‘$(CC)’, so the value of the ‘CC’ make variable selects the
compiler name.
Some cp implementations (e.g., BSD/OS 4.2) do not allow trailing slashes at the end of nonexistent destination directories. To avoid this problem, omit the trailing slashes. For example, use ‘cp -R source /tmp/newdir’ rather than ‘cp -R source /tmp/newdir/’ if /tmp/newdir does not exist.
The ancient SunOS 4 cp does not support -f, although its mv does.
Traditionally, file timestamps had 1-second resolution, and ‘cp
-p’ copied the timestamps exactly. However, many modern file systems
have timestamps with 1-nanosecond resolution. Unfortunately, ‘cp
-p’ implementations truncate timestamps when copying files, so this
can result in the destination file appearing to be older than the
source. The exact amount of truncation depends on the resolution of
the system calls that cp uses; traditionally this was
utime
, which has 1-second resolution, but some newer
cp implementations use utimes
, which has
1-microsecond resolution. These newer implementations include GNU
Core Utilities 5.0.91 or later, and Solaris 8 (sparc) patch 109933-02 or
later. Unfortunately as of January 2006 there is still no system
call to set timestamps to the full nanosecond resolution.
Bob Proulx notes that ‘cp -p’ always tries to copy ownerships. But whether it actually does copy ownerships or not is a system dependent policy decision implemented by the kernel. If the kernel allows it then it happens. If the kernel does not allow it then it does not happen. It is not something cp itself has control over.
In Unix System V any user can chown files to any other user, and System V also has a non-sticky /tmp. That probably derives from the heritage of System V in a business environment without hostile users. BSD changed this to be a more secure model where only root can chown files and a sticky /tmp is used. That undoubtedly derives from the heritage of BSD in a campus environment.
GNU/Linux and Solaris by default follow BSD, but
can be configured to allow a System V style chown. On the
other hand, HP-UX follows System V, but can
be configured to use the modern security model and disallow
chown. Since it is an administrator-configurable parameter
you can't use the name of the kernel as an indicator of the behavior.
$ uname -a OSF1 medusa.sis.pasteur.fr V5.1 732 alpha $ date "+%s" %s
Some implementations, such as Tru64's, fail when comparing to
/dev/null. Use an empty file instead.
AS_DIRNAME
(see Programming in M4sh). For example:
dir=`dirname "$file"` # This is not portable. dir=`AS_DIRNAME(["$file"])` # This is more portable.
grep -E
. Also, some traditional implementations do
not work on long input lines. To work around these problems, invoke
AC_PROG_EGREP
and then use $EGREP
.
Portable extended regular expressions should use ‘\’ only to escape characters in the string ‘$()*+.?[\^{|’. For example, ‘\}’ is not portable, even though it typically matches ‘}’.
The empty alternative is not portable. Use ‘?’ instead. For instance with Digital Unix v5.0:
> printf "foo\n|foo\n" | $EGREP '^(|foo|bar)$' |foo > printf "bar\nbar|\n" | $EGREP '^(foo|bar|)$' bar| > printf "foo\nfoo|\n|bar\nbar\n" | $EGREP '^(foo||bar)$' foo |bar
$EGREP also suffers the limitations of grep.
Don't use length
, substr
, match
and index
.
expr '' \| ''
Posix 1003.2-1992 returns the empty string for this case, but traditional Unix returns ‘0’ (Solaris is one such example). In Posix 1003.1-2001, the specification was changed to match traditional Unix's behavior (which is bizarre, but it's too late to fix this). Please note that the same problem does arise when the empty string results from a computation, as in:
expr bar : foo \| foo : bar
Avoid this portability problem by avoiding the empty string.
Portable expr regular expressions should not begin with ‘^’. Patterns are automatically anchored so leading ‘^’ is not needed anyway.
The Posix standard is ambiguous as to whether ‘expr 'a' : '\(b\)'’ outputs ‘0’ or the empty string. In practice, it outputs the empty string on most platforms, but portable scripts should not assume this. For instance, the QNX 4.25 native expr returns ‘0’.
One might think that a way to get a uniform behavior would be to use the empty string as a default value:
expr a : '\(b\)' \| ''
Unfortunately this behaves exactly as the original expression; see the expr (‘|’) entry for more information.
Ancient expr implementations (e.g., SunOS 4 expr and Solaris 8 /usr/ucb/expr) have a silly length limit that causes expr to fail if the matched substring is longer than 120 bytes. In this case, you might want to fall back on ‘echo|sed’ if expr fails. Nowadays this is of practical importance only for the rare installer who mistakenly puts /usr/ucb before /usr/bin in PATH.
On Mac OS X 10.4, expr mishandles the pattern ‘[^-]’ in some cases. For example, the command
expr Xpowerpc-apple-darwin8.1.0 : 'X[^-]*-[^-]*-\(.*\)'
outputs ‘apple-darwin8.1.0’ rather than the correct ‘darwin8.1.0’. This particular case can be worked around by substituting ‘[^--]’ for ‘[^-]’.
Don't leave, there is some more!
The QNX 4.25 expr, in addition of preferring ‘0’ to the empty string, has a funny behavior in its exit status: it's always 1 when parentheses are used!
$ val=`expr 'a' : 'a'`; echo "$?: $val" 0: 1 $ val=`expr 'a' : 'b'`; echo "$?: $val" 1: 0 $ val=`expr 'a' : '\(a\)'`; echo "?: $val" 1: a $ val=`expr 'a' : '\(b\)'`; echo "?: $val" 1: 0
In practice this can be a big problem if you are ready to catch failures of expr programs with some other method (such as using sed), since you may get twice the result. For instance
$ expr 'a' : '\(a\)' || echo 'a' | sed 's/^\(a\)$/\1/'
outputs ‘a’ on most hosts, but ‘aa’ on QNX 4.25. A simple workaround consists of testing expr and using a variable set to expr or to false according to the result.
Tru64 expr incorrectly treats the result as a number, if it can be interpreted that way:
$ expr 00001 : '.*\(...\)' 1
grep -F
. Also, some traditional implementations do
not work on long input lines. To work around these problems, invoke
AC_PROG_FGREP
and then use $FGREP
.
The replacement of ‘{}’ is guaranteed only if the argument is exactly {}, not if it's only a part of an argument. For instance on DU, and HP-UX 10.20 and HP-UX 11:
$ touch foo $ find . -name foo -exec echo "{}-{}" \; {}-{}
while GNU find reports ‘./foo-./foo’.
Some of the options required by Posix are not portable in practice.
Don't use ‘grep -q’ to suppress output, because many grep
implementations (e.g., Solaris) do not support -q.
Don't use ‘grep -s’ to suppress output either, because Posix
says -s does not suppress output, only some error messages;
also, the -s option of traditional grep behaved
like -q does in most modern implementations. Instead,
redirect the standard output and standard error (in case the file
doesn't exist) of grep
to /dev/null. Check the exit
status of grep
to determine whether it found a match.
Some traditional grep implementations do not work on long
input lines. On AIX the default grep
silently truncates long
lines on the input before matching.
Also, many implementations do not support multiple regexps
with -e: they either reject -e entirely (e.g., Solaris)
or honor only the last pattern (e.g., IRIX 6.5 and NeXT). To
work around these problems, invoke AC_PROG_GREP
and then use
$GREP
.
Another possible workaround for the multiple -e problem is to separate the patterns by newlines, for example:
grep 'foo bar' in.txt
except that this fails with traditional grep implementations and with OpenBSD 3.8 grep.
Traditional grep implementations (e.g., Solaris) do not
support the -E or -F options. To work around these
problems, invoke AC_PROG_EGREP
and then use $EGREP
, and
similarly for AC_PROG_FGREP
and $FGREP
. Even if you are
willing to require support for Posix grep, your script should
not use both -E and -F, since Posix does not allow
this combination.
Portable grep regular expressions should use ‘\’ only to
escape characters in the string ‘$()*.0123456789[\^{}’. For example,
alternation, ‘\|’, is common but Posix does not require its
support in basic regular expressions, so it should be avoided in
portable scripts. Solaris grep does not support it.
Similarly, ‘\+’ and ‘\?’ should be avoided.
cat >file <<'EOF' 1 x 2 y EOF cat file | join file -
Use ‘join - file’ instead.
For versions of the DJGPP before 2.04,
ln emulates symbolic links
to executables by generating a stub that in turn calls the real
program. This feature also works with nonexistent files like in the
Posix spec. So ‘ln -s file link’ generates link.exe,
which attempts to call file.exe if run. But this feature only
works for executables, so ‘cp -p’ is used instead for these
systems. DJGPP versions 2.04 and later have full support
for symbolic links.
On ancient hosts, ‘ls foo’ sent the diagnostic ‘foo not found’
to standard output if foo did not exist. Hence a shell command
like ‘sources=`ls *.c 2>/dev/null`’ did not always work, since it
was equivalent to ‘sources='*.c not found'’ in the absence of
‘.c’ files. This is no longer a practical problem, since current
ls implementations send diagnostics to standard error.
AS_MKDIR_P(
file-name)
(see Programming in M4sh)
or AC_PROG_MKDIR_P
(see Particular Programs).
Posix does not clearly specify whether ‘mkdir -p foo’ should succeed when foo is a symbolic link to an already-existing directory. The GNU Core Utilities 5.1.0 mkdir succeeds, but Solaris mkdir fails.
Traditional mkdir -p
implementations suffer from race conditions.
For example, if you invoke mkdir -p a/b
and mkdir -p a/c
at the same time, both processes might detect that a is missing,
one might create a, then the other might try to create a
and fail with a File exists
diagnostic. The GNU Core
Utilities (‘fileutils’ version 4.1), FreeBSD 5.0,
NetBSD 2.0.2, and OpenBSD 2.4 are known to be
race-free when two processes invoke mkdir -p
simultaneously, but
earlier versions are vulnerable. Solaris mkdir is still
vulnerable as of Solaris 10, and other traditional Unix systems are
probably vulnerable too. This possible race is harmful in parallel
builds when several Make rules call mkdir -p
to
construct directories. You may use
install-sh -d
as a safe replacement, provided this script is
recent enough; the copy shipped with Autoconf 2.60 and Automake 1.10 is
OK, but copies from older versions are vulnerable.
Here is sample code to create a new temporary directory safely:
# Create a temporary directory $tmp in $TMPDIR (default /tmp). # Use mktemp if possible; otherwise fall back on mkdir, # with $RANDOM to make collisions less likely. : ${TMPDIR=/tmp} { tmp=` (umask 077 && mktemp -d "$TMPDIR/fooXXXXXX") 2>/dev/null ` && test -n "$tmp" && test -d "$tmp" } || { tmp=$TMPDIR/foo$$-$RANDOM (umask 077 && mkdir "$tmp") } || exit $?
Moving individual files between file systems is portable (it was in Unix version 6), but it is not always atomic: when doing ‘mv new existing’, there's a critical section where neither the old nor the new version of existing actually exists.
On some systems moving files from /tmp can sometimes cause undesirable (but perfectly valid) warnings, even if you created these files. This is because /tmp belongs to a group that ordinary users are not members of, and files created in /tmp inherit the group of /tmp. When the file is copied, mv issues a diagnostic without failing:
$ touch /tmp/foo $ mv /tmp/foo . error-->mv: ./foo: set owner/group (was: 100/0): Operation not permitted $ echo $? 0 $ ls foo foo
This annoying behavior conforms to Posix, unfortunately.
Moving directories across mount points is not portable, use cp and rm.
Moving/Deleting open files isn't portable. The following can't be done on DOS variants:
exec > foo mv foo bar
nor can
exec > foo rm -f foo
This problem no longer exists in Mac OS X 10.4.3.
Avoid empty patterns within parentheses (i.e., ‘\(\)’). Posix does not require support for empty patterns, and Unicos 9 sed rejects them.
Unicos 9 sed loops endlessly on patterns like ‘.*\n.*’.
Sed scripts should not use branch labels longer than 8 characters and should not contain comments. HP-UX sed has a limit of 99 commands (not counting ‘:’ commands) and 48 labels, which can not be circumvented by using more than one script file. It can execute up to 19 reads with the ‘r’ command per cycle. Solaris /usr/ucb/sed rejects usages that exceed an limit of about 6000 bytes for the internal representation of commands.
Avoid redundant ‘;’, as some sed implementations, such as NetBSD 1.4.2's, incorrectly try to interpret the second ‘;’ as a command:
$ echo a | sed 's/x/x/;;s/x/x/' sed: 1: "s/x/x/;;s/x/x/": invalid command code ;
Input should not have unreasonably long lines, since some sed implementations have an input buffer limited to 4000 bytes.
Portable sed regular expressions should use ‘\’ only to escape characters in the string ‘$()*.0123456789[\^n{}’. For example, alternation, ‘\|’, is common but Posix does not require its support, so it should be avoided in portable scripts. Solaris sed does not support alternation; e.g., ‘sed '/a\|b/d'’ deletes only lines that contain the literal string ‘a|b’. Similarly, ‘\+’ and ‘\?’ should be avoided.
Anchors (‘^’ and ‘$’) inside groups are not portable.
Nested parenthesization in patterns (e.g., ‘\(\(a*\)b*)\)’) is quite portable to current hosts, but was not supported by some ancient sed implementations like SVR3.
Some sed implementations, e.g., Solaris, restrict the special role of the asterisk to one-character regular expressions. This may lead to unexpected behavior:
$ echo '1*23*4' | /usr/bin/sed 's/\(.\)*/x/g' x2x4 $ echo '1*23*4' | /usr/xpg4/bin/sed 's/\(.\)*/x/g' x
The -e option is portable. Some people prefer to use it:
sed -e 'command-1' \ -e 'command-2'
as opposed to the equivalent:
sed ' command-1 command-2 '
The following usage is sometimes equivalent:
sed 'command-1;command-2'
but Posix says that this use of a semicolon has undefined effect if command-1's verb is ‘{’, ‘a’, ‘b’, ‘c’, ‘i’, ‘r’, ‘t’, ‘w’, ‘:’, or ‘#’, so you should use semicolon only with simple scripts that do not use these verbs.
Commands inside { } brackets are further restricted. Posix says that they cannot be preceded by addresses, ‘!’, or ‘;’, and that each command must be followed immediately by a newline, without any intervening blanks or semicolons. The closing bracket must be alone on a line, other than white space preceding or following it.
Contrary to yet another urban legend, you may portably use ‘&’ in
the replacement part of the s
command to mean “what was
matched”. All descendants of Unix version 7 sed
(at least; we
don't have first hand experience with older sed implementations) have
supported it.
Posix requires that you must not have any white space between ‘!’ and the following command. It is OK to have blanks between the address and the ‘!’. For instance, on Solaris:
$ echo "foo" | sed -n '/bar/ ! p' error-->Unrecognized command: /bar/ ! p $ echo "foo" | sed -n '/bar/! p' error-->Unrecognized command: /bar/! p $ echo "foo" | sed -n '/bar/ !p' foo
Posix also says that you should not combine ‘!’ and ‘;’. If you use ‘!’, it is best to put it on a command that is delimited by newlines rather than ‘;’.
Also note that Posix requires that the ‘b’, ‘t’, ‘r’, and
‘w’ commands be followed by exactly one space before their argument.
On the other hand, no white space is allowed between ‘:’ and the
subsequent label name.
s/keep me/kept/g # a t end # b s/.*/deleted/g # c :end # d
on
delete me # 1 delete me # 2 keep me # 3 delete me # 4
you get
deleted delete me kept deleted
instead of
deleted deleted kept deleted
Why? When processing line 1, (c) matches, therefore sets the ‘t’ flag, and the output is produced. When processing line 2, the ‘t’ flag is still set (this is the bug). Command (a) fails to match, but sed is not supposed to clear the ‘t’ flag when a substitution fails. Command (b) sees that the flag is set, therefore it clears it, and jumps to (d), hence you get ‘delete me’ instead of ‘deleted’. When processing line (3), ‘t’ is clear, (a) matches, so the flag is set, hence (b) clears the flags and jumps. Finally, since the flag is clear, line 4 is processed properly.
There are two things one should remember about ‘t’ in sed. Firstly, always remember that ‘t’ jumps if some substitution succeeded, not only the immediately preceding substitution. Therefore, always use a fake ‘t clear’ followed by a ‘:clear’ on the next line, to reset the ‘t’ flag where needed.
Secondly, you cannot rely on sed to clear the flag at each new cycle.
One portable implementation of the script above is:
t clear :clear s/keep me/kept/g t end s/.*/deleted/g :end
utime
or
utimes
system call, which can result in the same kind of
timestamp truncation problems that ‘cp -p’ has.
On ancient BSD systems, touch or any command that results in an empty file does not update the timestamps, so use a command like echo as a workaround. Also, GNU touch 3.16r (and presumably all before that) fails to work on SunOS 4.1.3 when the empty file is on an NFS-mounted 4.2 volume. However, these problems are no longer of practical concern.
Writing portable makefiles is an art. Since a makefile's commands are executed by the shell, you must consider the shell portability issues already mentioned. However, other issues are specific to make itself.
$<
in Ordinary Make RulesPosix says that the ‘$<’ construct in makefiles can be used only in inference rules and in the ‘.DEFAULT’ rule; its meaning in ordinary rules is unspecified. Solaris make for instance replaces it with the empty string. OpenBSD (3.0 and later) make diagnoses these uses and errors out.
Since 1992 Posix has required that make must invoke each command with the equivalent of a ‘sh -c’ subshell. However, many make implementations, including BSD make through 2004, use ‘sh -e -c’ instead, and the -e option causes the subshell to exit immediately if a subsidiary simple-command fails. For example, the command ‘touch T; rm -f U’ always attempts to remove U with Posix make, but incompatible make implementations skip the rm if the touch fails. One way to work around this is to reword the affected simple-commands so that they always succeed, e.g., ‘touch T || :; rm -f U’. However, even this approach can run into common bugs in BSD implementations of the -e option of sh and set (see Limitations of Builtins), so if you are worried about porting to buggy BSD shells it may be simpler to migrate complicated make actions into separate scripts.
Posix limits macro names to nonempty strings containing only ASCII letters and digits, ‘.’, and ‘_’. Many make implementations allow a wider variety of characters, but portable makefiles should avoid them. It is portable to start a name with a special character, e.g., ‘$(.FOO)’.
Some ancient make implementations don't support leading underscores in macro names. An example is NEWS-OS 4.2R.
$ cat Makefile _am_include = # _am_quote = all:; @echo this is test $ make Make: Must be a separator on rules line 2. Stop. $ cat Makefile2 am_include = # am_quote = all:; @echo this is test $ make -f Makefile2 this is test
However, this problem is no longer of practical concern.
On some versions of HP-UX, make reads multiple newlines following a backslash, continuing to the next non-empty line. For example,
FOO = one \ BAR = two test: : FOO is "$(FOO)" : BAR is "$(BAR)"
shows FOO
equal to one BAR = two
. Other implementations
sensibly let a backslash continue only to the immediately following
line.
According to Posix, Make comments start with #
and continue until an unescaped newline is reached.
$ cat Makefile # A = foo \ bar \ baz all: @echo ok $ make # GNU make ok
However this is not always the case. Some implementations
discard everything from #
through the end of the line, ignoring any
trailing backslash.
$ pmake # BSD make "Makefile", line 3: Need an operator Fatal errors encountered -- cannot continue
Therefore, if you want to comment out a multi-line definition, prefix each
line with #
, not only the first.
# A = foo \ # bar \ # baz
OSF/1 4.0d's make cannot process makefiles with lines
longer than 38912 bytes. It exits with a Line too long
diagnostic. A later version, Tru64 5.1's make has been
reported to crash with lines around 20 kB.
make macro=value
and SubmakesA command-line variable definition such as foo=bar
overrides any
definition of foo
in a makefile. Some make
implementations (such as GNU make) propagate this
override to subsidiary invocations of make. Some other
implementations do not pass the substitution along to submakes.
$ cat Makefile foo = foo one: @echo $(foo) $(MAKE) two two: @echo $(foo) $ make foo=bar # GNU make 3.79.1 bar make two make[1]: Entering directory `/home/adl' bar make[1]: Leaving directory `/home/adl' $ pmake foo=bar # BSD make bar pmake two foo
You have a few possibilities if you do want the foo=bar
override
to propagate to submakes. One is to use the -e
option, which causes all environment variables to have precedence over
the makefile macro definitions, and declare foo as an environment
variable:
$ env foo=bar make -e
The -e option is propagated to submakes automatically,
and since the environment is inherited between make
invocations, the foo
macro is overridden in
submakes as expected.
This syntax (foo=bar make -e
) is portable only when used
outside of a makefile, for instance from a script or from the
command line. When run inside a make rule, GNU
make 3.80 and prior versions forget to propagate the
-e option to submakes.
Moreover, using -e could have unexpected side effects if your
environment contains some other macros usually defined by the
makefile. (See also the note about make -e
and SHELL
below.)
Another way to propagate overrides to submakes is to do it manually, from your makefile:
foo = foo one: @echo $(foo) $(MAKE) foo=$(foo) two two: @echo $(foo)
You need to foresee all macros that a user might want to override if you do that.
Posix requires make to use MAKEFLAGS
to affect the
current and recursive invocations of make, but allows implementations
several formats for the variable. It is tricky to parse
$MAKEFLAGS
to determine whether -s for silent execution
or -k for continued execution are in effect. For example, you
cannot assume that the first space-separated word in $MAKEFLAGS
contains single-letter options, since in the Cygwin version of
GNU make it is either --unix or
--win32 with the second word containing single-letter options.
$ cat Makefile all: @echo MAKEFLAGS = $(MAKEFLAGS) $ make MAKEFLAGS = --unix $ make -k MAKEFLAGS = --unix -k
SHELL
Posix-compliant make internally uses the $(SHELL)
macro to spawn shell processes and execute Make rules. This
is a builtin macro supplied by make, but it can be modified
by a makefile or by a command-line argument.
Not all make implementations define this SHELL
macro.
OSF/Tru64
make is an example; this implementation always uses
/bin/sh
. So it's a good idea to always define SHELL
in
your makefiles. If you use Autoconf, do
SHELL = @SHELL@
Do not force SHELL = /bin/sh
because that is not correct
everywhere. For instance DJGPP lacks /bin/sh
, and when
its GNU make
port sees such a setting it enters a special
emulation mode where features like pipes and redirections are emulated
on top of DOS's command.com. Unfortunately this emulation is
incomplete; for instance it does not handle command substitutions.
On DJGPP SHELL
should point to Bash.
Posix-compliant make should never acquire the value of
$(SHELL) from the environment, even when make -e
is used
(otherwise, think about what would happen to your rules if
SHELL=/bin/tcsh
).
However not all make implementations have this exception.
For instance it's not surprising that OSF/Tru64 make doesn't
protect SHELL
, since it doesn't use it.
$ cat Makefile SHELL = /bin/sh FOO = foo all: @echo $(SHELL) @echo $(FOO) $ env SHELL=/bin/tcsh FOO=bar make -e # OSF1 V4.0 Make /bin/tcsh bar $ env SHELL=/bin/tcsh FOO=bar gmake -e # GNU make /bin/sh bar
Some make treat anything starting with a tab as a command for
the current rule, even if the tab is immediately followed by a #
.
The make from Tru64 Unix V5.1 is one of them. The following
makefile runs # foo
through the shell.
all: # foo
Never name one of your subdirectories obj/ if you don't like surprises.
If an obj/ directory exists, BSD make enters it before reading the makefile. Hence the makefile in the current directory is not read.
$ cat Makefile all: echo Hello $ cat obj/Makefile all: echo World $ make # GNU make echo Hello Hello $ pmake # BSD make echo World World
make -k
Do not rely on the exit status of make -k
. Some implementations
reflect whether they encountered an error in their exit status; other
implementations always succeed.
$ cat Makefile all: false $ make -k; echo exit status: $? # GNU make false make: *** [all] Error 1 exit status: 2 $ pmake -k; echo exit status: $? # BSD make false *** Error code 1 (continuing) exit status: 0
VPATH
and Make
Posix does not specify the semantics of VPATH
. Typically,
make supports VPATH
, but its implementation is not
consistent.
Autoconf and Automake support makefiles whose usages of VPATH
are
portable to recent-enough popular implementations of make, but
to keep the resulting makefiles portable, a package's makefile
prototypes must take the following issues into account. These issues
are complicated and are often poorly understood, and installers who use
VPATH
should expect to find many bugs in this area. If you use
VPATH
, the simplest way to avoid these portability bugs is to
stick with GNU make, since it is the most
commonly-used make among Autoconf users.
Here are some known issues with some VPATH
implementations.
VPATH
and Double-colon Rules
With ancient versions of Sun make,
any assignment to VPATH
causes make to execute only
the first set of double-colon rules.
However, this problem is no longer of practical concern.
$<
Not Supported in Explicit Rules
Using $<
in explicit rules is not portable.
The prerequisite file must be named explicitly in the rule. If you want
to find the prerequisite via a VPATH
search, you have to code the
whole thing manually. See Build Directories.
Some make implementations, such as Solaris make and
OSF1/Tru64 make, search for prerequisites in VPATH
and
then rewrite each occurrence as a plain word in the rule.
For instance:
# This isn't portable to GNU make. VPATH = ../pkg/src f.c: if.c cp if.c f.c
executes cp ../pkg/src/if.c f.c
if if.c is
found in ../pkg/src.
However, this rule leads to real problems in practice. For example, if the source directory contains an ordinary file named test that is used in a dependency, Solaris make rewrites commands like ‘if test -r foo; ...’ to ‘if ../pkg/src/test -r foo; ...’, which is typically undesirable. To avoid this problem, portable makefiles should never mention a source file whose name is that of a shell keyword like until or a shell command like cat or gcc or test.
Because of these problems GNU make and many other
make implementations do not rewrite commands, so portable
makefiles should
search VPATH
manually. It is tempting to write this:
# This isn't portable to Solaris make. VPATH = ../pkg/src f.c: if.c cp `test -f if.c || echo $(VPATH)/`if.c f.c
However, the “prerequisite rewriting” still applies here. So if if.c is in ../pkg/src, Solaris make and OSF1/Tru64 make executes
cp `test -f ../pkg/src/if.c || echo ../pkg/src/`if.c f.c
which reduces to
cp if.c f.c
and thus fails. Oops.
A simple workaround, and good practice anyway, is to use ‘$?’ and ‘$@’ when possible:
VPATH = ../pkg/src f.c: if.c cp $? $@
but this does not generalize well to commands with multiple prerequisites. A more general workaround is to rewrite the rule so that the prerequisite if.c never appears as a plain word. For example, these three rules would be safe, assuming if.c is in ../pkg/src and the other files are in the working directory:
VPATH = ../pkg/src f.c: if.c f1.c cat `test -f ./if.c || echo $(VPATH)/`if.c f1.c >$@ g.c: if.c g1.c cat `test -f 'if.c' || echo $(VPATH)/`if.c g1.c >$@ h.c: if.c h1.c cat `test -f "if.c" || echo $(VPATH)/`if.c h1.c >$@
Things get worse when your prerequisites are in a macro.
VPATH = ../pkg/src HEADERS = f.h g.h h.h install-HEADERS: $(HEADERS) for i in $(HEADERS); do \ $(INSTALL) -m 644 \ `test -f $$i || echo $(VPATH)/`$$i \ $(DESTDIR)$(includedir)/$$i; \ done
The above install-HEADERS
rule is not Solaris-proof because for
i in $(HEADERS);
is expanded to for i in f.h g.h h.h;
where f.h
and g.h
are plain words and are hence
subject to VPATH
adjustments.
If the three files are in ../pkg/src, the rule is run as:
for i in ../pkg/src/f.h ../pkg/src/g.h h.h; do \ install -m 644 \ `test -f $i || echo ../pkg/src/`$i \ /usr/local/include/$i; \ done
where the two first install calls fail. For instance,
consider the f.h
installation:
install -m 644 \ `test -f ../pkg/src/f.h || \ echo ../pkg/src/ \ `../pkg/src/f.h \ /usr/local/include/../pkg/src/f.h;
It reduces to:
install -m 644 \ ../pkg/src/f.h \ /usr/local/include/../pkg/src/f.h;
Note that the manual VPATH
search did not cause any problems here;
however this command installs f.h in an incorrect directory.
Trying to quote $(HEADERS)
in some way, as we did for
foo.c
a few makefiles ago, does not help:
install-HEADERS: $(HEADERS) headers='$(HEADERS)'; \ for i in $$headers; do \ $(INSTALL) -m 644 \ `test -f $$i || echo $(VPATH)/`$$i \ $(DESTDIR)$(includedir)/$$i; \ done
Now, headers='$(HEADERS)'
macroexpands to:
headers='f.h g.h h.h'
but g.h
is still a plain word. (As an aside, the idiom
headers='$(HEADERS)'; for i in $$headers;
is a good
idea if $(HEADERS)
can be empty, because some shells diagnose a
syntax error on for i in;
.)
One workaround is to strip this unwanted ../pkg/src/ prefix manually:
VPATH = ../pkg/src HEADERS = f.h g.h h.h install-HEADERS: $(HEADERS) headers='$(HEADERS)'; \ for i in $$headers; do \ i=`expr "$$i" : '$(VPATH)/\(.*\)'`; $(INSTALL) -m 644 \ `test -f $$i || echo $(VPATH)/`$$i \ $(DESTDIR)$(includedir)/$$i; \ done
Automake does something similar. However the above hack works only if
the files listed in HEADERS
are in the current directory or a
subdirectory; they should not be in an enclosing directory. If we had
HEADERS = ../f.h
, the above fragment would fail in a VPATH
build with OSF1/Tru64 make. The reason is that not only does
OSF1/Tru64 make rewrite dependencies, but it also simplifies
them. Hence ../f.h
becomes ../pkg/f.h
instead of
../pkg/src/../f.h
. This obviously defeats any attempt to strip
a leading ../pkg/src/ component.
The following example makes the behavior of OSF1/Tru64 make more apparent.
$ cat Makefile VPATH = sub all: ../foo echo ../foo $ ls Makefile foo $ make echo foo foo
Dependency ../foo was found in sub/../foo, but OSF1/Tru64 make simplified it as foo. (Note that the sub/ directory does not even exist, this just means that the simplification occurred before the file was checked for.)
For the record here is how SunOS 4 make behaves on this example.
$ make make: Fatal error: Don't know how to make target `../foo' $ mkdir sub $ make echo sub/../foo sub/../foo
When a prerequisite is a subdirectory of VPATH
, Tru64
make creates it in the current directory.
$ mkdir -p foo/bar build $ cd build $ cat >Makefile <<END VPATH = .. all: foo/bar END $ make mkdir foo mkdir foo/bar
This can yield unexpected results if a rule uses a manual VPATH
search as presented before.
VPATH = .. all : foo/bar command `test -d foo/bar || echo ../`foo/bar
The above command is run on the empty foo/bar directory that was created in the current directory.
GNU make uses a complex algorithm to decide when it
should use files found via a VPATH
search. See How Directory Searches are Performed.
If a target needs to be rebuilt, GNU make discards the
file name found during the VPATH
search for this target, and
builds the file locally using the file name given in the makefile.
If a target does not need to be rebuilt, GNU make uses the
file name found during the VPATH
search.
Other make implementations, like NetBSD make, are
easier to describe: the file name found during the VPATH
search
is used whether the target needs to be rebuilt or not. Therefore
new files are created locally, but existing files are updated at their
VPATH
location.
OpenBSD and FreeBSD make, however,
never perform a
VPATH
search for a dependency that has an explicit rule.
This is extremely annoying.
When attempting a VPATH
build for an autoconfiscated package
(e.g., mkdir build && cd build && ../configure
), this means
GNU
make builds everything locally in the build
directory, while BSD make builds new files locally and
updates existing files in the source directory.
$ cat Makefile VPATH = .. all: foo.x bar.x foo.x bar.x: newer.x @echo Building $@ $ touch ../bar.x $ touch ../newer.x $ make # GNU make Building foo.x Building bar.x $ pmake # NetBSD make Building foo.x Building ../bar.x $ fmake # FreeBSD make, OpenBSD make Building foo.x Building bar.x $ tmake # Tru64 make Building foo.x Building bar.x $ touch ../bar.x $ make # GNU make Building foo.x $ pmake # NetBSD make Building foo.x $ fmake # FreeBSD make, OpenBSD make Building foo.x Building bar.x $ tmake # Tru64 make Building foo.x Building bar.x
Note how NetBSD make updates ../bar.x in its VPATH location, and how FreeBSD, OpenBSD, and Tru64 make always update bar.x, even when ../bar.x is up to date.
Another point worth mentioning is that once GNU make has
decided to ignore a VPATH
file name (e.g., it ignored
../bar.x in the above example) it continues to ignore it when
the target occurs as a prerequisite of another rule.
The following example shows that GNU make does not look up
bar.x in VPATH
before performing the .x.y
rule,
because it ignored the VPATH
result of bar.x while running
the bar.x: newer.x
rule.
$ cat Makefile VPATH = .. all: bar.y bar.x: newer.x @echo Building $@ .SUFFIXES: .x .y .x.y: cp $< $@ $ touch ../bar.x $ touch ../newer.x $ make # GNU make Building bar.x cp bar.x bar.y cp: cannot stat `bar.x': No such file or directory make: *** [bar.y] Error 1 $ pmake # NetBSD make Building ../bar.x cp ../bar.x bar.y $ rm bar.y $ fmake # FreeBSD make, OpenBSD make echo Building bar.x cp bar.x bar.y cp: cannot stat `bar.x': No such file or directory *** Error code 1 $ tmake # Tru64 make Building bar.x cp: bar.x: No such file or directory *** Exit 1
Note that if you drop away the command from the bar.x: newer.x
rule, GNU make magically starts to work: it
knows that bar.x
hasn't been updated, therefore it doesn't
discard the result from VPATH
(../bar.x) in succeeding
uses. Tru64 also works, but FreeBSD and OpenBSD
still don't.
$ cat Makefile VPATH = .. all: bar.y bar.x: newer.x .SUFFIXES: .x .y .x.y: cp $< $@ $ touch ../bar.x $ touch ../newer.x $ make # GNU make cp ../bar.x bar.y $ rm bar.y $ pmake # NetBSD make cp ../bar.x bar.y $ rm bar.y $ fmake # FreeBSD make, OpenBSD make cp bar.x bar.y cp: cannot stat `bar.x': No such file or directory *** Error code 1 $ tmake # Tru64 make cp ../bar.x bar.y
It seems the sole solution that would please every make
implementation is to never rely on VPATH
searches for targets.
In other words, VPATH
should be reserved to unbuilt sources.
A Single Suffix Rule is basically a usual suffix (inference) rule (‘.from.to:’), but which destination suffix is empty (‘.from:’).
Separated dependencies simply refers to listing the prerequisite of a target, without defining a rule. Usually one can list on the one hand side, the rules, and on the other hand side, the dependencies.
Solaris make does not support separated dependencies for targets defined by single suffix rules:
$ cat Makefile .SUFFIXES: .in foo: foo.in .in: cp $< $@ $ touch foo.in $ make $ ls Makefile foo.in
while GNU Make does:
$ gmake cp foo.in foo $ ls Makefile foo foo.in
Note it works without the ‘foo: foo.in’ dependency.
$ cat Makefile .SUFFIXES: .in .in: cp $< $@ $ make foo cp foo.in foo
and it works with double suffix inference rules:
$ cat Makefile foo.out: foo.in .SUFFIXES: .in .out .in.out: cp $< $@ $ make cp foo.in foo.out
As a result, in such a case, you have to write target rules.
Traditionally, file timestamps had 1-second resolution, and make used those timestamps to determine whether one file was newer than the other. However, many modern file systems have timestamps with 1-nanosecond resolution. Some make implementations look at the entire timestamp; others ignore the fractional part, which can lead to incorrect results. Normally this is not a problem, but in some extreme cases you may need to use tricks like ‘sleep 1’ to work around timestamp truncation bugs.
Commands like ‘cp -p’ and ‘touch -r’ typically do not copy file timestamps to their full resolutions (see Limitations of Usual Tools). Hence you should be wary of rules like this:
dest: src cp -p src dest
as dest often appears to be older than src after the timestamp is truncated, and this can cause make to do needless rework the next time it is invoked. To work around this problem, you can use a timestamp file, e.g.:
dest-stamp: src cp -p src dest date >dest-stamp
C and C++ programs often use low-level features of the underlying system, and therefore are often more difficult to make portable to other platforms.
Several standards have been developed to help make your programs more portable. If you write programs with these standards in mind, you can have greater confidence that your programs work on a wide variety of systems. See Language Standards Supported by GCC, for a list of C-related standards. Many programs also assume the Posix standard.
Some old code is written to be portable to K&R C, which predates any C standard. K&R C compilers are no longer of practical interest, though, and the rest of section assumes at least C89, the first C standard.
Program portability is a huge topic, and this section can only briefly introduce common pitfalls. See Portability between System Types, for more information.
Autoconf tests and ordinary programs often need to test what is allowed on a system, and therefore they may need to deliberately exceed the boundaries of what the standards allow, if only to see whether an optional feature is present. When you write such a program, you should keep in mind the difference between constraints, unspecified behavior, and undefined behavior.
In C, a constraint is a rule that the compiler must enforce. An example constraint is that C programs must not declare a bit-field with negative width. Tests can therefore reliably assume that programs with negative-width bit-fields are rejected by a compiler that conforms to the standard.
Unspecified behavior is valid behavior, where the standard allows multiple possibilities. For example, the order of evaluation of function arguments is unspecified. Some unspecified behavior is implementation-defined, i.e., documented by the implementation, but since Autoconf tests cannot read the documentation they cannot distinguish between implementation-defined and other unspecified behavior. It is common for Autoconf tests to probe implementations to determine otherwise-unspecified behavior.
Undefined behavior is invalid behavior, where the standard allows the implementation to do anything it pleases. For example, dereferencing a null pointer leads to undefined behavior. If possible, test programs should avoid undefined behavior, since a program with undefined behavior might succeed on a test that should fail.
The above rules apply to programs that are intended to conform to the standard. However, strictly-conforming programs are quite rare, since the standards are so limiting. A major goal of Autoconf is to support programs that use implementation features not described by the standard, and it is fairly common for test programs to violate the above rules, if the programs work well enough in practice.
In C, signed integer overflow leads to undefined behavior. However,
many programs and Autoconf tests assume that signed integer overflow after
addition, subtraction, or multiplication silently
wraps around modulo a power of two, using two's complement arithmetic,
so long as you cast the resulting value
to an integer type or store it into an integer variable. Such programs
are portable to the vast majority of modern platforms. However, signed
integer division is not always harmless: for example, on CPUs of the
i386 family, dividing INT_MIN
by -1
yields a SIGFPE signal
which by default terminates the program. Worse, taking the remainder
of these two values typically yields the same signal on these CPUs,
even though the C standard requires INT_MIN % -1
to yield zero
because the expression does not overflow.
GCC users might consider using the -ftrapv option if they are worried about porting their code to the rare platforms where signed integer overflow does not wrap around after addition, subtraction, or multiplication.
Unsigned integer overflow reliably wraps around modulo the word size. This is guaranteed by the C standard and is portable in practice.
Most modern hosts reliably fail when you attempt to dereference a null pointer.
On almost all modern hosts, null pointers use an all-bits-zero internal
representation, so you can reliably use memset
with 0 to set all
the pointers in an array to null values.
If p
is a null pointer to an object type, the C expression
p + 0
always evaluates to p
on modern hosts, even though
the standard says that it has undefined behavior.
Buffer overruns and subscript errors are the most common dangerous errors in C programs. They result in undefined behavior because storing outside an array typically modifies storage that is used by some other object, and most modern systems lack runtime checks to catch these errors. Programs should not rely on buffer overruns being caught.
There is one exception to the usual rule that a portable program cannot
address outside an array. In C, it is valid to compute the address just
past an object, e.g., &a[N]
where a
has N
elements,
so long as you do not dereference the resulting pointer. But it is not
valid to compute the address just before an object, e.g., &a[-1]
;
nor is it valid to compute two past the end, e.g., &a[N+1]
. On
most platforms &a[-1] < &a[0] && &a[N] < &a[N+1]
, but this is not
reliable in general, and it is usually easy enough to avoid the
potential portability problem, e.g., by allocating an extra unused array
element at the start or end.
Valgrind can catch many overruns. GCC users might also consider using the -fmudflap option to catch overruns.
Buffer overruns are usually caused by off-by-one errors, but there are more subtle ways to get them.
Using int
values to index into an array or compute array sizes
causes problems on typical 64-bit hosts where an array index might
be 2^31 or larger. Index values of type size_t
avoid this
problem, but cannot be negative. Index values of type ptrdiff_t
are signed, and are wide enough in practice.
If you add or multiply two numbers to calculate an array size, e.g.,
malloc (x * sizeof y + z)
, havoc ensues if the addition or
multiplication overflows.
Many implementations of the alloca
function silently misbehave
and can generate buffer overflows if given sizes that are too large.
The size limits are implementation dependent, but are at least 4000
bytes on all platforms that we know about.
The standard functions asctime
, asctime_r
, ctime
,
ctime_r
, and gets
are prone to buffer overflows, and
portable code should not use them unless the inputs are known to be
within certain limits. The time-related functions can overflow their
buffers if given timestamps out of range (e.g., a year less than -999
or greater than 9999). Time-related buffer overflows cannot happen with
recent-enough versions of the GNU C library, but are possible
with other
implementations. The gets
function is the worst, since it almost
invariably overflows its buffer when presented with an input line larger
than the buffer.
Almost all modern systems use IEEE-754 floating point, and it is safe to assume IEEE-754 in most portable code these days. For more information, please see David Goldberg's classic paper What Every Computer Scientist Should Know About Floating-Point Arithmetic.
A C or C++ program can exit with status N by returning
N from the main
function. Portable programs are supposed
to exit either with status 0 or EXIT_SUCCESS
to succeed, or with
status EXIT_FAILURE
to fail, but in practice it is portable to
fail by exiting with status 1, and test programs that assume Posix can
fail by exiting with status values from 1 through 255. Programs on
SunOS 2.0 (1985) through 3.5.2 (1988) incorrectly exited with zero
status when main
returned nonzero, but ancient systems like these
are no longer of practical concern.
A program can also exit with status N by passing N to the
exit
function, and a program can fail by calling the abort
function. If a program is specialized to just some platforms, it can fail
by calling functions specific to those platforms, e.g., _exit
(Posix) and _Exit
(C99). However, like other functions, an exit
function should be declared, typically by including a header. For
example, if a C program calls exit
, it should include stdlib.h
either directly or via the default includes (see Default Includes).
A program can fail due to undefined behavior such as dereferencing a null pointer, but this is not recommended as undefined behavior allows an implementation to do whatever it pleases and this includes exiting successfully.
A few kinds of features can't be guessed automatically by running test
programs. For example, the details of the object-file format, or
special options that need to be passed to the compiler or linker. You
can check for such features using ad-hoc means, such as having
configure check the output of the uname
program, or
looking for libraries that are unique to particular systems. However,
Autoconf provides a uniform method for handling unguessable features.
Like other GNU configure scripts, Autoconf-generated configure scripts can make decisions based on a canonical name for the system type, which has the form: ‘cpu-vendor-os’, where os can be ‘system’ or ‘kernel-system’
configure can usually guess the canonical name for the type of
system it's running on. To do so it runs a script called
config.guess, which infers the name using the uname
command or symbols predefined by the C preprocessor.
Alternately, the user can specify the system type with command line arguments to configure. Doing so is necessary when cross-compiling. In the most complex case of cross-compiling, three system types are involved. The options to specify them are:
If you mean to override the result of config.guess, use --build, not --host, since the latter enables cross-compilation. For historical reasons, passing --host also changes the build type. Therefore, whenever you specify --host, be sure to specify --build too; this will be fixed in the future. So, to enter cross-compilation mode, use a command like this
./configure --build=i686-pc-linux-gnu --host=m68k-coff
Note that if you do not specify --host, configure fails if it can't run the code generated by the specified compiler. For example, configuring as follows fails:
./configure CC=m68k-coff-gcc
In the future, when cross-compiling Autoconf will not
accept tools (compilers, linkers, assemblers) whose name is not
prefixed with the host type. The only case when this may be
useful is when you really are not cross-compiling, but only
building for a least-common-denominator architecture: an example
is building for i386-pc-linux-gnu
while running on an
i686-pc-linux-gnu
architecture. In this case, some particular
pairs might be similar enough to let you get away with the system
compilers, but in general the compiler might make bogus assumptions
on the host: if you know what you are doing, please create symbolic
links from the host compiler to the build compiler.
configure recognizes short aliases for many system types; for example, ‘decstation’ can be used instead of ‘mips-dec-ultrix4.2’. configure runs a script called config.sub to canonicalize system type aliases.
This section deliberately omits the description of the obsolete interface; see Hosts and Cross-Compilation.
The following macros make the system type available to configure scripts.
The variables ‘build_alias’, ‘host_alias’, and
‘target_alias’ are always exactly the arguments of --build,
--host, and --target; in particular, they are left empty
if the user did not use them, even if the corresponding
AC_CANONICAL
macro was run. Any configure script may use these
variables anywhere. These are the variables that should be used when in
interaction with the user.
If you need to recognize some special environments based on their system type, run the following macros to get canonical system names. These variables are not set before the macro call.
If you use these macros, you must distribute config.guess and
config.sub along with your source code. See Output, for
information about the AC_CONFIG_AUX_DIR
macro which you can use
to control in which directory configure looks for those scripts.
Compute the canonical build-system type variable,
build
, and its three individual partsbuild_cpu
,build_vendor
, andbuild_os
.If --build was specified, then
build
is the canonicalization ofbuild_alias
by config.sub, otherwise it is determined by the shell script config.guess.
Compute the canonical host-system type variable,
host
, and its three individual partshost_cpu
,host_vendor
, andhost_os
.If --host was specified, then
host
is the canonicalization ofhost_alias
by config.sub, otherwise it defaults tobuild
.
Compute the canonical target-system type variable,
target
, and its three individual partstarget_cpu
,target_vendor
, andtarget_os
.If --target was specified, then
target
is the canonicalization oftarget_alias
by config.sub, otherwise it defaults tohost
.
Note that there can be artifacts due to the backward compatibility code. See See Hosts and Cross-Compilation, for more.
In configure.ac the system type is generally used by one or more
case
statements to select system-specifics. Shell wildcards can
be used to match a group of system types.
For example, an extra assembler code object file could be chosen, giving
access to a CPU cycle counter register. $(CYCLE_OBJ)
in the
following would be used in a makefile to add the object to a
program or library.
case $host in alpha*-*-*) CYCLE_OBJ=rpcc.o ;; i?86-*-*) CYCLE_OBJ=rdtsc.o ;; *) CYCLE_OBJ= ;; esac AC_SUBST([CYCLE_OBJ])
AC_CONFIG_LINKS
(see Configuration Links) is another good way
to select variant source files, for example optimized code for some
CPUs. The configured CPU type doesn't always indicate exact CPU types,
so some runtime capability checks may be necessary too.
case $host in alpha*-*-*) AC_CONFIG_LINKS([dither.c:alpha/dither.c]) ;; powerpc*-*-*) AC_CONFIG_LINKS([dither.c:powerpc/dither.c]) ;; *-*-*) AC_CONFIG_LINKS([dither.c:generic/dither.c]) ;; esac
The host system type can also be used to find cross-compilation tools
with AC_CHECK_TOOL
(see Generic Programs).
The above examples all show ‘$host’, since this is where the code is going to run. Only rarely is it necessary to test ‘$build’ (which is where the build is being done).
Whenever you're tempted to use ‘$host’ it's worth considering whether some sort of probe would be better. New system types come along periodically or previously missing features are added. Well-written probes can adapt themselves to such things, but hard-coded lists of names can't. Here are some guidelines,
‘$target’ is for use by a package creating a compiler or similar. For ordinary packages it's meaningless and should not be used. It indicates what the created compiler should generate code for, if it can cross-compile. ‘$target’ generally selects various hard-coded CPU and system conventions, since usually the compiler or tools under construction themselves determine how the target works.
configure scripts support several kinds of local configuration decisions. There are ways for users to specify where external software packages are, include or exclude optional features, install programs under modified names, and set default values for configure options.
Users consult ‘configure --help’ to learn of configuration decisions specific to your package. By default, configure breaks this output into sections for each type of option; within each section, help strings appear in the order configure.ac defines them:
Optional Features: ... --enable-bar include bar Optional Packages: ... --with-foo use foo
Request an alternate --help format, in which options of all types appear together, in the order defined. Call this macro before any
AC_ARG_ENABLE
orAC_ARG_WITH
.Optional Features and Packages: ... --enable-bar include bar --with-foo use foo
Some packages require, or can optionally use, other software packages that are already installed. The user can give configure command line options to specify which such external software to use. The options have one of these forms:
--with-package[=arg] --without-package
For example, --with-gnu-ld means work with the GNU linker instead of some other linker. --with-x means work with The X Window System.
The user can give an argument by following the package name with ‘=’ and the argument. Giving an argument of ‘no’ is for packages that are used by default; it says to not use the package. An argument that is neither ‘yes’ nor ‘no’ could include a name or number of a version of the other package, to specify more precisely which other package this program is supposed to work with. If no argument is given, it defaults to ‘yes’. --without-package is equivalent to --with-package=no.
configure scripts do not complain about --with-package options that they do not support. This behavior permits configuring a source tree containing multiple packages with a top-level configure script when the packages support different options, without spurious error messages about options that some of the packages support. An unfortunate side effect is that option spelling errors are not diagnosed. No better approach to this problem has been suggested so far.
For each external software package that may be used, configure.ac
should call AC_ARG_WITH
to detect whether the configure
user asked to use it. Whether each package is used or not by default,
and which arguments are valid, is up to you.
If the user gave configure the option --with-package or --without-package, run shell commands action-if-given. If neither option was given, run shell commands action-if-not-given. The name package indicates another software package that this program should work with. It should consist only of alphanumeric characters and dashes.
The option's argument is available to the shell commands action-if-given in the shell variable
withval
, which is actually just the value of the shell variablewith_
package, with any - characters changed into ‘_’. You may use that variable instead, if you wish.The argument help-string is a description of the option that looks like this:
--with-readline support fancy command line editinghelp-string may be more than one line long, if more detail is needed. Just make sure the columns line up in ‘configure --help’. Avoid tabs in the help string. You'll need to enclose the help string in ‘[’ and ‘]’ in order to produce the leading blanks.
You should format your help-string with the macro
AS_HELP_STRING
(see Pretty Help Strings).The following example shows how to use the
AC_ARG_WITH
macro in a common situation. You want to let the user decide whether to enable support for an external library (e.g., the readline library); if the user specified neither --with-readline nor --without-readline, you want to enable support for readline only if the library is available on the system.AC_ARG_WITH([readline], [AS_HELP_STRING([--with-readline], [support fancy command line editing @<:@default=check@:>@])], [], [with_readline=check]) LIBREADLINE= AS_IF([test "x$with_readline" != xno], [AC_CHECK_LIB([readline], [main], [AC_SUBST([LIBREADLINE], ["-lreadline -lncurses"]) AC_DEFINE([HAVE_LIBREADLINE], [1], [Define if you have libreadline]) ], [if test "x$with_readline" != xcheck; then AC_MSG_FAILURE( [--with-readline was given, but test for readline failed]) fi ], -lncurses)])The next example shows how to use
AC_ARG_WITH
to give the user the possibility to enable support for the readline library, in case it is still experimental and not well tested, and is therefore disabled by default.AC_ARG_WITH([readline], [AS_HELP_STRING([--with-readline], [enable experimental support for readline])], [], [with_readline=no]) LIBREADLINE= AS_IF([test "x$with_readline" != xno], [AC_CHECK_LIB([readline], [main], [AC_SUBST([LIBREADLINE], ["-lreadline -lncurses"]) AC_DEFINE([HAVE_LIBREADLINE], [1], [Define if you have libreadline]) ], [AC_MSG_FAILURE( [--with-readline was given, but test for readline failed])], [-lncurses])])The last example shows how to use
AC_ARG_WITH
to give the user the possibility to disable support for the readline library, given that it is an important feature and that it should be enabled by default.AC_ARG_WITH([readline], [AS_HELP_STRING([--without-readline], [disable support for readline])], [], [with_readline=yes]) LIBREADLINE= AS_IF([test "x$with_readline" != xno], [AC_CHECK_LIB([readline], [main], [AC_SUBST([LIBREADLINE], ["-lreadline -lncurses"]) AC_DEFINE([HAVE_LIBREADLINE], [1], [Define if you have libreadline]) ], [AC_MSG_FAILURE( [readline test failed (--without-readline to disable)])], [-lncurses])])These three examples can be easily adapted to the case where
AC_ARG_ENABLE
should be preferred toAC_ARG_WITH
(see Package Options).
This is an obsolete version of
AC_ARG_WITH
that does not support providing a help string.
If a software package has optional compile-time features, the user can give configure command line options to specify whether to compile them. The options have one of these forms:
--enable-feature[=arg] --disable-feature
These options allow users to choose which optional features to build and install. --enable-feature options should never make a feature behave differently or cause one feature to replace another. They should only cause parts of the program to be built rather than left out.
The user can give an argument by following the feature name with ‘=’ and the argument. Giving an argument of ‘no’ requests that the feature not be made available. A feature with an argument looks like --enable-debug=stabs. If no argument is given, it defaults to ‘yes’. --disable-feature is equivalent to --enable-feature=no.
configure scripts do not complain about --enable-feature options that they do not support. This behavior permits configuring a source tree containing multiple packages with a top-level configure script when the packages support different options, without spurious error messages about options that some of the packages support. An unfortunate side effect is that option spelling errors are not diagnosed. No better approach to this problem has been suggested so far.
For each optional feature, configure.ac should call
AC_ARG_ENABLE
to detect whether the configure user asked
to include it. Whether each feature is included or not by default, and
which arguments are valid, is up to you.
If the user gave configure the option --enable-feature or --disable-feature, run shell commands action-if-given. If neither option was given, run shell commands action-if-not-given. The name feature indicates an optional user-level facility. It should consist only of alphanumeric characters and dashes.
The option's argument is available to the shell commands action-if-given in the shell variable
enableval
, which is actually just the value of the shell variableenable_
feature, with any - characters changed into ‘_’. You may use that variable instead, if you wish. The help-string argument is like that ofAC_ARG_WITH
(see External Software).You should format your help-string with the macro
AS_HELP_STRING
(see Pretty Help Strings).See the examples suggested with the definition of
AC_ARG_WITH
(see External Software) to get an idea of possible applications ofAC_ARG_ENABLE
.
This is an obsolete version of
AC_ARG_ENABLE
that does not support providing a help string.
Properly formatting the ‘help strings’ which are used in
AC_ARG_WITH
(see External Software) and AC_ARG_ENABLE
(see Package Options) can be challenging. Specifically, you want
your own ‘help strings’ to line up in the appropriate columns of
‘configure --help’ just like the standard Autoconf ‘help
strings’ do. This is the purpose of the AS_HELP_STRING
macro.
Expands into an help string that looks pretty when the user executes ‘configure --help’. It is typically used in
AC_ARG_WITH
(see External Software) orAC_ARG_ENABLE
(see Package Options). The following example makes this clearer.AC_ARG_WITH([foo], [AS_HELP_STRING([--with-foo], [use foo (default is no)])], [use_foo=$withval], [use_foo=no])The second argument of
AS_HELP_STRING
is not a literal, and should not be double quoted. See Autoconf Language, for a more detailed explanation. Then the last few lines of ‘configure --help’ appear like this:--enable and --with options recognized: --with-foo use foo (default is no)The
AS_HELP_STRING
macro is particularly helpful when the left-hand-side and/or right-hand-side are composed of macro arguments, as shown in the following example.AC_DEFUN([MY_ARG_WITH], [AC_ARG_WITH([$1], [AS_HELP_STRING([--with-$1], [use $1 (default is $2)])], [use_[]$1=$withval], [use_[]$1=$2])])
Some software packages require complex site-specific information. Some examples are host names to use for certain services, company names, and email addresses to contact. Since some configuration scripts generated by Metaconfig ask for such information interactively, people sometimes wonder how to get that information in Autoconf-generated configuration scripts, which aren't interactive.
Such site configuration information should be put in a file that is
edited only by users, not by programs. The location of the file
can either be based on the prefix
variable, or be a standard
location such as the user's home directory. It could even be specified
by an environment variable. The programs should examine that file at
runtime, rather than at compile time. Runtime configuration is more
convenient for users and makes the configuration process simpler than
getting the information while configuring. See Variables for Installation Directories, for more information on where to put data files.
Autoconf supports changing the names of programs when installing them.
In order to use these transformations, configure.ac must call the
macro AC_ARG_PROGRAM
.
Place in output variable
program_transform_name
a sequence ofsed
commands for changing the names of installed programs.If any of the options described below are given to configure, program names are transformed accordingly. Otherwise, if
AC_CANONICAL_TARGET
has been called and a --target value is given, the target type followed by a dash is used as a prefix. Otherwise, no program name transformation is done.
You can specify name transformations by giving configure these command line options:
sed
substitution expression on the names.
These transformations are useful with programs that can be part of a cross-compilation development environment. For example, a cross-assembler running on a Sun 4 configured with --target=i960-vxworks is normally installed as i960-vxworks-as, rather than as, which could be confused with a native Sun 4 assembler.
You can force a program name to begin with g, if you don't want
GNU programs installed on your system to shadow other programs with
the same name. For example, if you configure GNU diff
with
--program-prefix=g, then when you run ‘make install’ it is
installed as /usr/local/bin/gdiff.
As a more sophisticated example, you could use
--program-transform-name='s/^/g/; s/^gg/g/; s/^gless/less/'
to prepend ‘g’ to most of the program names in a source tree,
excepting those like gdb
that already have one and those like
less
and lesskey
that aren't GNU programs. (That is
assuming that you have a source tree containing those programs that is
set up to use this feature.)
One way to install multiple versions of some programs simultaneously is to append a version number to the name of one or both. For example, if you want to keep Autoconf version 1 around for awhile, you can configure Autoconf version 2 using --program-suffix=2 to install the programs as /usr/local/bin/autoconf2, /usr/local/bin/autoheader2, etc. Nevertheless, pay attention that only the binaries are renamed, therefore you'd have problems with the library files which might overlap.
Here is how to use the variable program_transform_name
in a
Makefile.in:
PROGRAMS = cp ls rm transform = @program_transform_name@ install: for p in $(PROGRAMS); do \ $(INSTALL_PROGRAM) $$p $(DESTDIR)$(bindir)/`echo $$p | \ sed '$(transform)'`; \ done uninstall: for p in $(PROGRAMS); do \ rm -f $(DESTDIR)$(bindir)/`echo $$p | sed '$(transform)'`; \ done
It is guaranteed that program_transform_name
is never empty, and
that there are no useless separators. Therefore you may safely embed
program_transform_name
within a sed program using ‘;’:
transform = @program_transform_name@ transform_exe = s/$(EXEEXT)$$//;$(transform);s/$$/$(EXEEXT)/
Whether to do the transformations on documentation files (Texinfo or
man
) is a tricky question; there seems to be no perfect answer,
due to the several reasons for name transforming. Documentation is not
usually particular to a specific architecture, and Texinfo files do not
conflict with system documentation. But they might conflict with
earlier versions of the same files, and man
pages sometimes do
conflict with system documentation. As a compromise, it is probably
best to do name transformations on man
pages but not on Texinfo
manuals.
Autoconf-generated configure scripts allow your site to provide default values for some configuration values. You do this by creating site- and system-wide initialization files.
If the environment variable CONFIG_SITE
is set, configure
uses its value as the name of a shell script to read. Otherwise, it
reads the shell script prefix/share/config.site if it exists,
then prefix/etc/config.site if it exists. Thus,
settings in machine-specific files override those in machine-independent
ones in case of conflict.
Site files can be arbitrary shell scripts, but only certain kinds of
code are really appropriate to be in them. Because configure
reads any cache file after it has read any site files, a site file can
define a default cache file to be shared between all Autoconf-generated
configure scripts run on that system (see Cache Files). If
you set a default cache file in a site file, it is a good idea to also
set the output variable CC
in that site file, because the cache
file is only valid for a particular compiler, but many systems have
several available.
You can examine or override the value set by a command line option to
configure in a site file; options set shell variables that have
the same names as the options, with any dashes turned into underscores.
The exceptions are that --without- and --disable- options
are like giving the corresponding --with- or --enable-
option and the value ‘no’. Thus, --cache-file=localcache
sets the variable cache_file
to the value ‘localcache’;
--enable-warnings=no or --disable-warnings sets the variable
enable_warnings
to the value ‘no’; --prefix=/usr sets the
variable prefix
to the value ‘/usr’; etc.
Site files are also good places to set default values for other output
variables, such as CFLAGS
, if you need to give them non-default
values: anything you would normally do, repetitively, on the command
line. If you use non-default values for prefix or
exec_prefix (wherever you locate the site file), you can set them
in the site file if you specify it with the CONFIG_SITE
environment variable.
You can set some cache values in the site file itself. Doing this is useful if you are cross-compiling, where it is impossible to check features that require running a test program. You could “prime the cache” by setting those values correctly for that system in prefix/etc/config.site. To find out the names of the cache variables you need to set, look for shell variables with ‘_cv_’ in their names in the affected configure scripts, or in the Autoconf M4 source code for those macros.
The cache file is careful to not override any variables set in the site
files. Similarly, you should not override command-line options in the
site files. Your code should check that variables such as prefix
and cache_file
have their default values (as set near the top of
configure) before changing them.
Here is a sample file /usr/share/local/gnu/share/config.site. The
command ‘configure --prefix=/usr/share/local/gnu’ would read this
file (if CONFIG_SITE
is not set to a different file).
# config.site for configure # # Change some defaults. test "$prefix" = NONE && prefix=/usr/share/local/gnu test "$exec_prefix" = NONE && exec_prefix=/usr/local/gnu test "$sharedstatedir" = '$prefix/com' && sharedstatedir=/var test "$localstatedir" = '$prefix/var' && localstatedir=/var # Give Autoconf 2.x generated configure scripts a shared default # cache file for feature test results, architecture-specific. if test "$cache_file" = /dev/null; then cache_file="$prefix/var/config.cache" # A cache file is only valid for one C compiler. CC=gcc fi
Below are instructions on how to configure a package that uses a configure script, suitable for inclusion as an INSTALL file in the package. A plain-text version of INSTALL which you may use comes with Autoconf.
Briefly, the shell commands ‘./configure; make; make install’ should configure, build, and install this package. The following more-detailed instructions are generic; see the README file for instructions specific to this package.
The configure shell script attempts to guess correct values for various system-dependent variables used during compilation. It uses those values to create a Makefile in each directory of the package. It may also create one or more .h files containing system-dependent definitions. Finally, it creates a shell script config.status that you can run in the future to recreate the current configuration, and a file config.log containing compiler output (useful mainly for debugging configure).
It can also use an optional file (typically called config.cache and enabled with --cache-file=config.cache or simply -C) that saves the results of its tests to speed up reconfiguring. Caching is disabled by default to prevent problems with accidental use of stale cache files.
If you need to do unusual things to compile the package, please try to figure out how configure could check whether to do them, and mail diffs or instructions to the address given in the README so they can be considered for the next release. If you are using the cache, and at some point config.cache contains results you don't want to keep, you may remove or edit it.
The file configure.ac (or configure.in) is used to create
configure by a program called autoconf
. You need
configure.ac if you want to change it or regenerate
configure using a newer version of autoconf
.
The simplest way to compile this package is:
cd
to the directory containing the package's source code and type
‘./configure’ to configure the package for your system.
Running configure might take a while. While running, it prints some messages telling which features it is checking for.
Some systems require unusual options for compilation or linking that the configure script does not know about. Run ‘./configure --help’ for details on some of the pertinent environment variables.
You can give configure initial values for configuration parameters by setting variables in the command line or in the environment. Here is an example:
./configure CC=c99 CFLAGS=-g LIBS=-lposix
See Defining Variables, for more details.
You can compile the package for more than one kind of computer at the same time, by placing the object files for each architecture in their own directory. To do this, you can use GNU make. cd to the directory where you want the object files and executables to go and run the configure script. configure automatically checks for the source code in the directory that configure is in and in ...
With a non-GNU make, it is safer to compile the package for one architecture at a time in the source code directory. After you have installed the package for one architecture, use ‘make distclean’ before reconfiguring for another architecture.
By default, ‘make install’ installs the package's commands under /usr/local/bin, include files under /usr/local/include, etc. You can specify an installation prefix other than /usr/local by giving configure the option --prefix=prefix.
You can specify separate installation prefixes for architecture-specific files and architecture-independent files. If you pass the option --exec-prefix=prefix to configure, the package uses prefix as the prefix for installing programs and libraries. Documentation and other data files still use the regular prefix.
In addition, if you use an unusual directory layout you can give options like --bindir=dir to specify different values for particular kinds of files. Run ‘configure --help’ for a list of the directories you can set and what kinds of files go in them.
If the package supports it, you can cause programs to be installed with an extra prefix or suffix on their names by giving configure the option --program-prefix=PREFIX or --program-suffix=SUFFIX.
Some packages pay attention to --enable-feature options to configure, where feature indicates an optional part of the package. They may also pay attention to --with-package options, where package is something like ‘gnu-as’ or ‘x’ (for the X Window System). The README should mention any --enable- and --with- options that the package recognizes.
For packages that use the X Window System, configure can usually find the X include and library files automatically, but if it doesn't, you can use the configure options --x-includes=dir and --x-libraries=dir to specify their locations.
There may be some features configure cannot figure out automatically, but needs to determine by the type of machine the package will run on. Usually, assuming the package is built to be run on the same architectures, configure can figure that out, but if it prints a message saying it cannot guess the machine type, give it the --build=type option. type can either be a short name for the system type, such as ‘sun4’, or a canonical name which has the form:
cpu-company-system
where system can have one of these forms:
os kernel-os
See the file config.sub for the possible values of each field. If config.sub isn't included in this package, then this package doesn't need to know the machine type.
If you are building compiler tools for cross-compiling, you should use the option --target=type to select the type of system they will produce code for.
If you want to use a cross compiler, that generates code for a platform different from the build platform, you should specify the host platform (i.e., that on which the generated programs will eventually be run) with --host=type.
If you want to set default values for configure scripts to
share, you can create a site shell script called config.site that
gives default values for variables like CC
, cache_file
,
and prefix
. configure looks for
prefix/share/config.site if it exists, then
prefix/etc/config.site if it exists. Or, you can set the
CONFIG_SITE
environment variable to the location of the site
script. A warning: not all configure scripts look for a site
script.
Variables not defined in a site shell script can be set in the environment passed to configure. However, some packages may run configure again during the build, and the customized values of these variables may be lost. In order to avoid this problem, you should set them in the configure command line, using ‘VAR=value’. For example:
./configure CC=/usr/local2/bin/gcc
causes the specified gcc to be used as the C compiler (unless it is overridden in the site shell script).
Unfortunately, this technique does not work for CONFIG_SHELL due to an Autoconf bug. Until the bug is fixed you can use this workaround:
CONFIG_SHELL=/bin/bash /bin/bash ./configure CONFIG_SHELL=/bin/bash
configure recognizes the following options to control how it operates.
configure also accepts some other, not widely useful, options. Run ‘configure --help’ for more details.
The configure script creates a file named config.status, which actually configures, instantiates, the template files. It also records the configuration options that were specified when the package was last configured in case reconfiguring is needed.
Synopsis:
./config.status option... [file...]
It configures the files; if none are specified, all the templates are instantiated. The files must be specified without their dependencies, as in
./config.status foobar
not
./config.status foobar:foo.in:bar.in
The supported options are:
This option and the following ones provide one way for separately
distributed packages to share the values computed by configure.
Doing so can be useful if some of the packages need a superset of the
features that one of them, perhaps a common library, does. These
options allow a config.status file to create files other than the
ones that its configure.ac specifies, so it can be used for a
different package.
config.status checks several optional environment variables that can alter its behavior:
The shell with which to run configure for the --recheck option. It must be Bourne-compatible. The default is a shell that supports
LINENO
if available, and /bin/sh otherwise. Invoking configure by hand bypasses this setting, so you may need to use a command like ‘CONFIG_SHELL=/bin/bash /bin/bash ./configure’ to insure that the same shell is used everywhere. The absolute name of the shell should be passed.
The file name to use for the shell script that records the configuration. The default is ./config.status. This variable is useful when one package uses parts of another and the configure scripts shouldn't be merged because they are maintained separately.
You can use ./config.status in your makefiles. For example, in the dependencies given above (see Automatic Remaking), config.status is run twice when configure.ac has changed. If that bothers you, you can make each run only regenerate the files for that rule:
config.h: stamp-h stamp-h: config.h.in config.status ./config.status config.h echo > stamp-h Makefile: Makefile.in config.status ./config.status Makefile
The calling convention of config.status has changed; see Obsolete config.status Use, for details.
Autoconf changes, and throughout the years some constructs have been obsoleted. Most of the changes involve the macros, but in some cases the tools themselves, or even some concepts, are now considered obsolete.
You may completely skip this chapter if you are new to Autoconf. Its intention is mainly to help maintainers updating their packages by understanding how to move to more modern constructs.
config.status now supports arguments to specify the files to instantiate; see config.status Invocation, for more details. Before, environment variables had to be used.
The tags of the commands to execute. The default is the arguments given to
AC_OUTPUT
andAC_CONFIG_COMMANDS
in configure.ac.
The files in which to perform ‘@variable@’ substitutions. The default is the arguments given to
AC_OUTPUT
andAC_CONFIG_FILES
in configure.ac.
The files in which to substitute C
#define
statements. The default is the arguments given toAC_CONFIG_HEADERS
; if that macro was not called, config.status ignores this variable.
The symbolic links to establish. The default is the arguments given to
AC_CONFIG_LINKS
; if that macro was not called, config.status ignores this variable.
In config.status Invocation, using this old interface, the example would be:
config.h: stamp-h stamp-h: config.h.in config.status CONFIG_COMMANDS= CONFIG_LINKS= CONFIG_FILES= \ CONFIG_HEADERS=config.h ./config.status echo > stamp-h Makefile: Makefile.in config.status CONFIG_COMMANDS= CONFIG_LINKS= CONFIG_HEADERS= \ CONFIG_FILES=Makefile ./config.status
(If configure.ac does not call AC_CONFIG_HEADERS
, there is
no need to set CONFIG_HEADERS
in the make
rules. Equally
for CONFIG_COMMANDS
, etc.)
In order to produce config.h.in, autoheader needs to
build or to find templates for each symbol. Modern releases of Autoconf
use AH_VERBATIM
and AH_TEMPLATE
(see Autoheader Macros), but in older releases a file, acconfig.h, contained the
list of needed templates. autoheader copied comments and
#define
and #undef
statements from acconfig.h in
the current directory, if present. This file used to be mandatory if
you AC_DEFINE
any additional symbols.
Modern releases of Autoconf also provide AH_TOP
and
AH_BOTTOM
if you need to prepend/append some information to
config.h.in. Ancient versions of Autoconf had a similar feature:
if ./acconfig.h contains the string ‘@TOP@’,
autoheader copies the lines before the line containing
‘@TOP@’ into the top of the file that it generates. Similarly,
if ./acconfig.h contains the string ‘@BOTTOM@’,
autoheader copies the lines after that line to the end of the
file it generates. Either or both of those strings may be omitted. An
even older alternate way to produce the same effect in ancient versions
of Autoconf is to create the files file.top (typically
config.h.top) and/or file.bot in the current
directory. If they exist, autoheader copies them to the
beginning and end, respectively, of its output.
In former versions of Autoconf, the files used in preparing a software package for distribution were:
configure.ac --. .------> autoconf* -----> configure +---+ [aclocal.m4] --+ `---. [acsite.m4] ---' | +--> [autoheader*] -> [config.h.in] [acconfig.h] ----. | +-----' [config.h.top] --+ [config.h.bot] --'
Using only the AH_
macros, configure.ac should be
self-contained, and should not depend upon acconfig.h etc.
The autoupdate program updates a configure.ac file that calls Autoconf macros by their old names to use the current macro names. In version 2 of Autoconf, most of the macros were renamed to use a more uniform and descriptive naming scheme. See Macro Names, for a description of the new scheme. Although the old names still work (see Obsolete Macros, for a list of the old macros and the corresponding new names), you can make your configure.ac files more readable and make it easier to use the current Autoconf documentation if you update them to use the new macro names.
If given no arguments, autoupdate updates configure.ac,
backing up the original version with the suffix ~ (or the value
of the environment variable SIMPLE_BACKUP_SUFFIX
, if that is
set). If you give autoupdate an argument, it reads that file
instead of configure.ac and writes the updated file to the
standard output.
autoupdate accepts the following options:
Several macros are obsoleted in Autoconf, for various reasons (typically they failed to quote properly, couldn't be extended for more recent issues, etc.). They are still supported, but deprecated: their use should be avoided.
During the jump from Autoconf version 1 to version 2, most of the macros were renamed to use a more uniform and descriptive naming scheme, but their signature did not change. See Macro Names, for a description of the new naming scheme. Below, if there is just the mapping from old names to new names for these macros, the reader is invited to refer to the definition of the new macro for the signature and the description.
If the C compiler supports a working
long double
type with more range or precision than thedouble
type, defineHAVE_LONG_DOUBLE
.You should use
AC_TYPE_LONG_DOUBLE
orAC_TYPE_LONG_DOUBLE_WIDER
instead. See Particular Types.
Determine the system type and set output variables to the names of the canonical system types. See Canonicalizing, for details about the variables this macro sets.
The user is encouraged to use either
AC_CANONICAL_BUILD
, orAC_CANONICAL_HOST
, orAC_CANONICAL_TARGET
, depending on the needs. UsingAC_CANONICAL_TARGET
is enough to run the two other macros.
Autoconf, up to 2.13, used to provide this version of
AC_CHECK_TYPE
, deprecated because of its flaws. First, although it is a member of theCHECK
clan, it does more than just checking. Secondly, missing types are defined using#define
, nottypedef
, and this can lead to problems in the case of pointer types.This use of
AC_CHECK_TYPE
is obsolete and discouraged; see Generic Types, for the description of the current macro.If the type type is not defined, define it to be the C (or C++) builtin type default, e.g., ‘short int’ or ‘unsigned int’.
This macro is equivalent to:
AC_CHECK_TYPE([type], [], [AC_DEFINE_UNQUOTED([type], [default], [Define to `default' if <sys/types.h> does not define.])])In order to keep backward compatibility, the two versions of
AC_CHECK_TYPE
are implemented, selected by a simple heuristics:
- If there are three or four arguments, the modern version is used.
- If the second argument appears to be a C or C++ type, then the obsolete version is used. This happens if the argument is a C or C++ builtin type or a C identifier ending in ‘_t’, optionally followed by one of ‘[(* ’ and then by a string of zero or more characters taken from the set ‘[]()* _a-zA-Z0-9’.
- If the second argument is spelled with the alphabet of valid C and C++ types, the user is warned and the modern version is used.
- Otherwise, the modern version is used.
You are encouraged either to use a valid builtin type, or to use the equivalent modern code (see above), or better yet, to use
AC_CHECK_TYPES
together with#if !HAVE_LOFF_T typedef loff_t off_t; #endif
This is an obsolete version of
AC_TRY_COMPILE
itself replaced byAC_COMPILE_IFELSE
(see Running the Compiler), with the addition that it prints ‘checking for echo-text’ to the standard output first, if echo-text is non-empty. UseAC_MSG_CHECKING
andAC_MSG_RESULT
instead to print messages (see Printing Messages).
Check for the Cygwin environment in which case the shell variable
CYGWIN
is set to ‘yes’. Don't use this macro, the dignified means to check the nature of the host is usingAC_CANONICAL_HOST
. As a matter of fact this macro is defined as:AC_REQUIRE([AC_CANONICAL_HOST])[]dnl case $host_os in *cygwin* ) CYGWIN=yes;; * ) CYGWIN=no;; esacBeware that the variable
CYGWIN
has a special meaning when running Cygwin, and should not be changed. That's yet another reason not to use this macro.
AC_CHECK_DECLS([sys_siglist], [], [], [#include <signal.h> /* NetBSD declares sys_siglist in unistd.h. */ #if HAVE_UNISTD_H # include <unistd.h> #endif ])
Like calling
AC_FUNC_CLOSEDIR_VOID
andAC_HEADER_DIRENT
, but defines a different set of C preprocessor macros to indicate which header file is found:
Header Old Symbol New Symbol dirent.h DIRENT
HAVE_DIRENT_H
sys/ndir.h SYSNDIR
HAVE_SYS_NDIR_H
sys/dir.h SYSDIR
HAVE_SYS_DIR_H
ndir.h NDIR
HAVE_NDIR_H
If on DYNIX/ptx, add -lseq to output variable
LIBS
. This macro used to be defined asAC_CHECK_LIB([seq], [getmntent], [LIBS="-lseq $LIBS"])now it is just
AC_FUNC_GETMNTENT
.
Defined the output variable
EXEEXT
based on the output of the compiler, which is now done automatically. Typically set to empty string if Posix and ‘.exe’ if a DOS variant.
If
wait3
is found and fills in the contents of its third argument (a ‘struct rusage *’), which HP-UX does not do, defineHAVE_WAIT3
.These days portable programs should use
waitpid
, notwait3
, aswait3
has been removed from Posix.
This macro is equivalent to calling
AC_CHECK_LIB
with a function argument ofmain
. In addition, library can be written as any of ‘foo’, -lfoo, or ‘libfoo.a’. In all of those cases, the compiler is passed -lfoo. However, library cannot be a shell variable; it must be a literal name.
Formerly
AC_INIT
used to have a single argument, and was equivalent to:AC_INIT AC_CONFIG_SRCDIR(unique-file-in-source-dir)
If the C type
int
is 16 bits wide, defineINT_16_BITS
. Use ‘AC_CHECK_SIZEOF(int)’ instead.
If on irix (Silicon Graphics Unix), add -lsun to output
LIBS
. If you were using it to getgetmntent
, useAC_FUNC_GETMNTENT
instead. If you used it for the NIS versions of the password and group functions, use ‘AC_CHECK_LIB(sun, getpwnam)’. Up to Autoconf 2.13, it used to beAC_CHECK_LIB([sun], [getmntent], [LIBS="-lsun $LIBS"])now it is defined as
AC_FUNC_GETMNTENT AC_CHECK_LIB([sun], [getpwnam])
Select the language that is saved on the top of the stack, as set by
AC_LANG_SAVE
, remove it from the stack, and callAC_LANG(
language)
.
Remember the current language (as set by
AC_LANG
) on a stack. The current language does not change.AC_LANG_PUSH
is preferred.
This is an obsolete version of
AC_CONFIG_LINKS
. An updated version of:AC_LINK_FILES(config/$machine.h config/$obj_format.h, host.h object.h)is:
AC_CONFIG_LINKS([host.h:config/$machine.h object.h:config/$obj_format.h])
Define
LONG_64_BITS
if the C typelong int
is 64 bits wide. Use the generic macro ‘AC_CHECK_SIZEOF([long int])’ instead.
If the C compiler supports a working
long double
type with more range or precision than thedouble
type, defineHAVE_LONG_DOUBLE
.You should use
AC_TYPE_LONG_DOUBLE
orAC_TYPE_LONG_DOUBLE_WIDER
instead. See Particular Types.
Used to define
NEED_MEMORY_H
if themem
functions were defined in memory.h. Today it is equivalent to ‘AC_CHECK_HEADERS([memory.h])’. Adjust your code to depend uponHAVE_MEMORY_H
, notNEED_MEMORY_H
; see Standard Symbols.
Similar to
AC_CYGWIN
but checks for the MinGW compiler environment and setsMINGW32
.
Defined the output variable
OBJEXT
based on the output of the compiler, after .c files have been excluded. Typically set to ‘o’ if Posix, ‘obj’ if a DOS variant. Now the compiler checking macros handle this automatically.
Make M4 print a message to the standard error output warning that this-macro-name is obsolete, and giving the file and line number where it was called. this-macro-name should be the name of the macro that is calling
AC_OBSOLETE
. If suggestion is given, it is printed at the end of the warning message; for example, it can be a suggestion for what to use instead of this-macro-name.For instance
AC_OBSOLETE([$0], [; use AC_CHECK_HEADERS(unistd.h) instead])dnlYou are encouraged to use
AU_DEFUN
instead, since it gives better services to the user.
The use of
AC_OUTPUT
with argument is deprecated. This obsoleted interface is equivalent to:AC_CONFIG_FILES(file...) AC_CONFIG_COMMANDS([default], extra-cmds, init-cmds) AC_OUTPUT
Specify additional shell commands to run at the end of config.status, and shell commands to initialize any variables from configure. This macro may be called multiple times. It is obsolete, replaced by
AC_CONFIG_COMMANDS
.Here is an unrealistic example:
fubar=27 AC_OUTPUT_COMMANDS([echo this is extra $fubar, and so on.], [fubar=$fubar]) AC_OUTPUT_COMMANDS([echo this is another, extra, bit], [echo init bit])Aside from the fact that
AC_CONFIG_COMMANDS
requires an additional key, an important difference is thatAC_OUTPUT_COMMANDS
is quoting its arguments twice, unlikeAC_CONFIG_COMMANDS
. This means thatAC_CONFIG_COMMANDS
can safely be given macro calls as arguments:AC_CONFIG_COMMANDS(foo, [my_FOO()])Conversely, where one level of quoting was enough for literal strings with
AC_OUTPUT_COMMANDS
, you need two withAC_CONFIG_COMMANDS
. The following lines are equivalent:AC_OUTPUT_COMMANDS([echo "Square brackets: []"]) AC_CONFIG_COMMANDS([default], [[echo "Square brackets: []"]])
If on SCO Unix, add -lintl to output variable
LIBS
. This macro used to do this:AC_CHECK_LIB([intl], [strftime], [LIBS="-lintl $LIBS"])Now it just calls
AC_FUNC_STRFTIME
instead.
If the system automatically restarts a system call that is interrupted by a signal, define
HAVE_RESTARTABLE_SYSCALLS
. This macro does not check whether system calls are restarted in general—it checks whether a signal handler installed withsignal
(but notsigaction
) causes system calls to be restarted. It does not check whether system calls can be restarted when interrupted by signals that have no handler.These days portable programs should use
sigaction
withSA_RESTART
if they want restartable system calls. They should not rely onHAVE_RESTARTABLE_SYSCALLS
, since nowadays whether a system call is restartable is a dynamic issue, not a configuration-time issue.
AC_COMPILE_IFELSE( [AC_LANG_PROGRAM([[includes]], [[function-body]])], [action-if-true], [action-if-false])See Running the Compiler.
This macro double quotes both includes and function-body.
For C and C++, includes is any
#include
statements needed by the code in function-body (includes is ignored if the currently selected language is Fortran or Fortran 77). The compiler and compilation flags are determined by the current language (see Language Choice).
AC_PREPROC_IFELSE( [AC_LANG_SOURCE([[input]])], [action-if-true], [action-if-false])This macro double quotes the input.
AC_LINK_IFELSE( [AC_LANG_PROGRAM([[includes]], [[function-body]])], [action-if-true], [action-if-false])See Running the Compiler.
This macro double quotes both includes and function-body.
Depending on the current language (see Language Choice), create a test program to see whether a function whose body consists of function-body can be compiled and linked. If the file compiles and links successfully, run shell commands action-if-found, otherwise run action-if-not-found.
This macro double quotes both includes and function-body.
For C and C++, includes is any
#include
statements needed by the code in function-body (includes is ignored if the currently selected language is Fortran or Fortran 77). The compiler and compilation flags are determined by the current language (see Language Choice), and in additionLDFLAGS
andLIBS
are used for linking.
This macro is equivalent to ‘AC_LINK_IFELSE([AC_LANG_CALL([], [function])], [action-if-found], [action-if-not-found])’.
AC_RUN_IFELSE( [AC_LANG_SOURCE([[program]])], [action-if-true], [action-if-false], [action-if-cross-compiling])See Runtime.
Define
USG
if the BSD string functions are defined in strings.h. You should no longer depend uponUSG
, but onHAVE_STRING_H
; see Standard Symbols.
If the cache file is inconsistent with the current host, target and build system types, it used to execute cmd or print a default error message. This is now handled by default.
This macro used to add -lx to output variable
LIBS
if on Xenix. Also, if dirent.h is being checked for, added -ldir toLIBS
. Now it is merely an alias ofAC_HEADER_DIRENT
instead, plus some code to detect whether running xenix on which you should not depend:AC_MSG_CHECKING([for Xenix]) AC_EGREP_CPP([yes], [#if defined M_XENIX && !defined M_UNIX yes #endif], [AC_MSG_RESULT([yes]); XENIX=yes], [AC_MSG_RESULT([no]); XENIX=])
Autoconf version 2 is mostly backward compatible with version 1. However, it introduces better ways to do some things, and doesn't support some of the ugly things in version 1. So, depending on how sophisticated your configure.ac files are, you might have to do some manual work in order to upgrade to version 2. This chapter points out some problems to watch for when upgrading. Also, perhaps your configure scripts could benefit from some of the new features in version 2; the changes are summarized in the file NEWS in the Autoconf distribution.
If you have an aclocal.m4 installed with Autoconf (as opposed to in a particular package's source directory), you must rename it to acsite.m4. See autoconf Invocation.
If you distribute install.sh with your package, rename it to
install-sh so make
builtin rules don't inadvertently
create a file called install from it. AC_PROG_INSTALL
looks for the script under both names, but it is best to use the new name.
If you were using config.h.top, config.h.bot, or
acconfig.h, you still can, but you have less clutter if you
use the AH_
macros. See Autoheader Macros.
Add ‘@CFLAGS@’, ‘@CPPFLAGS@’, and ‘@LDFLAGS@’ in your Makefile.in files, so they can take advantage of the values of those variables in the environment when configure is run. Doing this isn't necessary, but it's a convenience for users.
Also add ‘@configure_input@’ in a comment to each input file for
AC_OUTPUT
, so that the output files contain a comment saying
they were produced by configure. Automatically selecting the
right comment syntax for all the kinds of files that people call
AC_OUTPUT
on became too much work.
Add config.log and config.cache to the list of files you
remove in distclean
targets.
If you have the following in Makefile.in:
prefix = /usr/local exec_prefix = $(prefix)
you must change it to:
prefix = @prefix@ exec_prefix = @exec_prefix@
The old behavior of replacing those variables without ‘@’ characters around them has been removed.
Many of the macros were renamed in Autoconf version 2. You can still use the old names, but the new ones are clearer, and it's easier to find the documentation for them. See Obsolete Macros, for a table showing the new names for the old macros. Use the autoupdate program to convert your configure.ac to using the new macro names. See autoupdate Invocation.
Some macros have been superseded by similar ones that do the job better,
but are not call-compatible. If you get warnings about calling obsolete
macros while running autoconf, you may safely ignore them, but
your configure script generally works better if you follow
the advice that is printed about what to replace the obsolete macros with. In
particular, the mechanism for reporting the results of tests has
changed. If you were using echo or AC_VERBOSE
(perhaps
via AC_COMPILE_CHECK
), your configure script's output
looks better if you switch to AC_MSG_CHECKING
and
AC_MSG_RESULT
. See Printing Messages. Those macros work best
in conjunction with cache variables. See Caching Results.
If you were checking the results of previous tests by examining the
shell variable DEFS
, you need to switch to checking the values of
the cache variables for those tests. DEFS
no longer exists while
configure is running; it is only created when generating output
files. This difference from version 1 is because properly quoting the
contents of that variable turned out to be too cumbersome and
inefficient to do every time AC_DEFINE
is called. See Cache Variable Names.
For example, here is a configure.ac fragment written for Autoconf version 1:
AC_HAVE_FUNCS(syslog) case "$DEFS" in *-DHAVE_SYSLOG*) ;; *) # syslog is not in the default libraries. See if it's in some other. saved_LIBS="$LIBS" for lib in bsd socket inet; do AC_CHECKING(for syslog in -l$lib) LIBS="-l$lib $saved_LIBS" AC_HAVE_FUNCS(syslog) case "$DEFS" in *-DHAVE_SYSLOG*) break ;; *) ;; esac LIBS="$saved_LIBS" done ;; esac
Here is a way to write it for version 2:
AC_CHECK_FUNCS([syslog]) if test $ac_cv_func_syslog = no; then # syslog is not in the default libraries. See if it's in some other. for lib in bsd socket inet; do AC_CHECK_LIB([$lib], [syslog], [AC_DEFINE([HAVE_SYSLOG]) LIBS="-l$lib $LIBS"; break]) done fi
If you were working around bugs in AC_DEFINE_UNQUOTED
by adding
backslashes before quotes, you need to remove them. It now works
predictably, and does not treat quotes (except back quotes) specially.
See Setting Output Variables.
All of the Boolean shell variables set by Autoconf macros now use ‘yes’ for the true value. Most of them use ‘no’ for false, though for backward compatibility some use the empty string instead. If you were relying on a shell variable being set to something like 1 or ‘t’ for true, you need to change your tests.
When defining your own macros, you should now use AC_DEFUN
instead of define
. AC_DEFUN
automatically calls
AC_PROVIDE
and ensures that macros called via AC_REQUIRE
do not interrupt other macros, to prevent nested ‘checking...’
messages on the screen. There's no actual harm in continuing to use the
older way, but it's less convenient and attractive. See Macro Definitions.
You probably looked at the macros that came with Autoconf as a guide for how to do things. It would be a good idea to take a look at the new versions of them, as the style is somewhat improved and they take advantage of some new features.
If you were doing tricky things with undocumented Autoconf internals (macros, variables, diversions), check whether you need to change anything to account for changes that have been made. Perhaps you can even use an officially supported technique in version 2 instead of kludging. Or perhaps not.
To speed up your locally written feature tests, add caching to them. See whether any of your tests are of general enough usefulness to encapsulate them into macros that you can share.
The introduction of the previous section (see Autoconf 1) perfectly suits this section...
Autoconf version 2.50 is mostly backward compatible with version 2.13. However, it introduces better ways to do some things, and doesn't support some of the ugly things in version 2.13. So, depending on how sophisticated your configure.ac files are, you might have to do some manual work in order to upgrade to version 2.50. This chapter points out some problems to watch for when upgrading. Also, perhaps your configure scripts could benefit from some of the new features in version 2.50; the changes are summarized in the file NEWS in the Autoconf distribution.
The most important changes are invisible to you: the implementation of most macros have completely changed. This allowed more factorization of the code, better error messages, a higher uniformity of the user's interface etc. Unfortunately, as a side effect, some construct which used to (miraculously) work might break starting with Autoconf 2.50. The most common culprit is bad quotation.
For instance, in the following example, the message is not properly quoted:
AC_INIT AC_CHECK_HEADERS(foo.h, , AC_MSG_ERROR(cannot find foo.h, bailing out)) AC_OUTPUT
Autoconf 2.13 simply ignores it:
$ autoconf-2.13; ./configure --silent creating cache ./config.cache configure: error: cannot find foo.h $
while Autoconf 2.50 produces a broken configure:
$ autoconf-2.50; ./configure --silent configure: error: cannot find foo.h ./configure: exit: bad non-numeric arg `bailing' ./configure: exit: bad non-numeric arg `bailing' $
The message needs to be quoted, and the AC_MSG_ERROR
invocation
too!
AC_INIT([Example], [1.0], [bug-example@example.org]) AC_CHECK_HEADERS([foo.h], [], [AC_MSG_ERROR([cannot find foo.h, bailing out])]) AC_OUTPUT
Many many (and many more) Autoconf macros were lacking proper quotation,
including no less than... AC_DEFUN
itself!
$ cat configure.in AC_DEFUN([AC_PROG_INSTALL], [# My own much better version ]) AC_INIT AC_PROG_INSTALL AC_OUTPUT $ autoconf-2.13 autoconf: Undefined macros: ***BUG in Autoconf--please report*** AC_FD_MSG ***BUG in Autoconf--please report*** AC_EPI configure.in:1:AC_DEFUN([AC_PROG_INSTALL], configure.in:5:AC_PROG_INSTALL $ autoconf-2.50 $
While Autoconf was relatively dormant in the late 1990s, Automake
provided Autoconf-like macros for a while. Starting with Autoconf 2.50
in 2001, Autoconf provided
versions of these macros, integrated in the AC_
namespace,
instead of AM_
. But in order to ease the upgrading via
autoupdate, bindings to such AM_
macros are provided.
Unfortunately older versions of Automake (e.g., Automake 1.4)
did not quote the names of these macros.
Therefore, when m4 finds something like
‘AC_DEFUN(AM_TYPE_PTRDIFF_T, ...)’ in aclocal.m4,
AM_TYPE_PTRDIFF_T
is
expanded, replaced with its Autoconf definition.
Fortunately Autoconf catches pre-AC_INIT
expansions, and
complains, in its own words:
$ cat configure.ac AC_INIT([Example], [1.0], [bug-example@example.org]) AM_TYPE_PTRDIFF_T $ aclocal-1.4 $ autoconf aclocal.m4:17: error: m4_defn: undefined macro: _m4_divert_diversion aclocal.m4:17: the top level autom4te: m4 failed with exit status: 1 $
Modern versions of Automake no longer define most of these macros, and properly quote the names of the remaining macros. If you must use an old Automake, do not depend upon macros from Automake as it is simply not its job to provide macros (but the one it requires itself):
$ cat configure.ac AC_INIT([Example], [1.0], [bug-example@example.org]) AM_TYPE_PTRDIFF_T $ rm aclocal.m4 $ autoupdate autoupdate: `configure.ac' is updated $ cat configure.ac AC_INIT([Example], [1.0], [bug-example@example.org]) AC_CHECK_TYPES([ptrdiff_t]) $ aclocal-1.4 $ autoconf $
Based on the experience of compiler writers, and after long public debates, many aspects of the cross-compilation chain have changed:
The relationship between build, host, and target have been cleaned up: the chain of default is now simply: target defaults to host, host to build, and build to the result of config.guess. Nevertheless, in order to ease the transition from 2.13 to 2.50, the following transition scheme is implemented. Do not rely on it, as it will be completely disabled in a couple of releases (we cannot keep it, as it proves to cause more problems than it cures).
They all default to the result of running config.guess, unless you specify either --build or --host. In this case, the default becomes the system type you specified. If you specify both, and they're different, configure enters cross compilation mode, so it doesn't run any tests that require execution.
Hint: if you mean to override the result of config.guess, prefer --build over --host. In the future, --host will not override the name of the build system type. Whenever you specify --host, be sure to specify --build too.
For backward compatibility, configure accepts a system type as an option by itself. Such an option overrides the defaults for build, host, and target system types. The following configure statement configures a cross toolchain that runs on NetBSD/alpha but generates code for GNU Hurd/sparc, which is also the build platform.
./configure --host=alpha-netbsd sparc-gnu
In Autoconf 2.13 and before, the variables
build
, host
,
and target
had a different semantics before and after the
invocation of AC_CANONICAL_BUILD
etc. Now, the argument of
--build is strictly copied into build_alias
, and is left
empty otherwise. After the AC_CANONICAL_BUILD
, build
is
set to the canonicalized build type. To ease the transition, before,
its contents is the same as that of build_alias
. Do not
rely on this broken feature.
For consistency with the backward compatibility scheme exposed above, when --host is specified but --build isn't, the build system is assumed to be the same as --host, and ‘build_alias’ is set to that value. Eventually, this historically incorrect behavior will go away.
The former scheme to enable cross-compilation proved to cause more harm than good, in particular, it used to be triggered too easily, leaving regular end users puzzled in front of cryptic error messages. configure could even enter cross-compilation mode only because the compiler was not functional. This is mainly because configure used to try to detect cross-compilation, instead of waiting for an explicit flag from the user.
Now, configure enters cross-compilation mode if and only if --host is passed.
That's the short documentation. To ease the transition between 2.13 and its successors, a more complicated scheme is implemented. Do not rely on the following, as it will be removed in the near future.
If you specify --host, but not --build, when configure performs the first compiler test it tries to run an executable produced by the compiler. If the execution fails, it enters cross-compilation mode. This is fragile. Moreover, by the time the compiler test is performed, it may be too late to modify the build-system type: other tests may have already been performed. Therefore, whenever you specify --host, be sure to specify --build too.
./configure --build=i686-pc-linux-gnu --host=m68k-coff
enters cross-compilation mode. The former interface, which consisted in setting the compiler to a cross-compiler without informing configure is obsolete. For instance, configure fails if it can't run the code generated by the specified compiler if you configure as follows:
./configure CC=m68k-coff-gcc
AC_LIBOBJ
vs. LIBOBJS
Up to Autoconf 2.13, the replacement of functions was triggered via the
variable LIBOBJS
. Since Autoconf 2.50, the macro
AC_LIBOBJ
should be used instead (see Generic Functions).
Starting at Autoconf 2.53, the use of LIBOBJS
is an error.
This change is mandated by the unification of the GNU Build System
components. In particular, the various fragile techniques used to parse
a configure.ac are all replaced with the use of traces. As a
consequence, any action must be traceable, which obsoletes critical
variable assignments. Fortunately, LIBOBJS
was the only problem,
and it can even be handled gracefully (read, “without your having to
change something”).
There were two typical uses of LIBOBJS
: asking for a replacement
function, and adjusting LIBOBJS
for Automake and/or Libtool.
As for function replacement, the fix is immediate: use
AC_LIBOBJ
. For instance:
LIBOBJS="$LIBOBJS fnmatch.o" LIBOBJS="$LIBOBJS malloc.$ac_objext"
should be replaced with:
AC_LIBOBJ([fnmatch]) AC_LIBOBJ([malloc])
When used with Automake 1.10 or newer, a suitable value for
LIBOBJDIR
is set so that the LIBOBJS
and LTLIBOBJS
can be referenced from any Makefile.am. Even without Automake,
arranging for LIBOBJDIR
to be set correctly enables
referencing LIBOBJS
and LTLIBOBJS
in another directory.
The LIBOJBDIR
feature is experimental.
AC_FOO_IFELSE
vs. AC_TRY_FOO
Since Autoconf 2.50, internal codes uses AC_PREPROC_IFELSE
,
AC_COMPILE_IFELSE
, AC_LINK_IFELSE
, and
AC_RUN_IFELSE
on one hand and AC_LANG_SOURCES
,
and AC_LANG_PROGRAM
on the other hand instead of the deprecated
AC_TRY_CPP
, AC_TRY_COMPILE
, AC_TRY_LINK
, and
AC_TRY_RUN
. The motivations where:
AC_TRY_COMPILE
etc. were double
quoting their arguments;
In addition to the change of syntax, the philosophy has changed too: while emphasis was put on speed at the expense of accuracy, today's Autoconf promotes accuracy of the testing framework at, ahem..., the expense of speed.
As a perfect example of what is not to be done, here is how to
find out whether a header file contains a particular declaration, such
as a typedef, a structure, a structure member, or a function. Use
AC_EGREP_HEADER
instead of running grep
directly on the
header file; on some systems the symbol might be defined in another
header file that the file you are checking includes.
As a (bad) example, here is how you should not check for C preprocessor
symbols, either defined by header files or predefined by the C
preprocessor: using AC_EGREP_CPP
:
AC_EGREP_CPP(yes, [#ifdef _AIX yes #endif ], is_aix=yes, is_aix=no)
The above example, properly written would (i) use
AC_LANG_PROGRAM
, and (ii) run the compiler:
AC_COMPILE_IFELSE([AC_LANG_PROGRAM( [[#if !defined _AIX error: This isn't AIX! #endif ]])], [is_aix=yes], [is_aix=no])
N.B.: This section describes an experimental feature which will be part of Autoconf in a forthcoming release. Although we believe Autotest is stabilizing, this documentation describes an interface which might change in the future: do not depend upon Autotest without subscribing to the Autoconf mailing lists.
It is paradoxical that portable projects depend on nonportable tools to run their test suite. Autoconf by itself is the paragon of this problem: although it aims at perfectly portability, up to 2.13 its test suite was using DejaGNU, a rich and complex testing framework, but which is far from being standard on Posix systems. Worse yet, it was likely to be missing on the most fragile platforms, the very platforms that are most likely to torture Autoconf and exhibit deficiencies.
To circumvent this problem, many package maintainers have developed their own testing framework, based on simple shell scripts whose sole outputs are exit status values describing whether the test succeeded. Most of these tests share common patterns, and this can result in lots of duplicated code and tedious maintenance.
Following exactly the same reasoning that yielded to the inception of Autoconf, Autotest provides a test suite generation framework, based on M4 macros building a portable shell script. The suite itself is equipped with automatic logging and tracing facilities which greatly diminish the interaction with bug reporters, and simple timing reports.
Autoconf itself has been using Autotest for years, and we do attest that it has considerably improved the strength of the test suite and the quality of bug reports. Other projects are known to use some generation of Autotest, such as Bison, Free Recode, Free Wdiff, GNU Tar, each of them with different needs, and this usage has validated Autotest as a general testing framework.
Nonetheless, compared to DejaGNU, Autotest is inadequate for interactive tool testing, which is probably its main limitation.
Generating testing or validation suites using Autotest is rather easy. The whole validation suite is held in a file to be processed through autom4te, itself using GNU M4 under the scene, to produce a stand-alone Bourne shell script which then gets distributed. Neither autom4te nor GNU M4 are needed at the installer's end.
Each test of the validation suite should be part of some test group. A test group is a sequence of interwoven tests that ought to be executed together, usually because one test in the group creates data files than a later test in the same group needs to read. Complex test groups make later debugging more tedious. It is much better to keep only a few tests per test group. Ideally there is only one test per test group.
For all but the simplest packages, some file such as testsuite.at does not fully hold all test sources, as these are often easier to maintain in separate files. Each of these separate files holds a single test group, or a sequence of test groups all addressing some common functionality in the package. In such cases, testsuite.at merely initializes the validation suite, and sometimes does elementary health checking, before listing include statements for all other test files. The special file package.m4, containing the identification of the package, is automatically included if found.
A convenient alternative consists in moving all the global issues
(local Autotest macros, elementary health checking, and AT_INIT
invocation) into the file local.at
, and making
testsuite.at be a simple list of m4_include
of sub test
suites. In such case, generating the whole test suite or pieces of it
is only a matter of choosing the autom4te command line
arguments.
The validation scripts that Autotest produces are by convention called testsuite. When run, testsuite executes each test group in turn, producing only one summary line per test to say if that particular test succeeded or failed. At end of all tests, summarizing counters get printed. One debugging directory is left for each test group which failed, if any: such directories are named testsuite.dir/nn, where nn is the sequence number of the test group, and they include:
AT_DATA
In the ideal situation, none of the tests fail, and consequently no debugging directory is left behind for validation.
It often happens in practice that individual tests in the validation
suite need to get information coming out of the configuration process.
Some of this information, common for all validation suites, is provided
through the file atconfig, automatically created by
AC_CONFIG_TESTDIR
. For configuration informations which your
testing environment specifically needs, you might prepare an optional
file named atlocal.in, instantiated by AC_CONFIG_FILES
.
The configuration process produces atconfig and atlocal
out of these two input files, and these two produced files are
automatically read by the testsuite script.
Here is a diagram showing the relationship between files.
Files used in preparing a software package for distribution:
[package.m4] -->. \ subfile-1.at ->. [local.at] ---->+ ... \ \ subfile-i.at ---->-- testsuite.at -->-- autom4te* -->testsuite ... / subfile-n.at ->'
Files used in configuring a software package:
.--> atconfig / [atlocal.in] --> config.status* --< \ `--> [atlocal]
Files created during the test suite execution:
atconfig -->. .--> testsuite.log \ / >-- testsuite* --< / \ [atlocal] ->' `--> [testsuite.dir]
When run, the test suite creates a log file named after itself, e.g., a test suite named testsuite creates testsuite.log. It contains a lot of information, usually more than maintainers actually need, but therefore most of the time it contains all that is needed:
CC
for subsequent runs.
Autoconf faced exactly the same problem, and solved it by asking
users to pass the variable definitions as command line arguments.
Autotest requires this rule, too, but has no means to enforce it; the log
then contains a trace of the variables that were changed by the user.
AT_TESTED
).
The testsuite.at is a Bourne shell script making use of special
Autotest M4 macros. It often contains a call to AT_INIT
near
its beginning followed by one call to m4_include
per source file
for tests. Each such included file, or the remainder of
testsuite.at if include files are not used, contain a sequence of
test groups. Each test group begins with a call to AT_SETUP
,
then an arbitrary number of shell commands or calls to AT_CHECK
,
and then completes with a call to AT_CLEANUP
.
Initialize Autotest. Giving a name to the test suite is encouraged if your package includes several test suites. In any case, the test suite always displays the package name and version. It also inherits the package bug report address.
State that, in addition to the Free Software Foundation's copyright on the Autotest macros, parts of your test suite are covered by copyright-notice.
The copyright-notice shows up in both the head of testsuite and in ‘testsuite --version’.
Log the file name and answer to --version of each program in space-separated list executables. Several invocations register new executables, in other words, don't fear registering one program several times.
Autotest test suites rely on PATH to find the tested program. This avoids the need to generate absolute names of the various tools, and makes it possible to test installed programs. Therefore, knowing which programs are being exercised is crucial to understanding problems in the test suite itself, or its occasional misuses. It is a good idea to also subscribe foreign programs you depend upon, to avoid incompatible diagnostics.
This macro starts a group of related tests, all to be executed in the same subshell. It accepts a single argument, which holds a few words (no more than about 30 or 40 characters) quickly describing the purpose of the test group being started.
Associate the space-separated list of keywords to the enclosing test group. This makes it possible to run “slices” of the test suite. For instance, if some of your test groups exercise some ‘foo’ feature, then using ‘AT_KEYWORDS(foo)’ lets you run ‘./testsuite -k foo’ to run exclusively these test groups. The title of the test group is automatically recorded to
AT_KEYWORDS
.Several invocations within a test group accumulate new keywords. In other words, don't fear registering the same keyword several times in a test group.
If the current test group fails, log the contents of file. Several identical calls within one test group have no additional effect.
Determine whether the test is expected to fail because it is a known bug (for unsupported features, you should skip the test). shell-condition is a shell expression such as a
test
command; you can instantiate this macro many times from within the same test group, and one of the conditions is enough to turn the test into an expected failure.
Initialize an input data file with given contents. Of course, the contents have to be properly quoted between square brackets to protect against included commas or spurious M4 expansion. The contents ought to end with an end of line.
Execute a test by performing given shell commands. These commands should normally exit with status, while producing expected stdout and stderr contents. If commands exit with status 77, then the whole test group is skipped. Otherwise, if this test fails, run shell commands run-if-fail or, if this test passes, run shell commands run-if-pass.
The commands must not redirect the standard output, nor the standard error.
If status, or stdout, or stderr is ‘ignore’, then the corresponding value is not checked.
The special value ‘expout’ for stdout means the expected output of the commands is the content of the file expout. If stdout is ‘stdout’, then the standard output of the commands is available for further tests in the file stdout. Similarly for stderr with ‘expout’ and ‘stderr’.
Autotest test suites support the following arguments:
clean
Make targets.
By default all tests are performed (or described with --list) in the default environment first silently, then verbosely, but the environment, set of tests, and verbosity level can be tuned:
The variable AUTOTEST_PATH
specifies the testing path to prepend
to PATH. Relative directory names (not starting with
‘/’) are considered to be relative to the top level of the
package being built. All directories are made absolute, first
starting from the top level build tree, then from the
source tree. For instance ‘./testsuite
AUTOTEST_PATH=tests:bin’ for a /src/foo-1.0 source package built
in /tmp/foo results in ‘/tmp/foo/tests:/tmp/foo/bin’ and
then ‘/src/foo-1.0/tests:/src/foo-1.0/bin’ being prepended to
PATH.
AT_SETUP
or AT_KEYWORDS
) that match all keywords
of the comma separated list keywords, case-insensitively. Use
‘!’ immediately before the keyword to invert the selection for this
keyword. By default, the keywords match whole words; enclose them in
‘.*’ to also match parts of words.
For example, running
./testsuite -k 'autoupdate,.*FUNC.*'
selects all tests tagged ‘autoupdate’ and with tags containing ‘FUNC’ (as in ‘AC_CHECK_FUNC’, ‘AC_FUNC_FNMATCH’, etc.), while
./testsuite -k '!autoupdate' -k '.*FUNC.*'
selects all tests not tagged ‘autoupdate’ or with tags
containing ‘FUNC’.
For putting Autotest into movement, you need some configuration and makefile machinery. We recommend, at least if your package uses deep or shallow hierarchies, that you use tests/ as the name of the directory holding all your tests and their makefile. Here is a check list of things to do.
AT_PACKAGE_STRING
, the
full signature of the package, and AT_PACKAGE_BUGREPORT
, the
address to which bug reports should be sent. For sake of completeness,
we suggest that you also define AT_PACKAGE_NAME
,
AT_PACKAGE_TARNAME
, and AT_PACKAGE_VERSION
.
See Initializing configure, for a description of these variables. We
suggest the following makefile excerpt:
$(srcdir)/package.m4: $(top_srcdir)/configure.ac { \ echo '# Signature of the current package.'; \ echo 'm4_define([AT_PACKAGE_NAME], [@PACKAGE_NAME@])'; \ echo 'm4_define([AT_PACKAGE_TARNAME], [@PACKAGE_TARNAME@])'; \ echo 'm4_define([AT_PACKAGE_VERSION], [@PACKAGE_VERSION@])'; \ echo 'm4_define([AT_PACKAGE_STRING], [@PACKAGE_STRING@])'; \ echo 'm4_define([AT_PACKAGE_BUGREPORT], [@PACKAGE_BUGREPORT@])'; \ } >'$(srcdir)/package.m4'
Be sure to distribute package.m4 and to put it into the source hierarchy: the test suite ought to be shipped!
AC_CONFIG_TESTDIR
.
An Autotest test suite is to be configured in directory. This macro requires the instantiation of directory/atconfig from directory/atconfig.in, and sets the default
AUTOTEST_PATH
to test-path (see testsuite Invocation).
AC_CONFIG_FILES
command includes substitution for
tests/atlocal.
With Automake, here is a minimal example about how to link ‘make check’ with a validation suite.
EXTRA_DIST = testsuite.at $(TESTSUITE) atlocal.in TESTSUITE = $(srcdir)/testsuite check-local: atconfig atlocal $(TESTSUITE) $(SHELL) '$(TESTSUITE)' $(TESTSUITEFLAGS) installcheck-local: atconfig atlocal $(TESTSUITE) $(SHELL) '$(TESTSUITE)' AUTOTEST_PATH='$(bindir)' \ $(TESTSUITEFLAGS) clean-local: test ! -f '$(TESTSUITE)' || \ $(SHELL) '$(TESTSUITE)' --clean AUTOTEST = $(AUTOM4TE) --language=autotest $(TESTSUITE): $(srcdir)/testsuite.at $(AUTOTEST) -I '$(srcdir)' -o $@.tmp $@.at mv $@.tmp $@
You might want to list explicitly the dependencies, i.e., the list of the files testsuite.at includes.
With strict Autoconf, you might need to add lines inspired from the following:
subdir = tests atconfig: $(top_builddir)/config.status cd $(top_builddir) && \ $(SHELL) ./config.status $(subdir)/$@ atlocal: $(srcdir)/atlocal.in $(top_builddir)/config.status cd $(top_builddir) && \ $(SHELL) ./config.status $(subdir)/$@
and manage to have atconfig.in and $(EXTRA_DIST)
distributed.
With all this in place, and if you have not initialized ‘TESTSUITEFLAGS’ within your makefile, you can fine-tune test suite execution with this variable, for example:
make check TESTSUITEFLAGS='-v -d -x 75 -k AC_PROG_CC CFLAGS=-g'
Several questions about Autoconf come up occasionally. Here some of them are addressed.
What are the restrictions on distributing configure
scripts that Autoconf generates? How does that affect my
programs that use them?
There are no restrictions on how the configuration scripts that Autoconf produces may be distributed or used. In Autoconf version 1, they were covered by the GNU General Public License. We still encourage software authors to distribute their work under terms like those of the GPL, but doing so is not required to use Autoconf.
Of the other files that might be used with configure, config.h.in is under whatever copyright you use for your configure.ac. config.sub and config.guess have an exception to the GPL when they are used with an Autoconf-generated configure script, which permits you to distribute them under the same terms as the rest of your package. install-sh is from the X Consortium and is not copyrighted.
Why does Autoconf require GNU M4?
Many M4 implementations have hard-coded limitations on the size and number of macros that Autoconf exceeds. They also lack several builtin macros that it would be difficult to get along without in a sophisticated application like Autoconf, including:
m4_builtin m4_indir m4_bpatsubst __file__ __line__
Autoconf requires version 1.4.4 or later of GNU M4.
Since only software maintainers need to use Autoconf, and since GNU M4 is simple to configure and install, it seems reasonable to require GNU M4 to be installed also. Many maintainers of GNU and other free software already have most of the GNU utilities installed, since they prefer them.
If Autoconf requires GNU M4 and GNU M4 has an Autoconf
configure script, how do I bootstrap? It seems like a chicken
and egg problem!
This is a misunderstanding. Although GNU M4 does come with a configure script produced by Autoconf, Autoconf is not required in order to run the script and install GNU M4. Autoconf is only required if you want to change the M4 configure script, which few people have to do (mainly its maintainer).
Why not use Imake instead of configure scripts?
Several people have written addressing this question, so I include adaptations of their explanations here.
The following answer is based on one written by Richard Pixley:
Autoconf generated scripts frequently work on machines that it has never been set up to handle before. That is, it does a good job of inferring a configuration for a new system. Imake cannot do this.Imake uses a common database of host specific data. For X11, this makes sense because the distribution is made as a collection of tools, by one central authority who has control over the database.
GNU tools are not released this way. Each GNU tool has a maintainer; these maintainers are scattered across the world. Using a common database would be a maintenance nightmare. Autoconf may appear to be this kind of database, but in fact it is not. Instead of listing host dependencies, it lists program requirements.
If you view the GNU suite as a collection of native tools, then the problems are similar. But the GNU development tools can be configured as cross tools in almost any host+target permutation. All of these configurations can be installed concurrently. They can even be configured to share host independent files across hosts. Imake doesn't address these issues.
Imake templates are a form of standardization. The GNU coding standards address the same issues without necessarily imposing the same restrictions.
Here is some further explanation, written by Per Bothner:
One of the advantages of Imake is that it easy to generate large makefiles using the ‘#include’ and macro mechanisms of cpp. However,cpp
is not programmable: it has limited conditional facilities, and no looping. Andcpp
cannot inspect its environment.All of these problems are solved by using
sh
instead ofcpp
. The shell is fully programmable, has macro substitution, can execute (or source) other shell scripts, and can inspect its environment.
Paul Eggert elaborates more:
With Autoconf, installers need not assume that Imake itself is already installed and working well. This may not seem like much of an advantage to people who are accustomed to Imake. But on many hosts Imake is not installed or the default installation is not working well, and requiring Imake to install a package hinders the acceptance of that package on those hosts. For example, the Imake template and configuration files might not be installed properly on a host, or the Imake build procedure might wrongly assume that all source files are in one big directory tree, or the Imake configuration might assume one compiler whereas the package or the installer needs to use another, or there might be a version mismatch between the Imake expected by the package and the Imake supported by the host. These problems are much rarer with Autoconf, where each package comes with its own independent configuration processor.Also, Imake often suffers from unexpected interactions between make and the installer's C preprocessor. The fundamental problem here is that the C preprocessor was designed to preprocess C programs, not makefiles. This is much less of a problem with Autoconf, which uses the general-purpose preprocessor M4, and where the package's author (rather than the installer) does the preprocessing in a standard way.
Finally, Mark Eichin notes:
Imake isn't all that extensible, either. In order to add new features to Imake, you need to provide your own project template, and duplicate most of the features of the existing one. This means that for a sophisticated project, using the vendor-provided Imake templates fails to provide any leverage—since they don't cover anything that your own project needs (unless it is an X11 program).On the other side, though:
The one advantage that Imake has over configure: Imakefile files tend to be much shorter (likewise, less redundant) than Makefile.in files. There is a fix to this, however—at least for the Kerberos V5 tree, we've modified things to call in common post.in and pre.in makefile fragments for the entire tree. This means that a lot of common things don't have to be duplicated, even though they normally are in configure setups.
#define
Installation Directories?My program needs library files, installed indatadir
and similar. If I useAC_DEFINE_UNQUOTED([DATADIR], [$datadir], [Define to the read-only architecture-independent data directory.])I get
#define DATADIR "${prefix}/share"
As already explained, this behavior is on purpose, mandated by the GNU Coding Standards, see Installation Directory Variables. There are several means to achieve a similar goal:
AC_DEFINE
but use your makefile to pass the
actual value of datadir
via compilation flags.
See Installation Directory Variables, for the details.
CPPFLAGS
:
CPPFLAGS = -DDATADIR='"$(datadir)"' @CPPFLAGS@
or create a dedicated header file:
DISTCLEANFILES = datadir.h datadir.h: Makefile echo '#define DATADIR "$(datadir)"' >$@
AC_DEFINE
but have configure compute the literal
value of datadir
and others. Many people have wrapped macros to
automate this task. For instance, the macro AC_DEFINE_DIR
from
the Autoconf Macro Archive.
This solution does not conform to the GNU Coding Standards.
prefix
, and try to
find prefix
at runtime, this way your package is relocatable.
Some macros are already available to address this issue: see
adl_COMPUTE_RELATIVE_PATHS
and
adl_COMPUTE_STANDARD_RELATIVE_PATHS
on the
Autoconf Macro Archive.
What is this directory autom4te.cache? Can I safely remove it?
In the GNU Build System, configure.ac plays a central role and is read by many tools: autoconf to create configure, autoheader to create config.h.in, automake to create Makefile.in, autoscan to check the completeness of configure.ac, autoreconf to check the GNU Build System components that are used. To “read configure.ac” actually means to compile it with M4, which can be a long process for complex configure.ac.
This is why all these tools, instead of running directly M4, invoke autom4te (see autom4te Invocation) which, while answering to a specific demand, stores additional information in autom4te.cache for future runs. For instance, if you run autoconf, behind the scenes, autom4te also stores information for the other tools, so that when you invoke autoheader or automake etc., reprocessing configure.ac is not needed. The speed up is frequently of 30%, and is increasing with the size of configure.ac.
But it is and remains being simply a cache: you can safely remove it.
Can I permanently get rid of it?
The creation of this cache can be disabled from ~/.autom4te.cfg, see Customizing autom4te, for more details. You should be aware that disabling the cache slows down the Autoconf test suite by 40%. The more GNU Build System components are used, the more the cache is useful; for instance running ‘autoreconf -f’ on the Core Utilities is twice slower without the cache although --force implies that the cache is not fully exploited, and eight times slower than without --force.
The most important guideline to bear in mind when checking for
features is to mimic as much as possible the intended use.
Unfortunately, old versions of AC_CHECK_HEADER
and
AC_CHECK_HEADERS
failed to follow this idea, and called
the preprocessor, instead of the compiler, to check for headers. As a
result, incompatibilities between headers went unnoticed during
configuration, and maintainers finally had to deal with this issue
elsewhere.
As of Autoconf 2.56 both checks are performed, and configure
complains loudly if the compiler and the preprocessor do not agree.
For the time being the result used is that of the preprocessor, to give
maintainers time to adjust their configure.ac, but in the
future, only the compiler will be considered.
Consider the following example:
$ cat number.h typedef int number; $ cat pi.h const number pi = 3; $ cat configure.ac AC_INIT([Example], [1.0], [bug-example@example.org]) AC_CHECK_HEADERS([pi.h]) $ autoconf -Wall $ ./configure checking for gcc... gcc checking for C compiler default output file name... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ISO C89... none needed checking how to run the C preprocessor... gcc -E checking for grep that handles long lines and -e... grep checking for egrep... grep -E checking for ANSI C header files... yes checking for sys/types.h... yes checking for sys/stat.h... yes checking for stdlib.h... yes checking for string.h... yes checking for memory.h... yes checking for strings.h... yes checking for inttypes.h... yes checking for stdint.h... yes checking for unistd.h... yes checking pi.h usability... no checking pi.h presence... yes configure: WARNING: pi.h: present but cannot be compiled configure: WARNING: pi.h: check for missing prerequisite headers? configure: WARNING: pi.h: see the Autoconf documentation configure: WARNING: pi.h: section "Present But Cannot Be Compiled" configure: WARNING: pi.h: proceeding with the preprocessor's result configure: WARNING: pi.h: in the future, the compiler will take precedence configure: WARNING: ## -------------------------------------- ## configure: WARNING: ## Report this to bug-example@example.org ## configure: WARNING: ## -------------------------------------- ## checking for pi.h... yes
The proper way the handle this case is using the fourth argument (see Generic Headers):
$ cat configure.ac AC_INIT([Example], [1.0], [bug-example@example.org]) AC_CHECK_HEADERS([number.h pi.h], [], [], [[#if HAVE_NUMBER_H # include <number.h> #endif ]]) $ autoconf -Wall $ ./configure checking for gcc... gcc checking for C compiler default output... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ANSI C... none needed checking for number.h... yes checking for pi.h... yes
See Particular Headers, for a list of headers with their prerequisite.
You may be wondering, Why was Autoconf originally written? How did it get into its present form? (Why does it look like gorilla spit?) If you're not wondering, then this chapter contains no information useful to you, and you might as well skip it. If you are wondering, then let there be light...
In June 1991 I was maintaining many of the GNU utilities for the Free Software Foundation. As they were ported to more platforms and more programs were added, the number of -D options that users had to select in the makefile (around 20) became burdensome. Especially for me—I had to test each new release on a bunch of different systems. So I wrote a little shell script to guess some of the correct settings for the fileutils package, and released it as part of fileutils 2.0. That configure script worked well enough that the next month I adapted it (by hand) to create similar configure scripts for several other GNU utilities packages. Brian Berliner also adapted one of my scripts for his CVS revision control system.
Later that summer, I learned that Richard Stallman and Richard Pixley were developing similar scripts to use in the GNU compiler tools; so I adapted my configure scripts to support their evolving interface: using the file name Makefile.in as the templates; adding ‘+srcdir’, the first option (of many); and creating config.status files.
As I got feedback from users, I incorporated many improvements, using Emacs to search and replace, cut and paste, similar changes in each of the scripts. As I adapted more GNU utilities packages to use configure scripts, updating them all by hand became impractical. Rich Murphey, the maintainer of the GNU graphics utilities, sent me mail saying that the configure scripts were great, and asking if I had a tool for generating them that I could send him. No, I thought, but I should! So I started to work out how to generate them. And the journey from the slavery of hand-written configure scripts to the abundance and ease of Autoconf began.
Cygnus configure, which was being developed at around that time, is table driven; it is meant to deal mainly with a discrete number of system types with a small number of mainly unguessable features (such as details of the object file format). The automatic configuration system that Brian Fox had developed for Bash takes a similar approach. For general use, it seems to me a hopeless cause to try to maintain an up-to-date database of which features each variant of each operating system has. It's easier and more reliable to check for most features on the fly—especially on hybrid systems that people have hacked on locally or that have patches from vendors installed.
I considered using an architecture similar to that of Cygnus configure, where there is a single configure script that reads pieces of configure.in when run. But I didn't want to have to distribute all of the feature tests with every package, so I settled on having a different configure made from each configure.in by a preprocessor. That approach also offered more control and flexibility.
I looked briefly into using the Metaconfig package, by Larry Wall, Harlan Stenn, and Raphael Manfredi, but I decided not to for several reasons. The Configure scripts it produces are interactive, which I find quite inconvenient; I didn't like the ways it checked for some features (such as library functions); I didn't know that it was still being maintained, and the Configure scripts I had seen didn't work on many modern systems (such as System V R4 and NeXT); it wasn't flexible in what it could do in response to a feature's presence or absence; I found it confusing to learn; and it was too big and complex for my needs (I didn't realize then how much Autoconf would eventually have to grow).
I considered using Perl to generate my style of configure
scripts, but decided that M4 was better suited to the job of simple
textual substitutions: it gets in the way less, because output is
implicit. Plus, everyone already has it. (Initially I didn't rely on
the GNU extensions to M4.) Also, some of my friends at the
University of Maryland had recently been putting M4 front ends on
several programs, including tvtwm
, and I was interested in trying
out a new language.
Since my configure scripts determine the system's capabilities automatically, with no interactive user intervention, I decided to call the program that generates them Autoconfig. But with a version number tacked on, that name would be too long for old Unix file systems, so I shortened it to Autoconf.
In the fall of 1991 I called together a group of fellow questers after the Holy Grail of portability (er, that is, alpha testers) to give me feedback as I encapsulated pieces of my handwritten scripts in M4 macros and continued to add features and improve the techniques used in the checks. Prominent among the testers were François Pinard, who came up with the idea of making an Autoconf shell script to run M4 and check for unresolved macro calls; Richard Pixley, who suggested running the compiler instead of searching the file system to find include files and symbols, for more accurate results; Karl Berry, who got Autoconf to configure TeX and added the macro index to the documentation; and Ian Lance Taylor, who added support for creating a C header file as an alternative to putting -D options in a makefile, so he could use Autoconf for his UUCP package. The alpha testers cheerfully adjusted their files again and again as the names and calling conventions of the Autoconf macros changed from release to release. They all contributed many specific checks, great ideas, and bug fixes.
In July 1992, after months of alpha testing, I released Autoconf 1.0, and converted many GNU packages to use it. I was surprised by how positive the reaction to it was. More people started using it than I could keep track of, including people working on software that wasn't part of the GNU Project (such as TCL, FSP, and Kerberos V5). Autoconf continued to improve rapidly, as many people using the configure scripts reported problems they encountered.
Autoconf turned out to be a good torture test for M4 implementations. Unix M4 started to dump core because of the length of the macros that Autoconf defined, and several bugs showed up in GNU M4 as well. Eventually, we realized that we needed to use some features that only GNU M4 has. 4.3BSD M4, in particular, has an impoverished set of builtin macros; the System V version is better, but still doesn't provide everything we need.
More development occurred as people put Autoconf under more stresses
(and to uses I hadn't anticipated). Karl Berry added checks for X11.
david zuhn contributed C++ support. François Pinard made it diagnose
invalid arguments. Jim Blandy bravely coerced it into configuring
GNU Emacs, laying the groundwork for several later improvements.
Roland McGrath got it to configure the GNU C Library, wrote the
autoheader script to automate the creation of C header file
templates, and added a --verbose option to configure.
Noah Friedman added the --autoconf-dir option and
AC_MACRODIR
environment variable. (He also coined the term
autoconfiscate to mean “adapt a software package to use
Autoconf”.) Roland and Noah improved the quoting protection in
AC_DEFINE
and fixed many bugs, especially when I got sick of
dealing with portability problems from February through June, 1993.
A long wish list for major features had accumulated, and the effect of
several years of patching by various people had left some residual
cruft. In April 1994, while working for Cygnus Support, I began a major
revision of Autoconf. I added most of the features of the Cygnus
configure that Autoconf had lacked, largely by adapting the
relevant parts of Cygnus configure with the help of david zuhn
and Ken Raeburn. These features include support for using
config.sub, config.guess, --host, and
--target; making links to files; and running configure
scripts in subdirectories. Adding these features enabled Ken to convert
GNU as
, and Rob Savoye to convert DejaGNU, to using
Autoconf.
I added more features in response to other peoples' requests. Many
people had asked for configure scripts to share the results of
the checks between runs, because (particularly when configuring a large
source tree, like Cygnus does) they were frustratingly slow. Mike
Haertel suggested adding site-specific initialization scripts. People
distributing software that had to unpack on MS-DOS asked for a way to
override the .in extension on the file names, which produced file
names like config.h.in containing two dots. Jim Avera did an
extensive examination of the problems with quoting in AC_DEFINE
and AC_SUBST
; his insights led to significant improvements.
Richard Stallman asked that compiler output be sent to config.log
instead of /dev/null, to help people debug the Emacs
configure script.
I made some other changes because of my dissatisfaction with the quality of the program. I made the messages showing results of the checks less ambiguous, always printing a result. I regularized the names of the macros and cleaned up coding style inconsistencies. I added some auxiliary utilities that I had developed to help convert source code packages to use Autoconf. With the help of François Pinard, I made the macros not interrupt each others' messages. (That feature revealed some performance bottlenecks in GNU M4, which he hastily corrected!) I reorganized the documentation around problems people want to solve. And I began a test suite, because experience had shown that Autoconf has a pronounced tendency to regress when we change it.
Again, several alpha testers gave invaluable feedback, especially François Pinard, Jim Meyering, Karl Berry, Rob Savoye, Ken Raeburn, and Mark Eichin.
Finally, version 2.0 was ready. And there was much rejoicing. (And I have free time again. I think. Yeah, right.)
Copyright © 2000,2001,2002 Free Software Foundation, Inc. 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed.
The purpose of this License is to make a manual, textbook, or other functional and useful document free in the sense of freedom: to assure everyone the effective freedom to copy and redistribute it, with or without modifying it, either commercially or noncommercially. Secondarily, this License preserves for the author and publisher a way to get credit for their work, while not being considered responsible for modifications made by others.
This License is a kind of “copyleft”, which means that derivative works of the document must themselves be free in the same sense. It complements the GNU General Public License, which is a copyleft license designed for free software.
We have designed this License in order to use it for manuals for free software, because free software needs free documentation: a free program should come with manuals providing the same freedoms that the software does. But this License is not limited to software manuals; it can be used for any textual work, regardless of subject matter or whether it is published as a printed book. We recommend this License principally for works whose purpose is instruction or reference.
This License applies to any manual or other work, in any medium, that contains a notice placed by the copyright holder saying it can be distributed under the terms of this License. Such a notice grants a world-wide, royalty-free license, unlimited in duration, to use that work under the conditions stated herein. The “Document”, below, refers to any such manual or work. Any member of the public is a licensee, and is addressed as “you”. You accept the license if you copy, modify or distribute the work in a way requiring permission under copyright law.
A “Modified Version” of the Document means any work containing the Document or a portion of it, either copied verbatim, or with modifications and/or translated into another language.
A “Secondary Section” is a named appendix or a front-matter section of the Document that deals exclusively with the relationship of the publishers or authors of the Document to the Document's overall subject (or to related matters) and contains nothing that could fall directly within that overall subject. (Thus, if the Document is in part a textbook of mathematics, a Secondary Section may not explain any mathematics.) The relationship could be a matter of historical connection with the subject or with related matters, or of legal, commercial, philosophical, ethical or political position regarding them.
The “Invariant Sections” are certain Secondary Sections whose titles are designated, as being those of Invariant Sections, in the notice that says that the Document is released under this License. If a section does not fit the above definition of Secondary then it is not allowed to be designated as Invariant. The Document may contain zero Invariant Sections. If the Document does not identify any Invariant Sections then there are none.
The “Cover Texts” are certain short passages of text that are listed, as Front-Cover Texts or Back-Cover Texts, in the notice that says that the Document is released under this License. A Front-Cover Text may be at most 5 words, and a Back-Cover Text may be at most 25 words.
A “Transparent” copy of the Document means a machine-readable copy, represented in a format whose specification is available to the general public, that is suitable for revising the document straightforwardly with generic text editors or (for images composed of pixels) generic paint programs or (for drawings) some widely available drawing editor, and that is suitable for input to text formatters or for automatic translation to a variety of formats suitable for input to text formatters. A copy made in an otherwise Transparent file format whose markup, or absence of markup, has been arranged to thwart or discourage subsequent modification by readers is not Transparent. An image format is not Transparent if used for any substantial amount of text. A copy that is not “Transparent” is called “Opaque”.
Examples of suitable formats for Transparent copies include plain ascii without markup, Texinfo input format, LaTeX input format, SGML or XML using a publicly available DTD, and standard-conforming simple HTML, PostScript or PDF designed for human modification. Examples of transparent image formats include PNG, XCF and JPG. Opaque formats include proprietary formats that can be read and edited only by proprietary word processors, SGML or XML for which the DTD and/or processing tools are not generally available, and the machine-generated HTML, PostScript or PDF produced by some word processors for output purposes only.
The “Title Page” means, for a printed book, the title page itself, plus such following pages as are needed to hold, legibly, the material this License requires to appear in the title page. For works in formats which do not have any title page as such, “Title Page” means the text near the most prominent appearance of the work's title, preceding the beginning of the body of the text.
A section “Entitled XYZ” means a named subunit of the Document whose title either is precisely XYZ or contains XYZ in parentheses following text that translates XYZ in another language. (Here XYZ stands for a specific section name mentioned below, such as “Acknowledgements”, “Dedications”, “Endorsements”, or “History”.) To “Preserve the Title” of such a section when you modify the Document means that it remains a section “Entitled XYZ” according to this definition.
The Document may include Warranty Disclaimers next to the notice which states that this License applies to the Document. These Warranty Disclaimers are considered to be included by reference in this License, but only as regards disclaiming warranties: any other implication that these Warranty Disclaimers may have is void and has no effect on the meaning of this License.
You may copy and distribute the Document in any medium, either commercially or noncommercially, provided that this License, the copyright notices, and the license notice saying this License applies to the Document are reproduced in all copies, and that you add no other conditions whatsoever to those of this License. You may not use technical measures to obstruct or control the reading or further copying of the copies you make or distribute. However, you may accept compensation in exchange for copies. If you distribute a large enough number of copies you must also follow the conditions in section 3.
You may also lend copies, under the same conditions stated above, and you may publicly display copies.
If you publish printed copies (or copies in media that commonly have printed covers) of the Document, numbering more than 100, and the Document's license notice requires Cover Texts, you must enclose the copies in covers that carry, clearly and legibly, all these Cover Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on the back cover. Both covers must also clearly and legibly identify you as the publisher of these copies. The front cover must present the full title with all words of the title equally prominent and visible. You may add other material on the covers in addition. Copying with changes limited to the covers, as long as they preserve the title of the Document and satisfy these conditions, can be treated as verbatim copying in other respects.
If the required texts for either cover are too voluminous to fit legibly, you should put the first ones listed (as many as fit reasonably) on the actual cover, and continue the rest onto adjacent pages.
If you publish or distribute Opaque copies of the Document numbering more than 100, you must either include a machine-readable Transparent copy along with each Opaque copy, or state in or with each Opaque copy a computer-network location from which the general network-using public has access to download using public-standard network protocols a complete Transparent copy of the Document, free of added material. If you use the latter option, you must take reasonably prudent steps, when you begin distribution of Opaque copies in quantity, to ensure that this Transparent copy will remain thus accessible at the stated location until at least one year after the last time you distribute an Opaque copy (directly or through your agents or retailers) of that edition to the public.
It is requested, but not required, that you contact the authors of the Document well before redistributing any large number of copies, to give them a chance to provide you with an updated version of the Document.
You may copy and distribute a Modified Version of the Document under the conditions of sections 2 and 3 above, provided that you release the Modified Version under precisely this License, with the Modified Version filling the role of the Document, thus licensing distribution and modification of the Modified Version to whoever possesses a copy of it. In addition, you must do these things in the Modified Version:
If the Modified Version includes new front-matter sections or appendices that qualify as Secondary Sections and contain no material copied from the Document, you may at your option designate some or all of these sections as invariant. To do this, add their titles to the list of Invariant Sections in the Modified Version's license notice. These titles must be distinct from any other section titles.
You may add a section Entitled “Endorsements”, provided it contains nothing but endorsements of your Modified Version by various parties—for example, statements of peer review or that the text has been approved by an organization as the authoritative definition of a standard.
You may add a passage of up to five words as a Front-Cover Text, and a passage of up to 25 words as a Back-Cover Text, to the end of the list of Cover Texts in the Modified Version. Only one passage of Front-Cover Text and one of Back-Cover Text may be added by (or through arrangements made by) any one entity. If the Document already includes a cover text for the same cover, previously added by you or by arrangement made by the same entity you are acting on behalf of, you may not add another; but you may replace the old one, on explicit permission from the previous publisher that added the old one.
The author(s) and publisher(s) of the Document do not by this License give permission to use their names for publicity for or to assert or imply endorsement of any Modified Version.
You may combine the Document with other documents released under this License, under the terms defined in section 4 above for modified versions, provided that you include in the combination all of the Invariant Sections of all of the original documents, unmodified, and list them all as Invariant Sections of your combined work in its license notice, and that you preserve all their Warranty Disclaimers.
The combined work need only contain one copy of this License, and multiple identical Invariant Sections may be replaced with a single copy. If there are multiple Invariant Sections with the same name but different contents, make the title of each such section unique by adding at the end of it, in parentheses, the name of the original author or publisher of that section if known, or else a unique number. Make the same adjustment to the section titles in the list of Invariant Sections in the license notice of the combined work.
In the combination, you must combine any sections Entitled “History” in the various original documents, forming one section Entitled “History”; likewise combine any sections Entitled “Acknowledgements”, and any sections Entitled “Dedications”. You must delete all sections Entitled “Endorsements.”
You may make a collection consisting of the Document and other documents released under this License, and replace the individual copies of this License in the various documents with a single copy that is included in the collection, provided that you follow the rules of this License for verbatim copying of each of the documents in all other respects.
You may extract a single document from such a collection, and distribute it individually under this License, provided you insert a copy of this License into the extracted document, and follow this License in all other respects regarding verbatim copying of that document.
A compilation of the Document or its derivatives with other separate and independent documents or works, in or on a volume of a storage or distribution medium, is called an “aggregate” if the copyright resulting from the compilation is not used to limit the legal rights of the compilation's users beyond what the individual works permit. When the Document is included in an aggregate, this License does not apply to the other works in the aggregate which are not themselves derivative works of the Document.
If the Cover Text requirement of section 3 is applicable to these copies of the Document, then if the Document is less than one half of the entire aggregate, the Document's Cover Texts may be placed on covers that bracket the Document within the aggregate, or the electronic equivalent of covers if the Document is in electronic form. Otherwise they must appear on printed covers that bracket the whole aggregate.
Translation is considered a kind of modification, so you may distribute translations of the Document under the terms of section 4. Replacing Invariant Sections with translations requires special permission from their copyright holders, but you may include translations of some or all Invariant Sections in addition to the original versions of these Invariant Sections. You may include a translation of this License, and all the license notices in the Document, and any Warranty Disclaimers, provided that you also include the original English version of this License and the original versions of those notices and disclaimers. In case of a disagreement between the translation and the original version of this License or a notice or disclaimer, the original version will prevail.
If a section in the Document is Entitled “Acknowledgements”, “Dedications”, or “History”, the requirement (section 4) to Preserve its Title (section 1) will typically require changing the actual title.
You may not copy, modify, sublicense, or distribute the Document except as expressly provided for under this License. Any other attempt to copy, modify, sublicense or distribute the Document is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance.
The Free Software Foundation may publish new, revised versions of the GNU Free Documentation License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. See http://www.gnu.org/copyleft/.
Each version of the License is given a distinguishing version number. If the Document specifies that a particular numbered version of this License “or any later version” applies to it, you have the option of following the terms and conditions either of that specified version or of any later version that has been published (not as a draft) by the Free Software Foundation. If the Document does not specify a version number of this License, you may choose any version ever published (not as a draft) by the Free Software Foundation.
To use this License in a document you have written, include a copy of the License in the document and put the following copyright and license notices just after the title page:
Copyright (C) year your name. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license is included in the section entitled ``GNU Free Documentation License''.
If you have Invariant Sections, Front-Cover Texts and Back-Cover Texts, replace the “with...Texts.” line with this:
with the Invariant Sections being list their titles, with the Front-Cover Texts being list, and with the Back-Cover Texts being list.
If you have Invariant Sections without Cover Texts, or some other combination of the three, merge those two alternatives to suit the situation.
If your document contains nontrivial examples of program code, we recommend releasing these examples in parallel under your choice of free software license, such as the GNU General Public License, to permit their use in free software.
This is an alphabetical list of the environment variables that Autoconf checks.
BIN_SH
: Special Shell VariablesCDPATH
: Special Shell VariablesCONFIG_COMMANDS
: Obsolete config.status UseCONFIG_FILES
: Obsolete config.status UseCONFIG_HEADERS
: Obsolete config.status UseCONFIG_LINKS
: Obsolete config.status UseCONFIG_SHELL
: config.status InvocationCONFIG_SITE
: Site DefaultsCONFIG_STATUS
: config.status InvocationDUALCASE
: Special Shell VariablesENV
: Special Shell VariablesIFS
: Special Shell VariablesLANG
: Special Shell VariablesLANGUAGE
: Special Shell VariablesLC_ADDRESS
: Special Shell VariablesLC_ALL
: Special Shell VariablesLC_COLLATE
: Special Shell VariablesLC_CTYPE
: Special Shell VariablesLC_IDENTIFICATION
: Special Shell VariablesLC_MEASUREMENT
: Special Shell VariablesLC_MESSAGES
: Special Shell VariablesLC_MONETARY
: Special Shell VariablesLC_NAME
: Special Shell VariablesLC_NUMERIC
: Special Shell VariablesLC_PAPER
: Special Shell VariablesLC_TELEPHONE
: Special Shell VariablesLC_TIME
: Special Shell VariablesM4
: autom4te InvocationMAIL
: Special Shell VariablesMAILPATH
: Special Shell VariablesNULLCMD
: Special Shell VariablesPATH_SEPARATOR
: Special Shell VariablesPS1
: Special Shell VariablesPS2
: Special Shell VariablesPS4
: Special Shell VariablesPWD
: Special Shell VariablesSIMPLE_BACKUP_SUFFIX
: autoupdate InvocationWARNINGS
: autom4te InvocationWARNINGS
: autoheader InvocationWARNINGS
: autoreconf InvocationWARNINGS
: autoconf InvocationXMKMF
: System ServicesThis is an alphabetical list of the variables that Autoconf can substitute into files that it creates, typically one or more makefiles. See Setting Output Variables, for more information on how this is done.
abs_builddir
: Preset Output Variablesabs_srcdir
: Preset Output Variablesabs_top_builddir
: Preset Output Variablesabs_top_srcdir
: Preset Output VariablesALLOCA
: Particular FunctionsAWK
: Particular Programsbindir
: Installation Directory Variablesbuild
: Canonicalizingbuild_alias
: Canonicalizingbuild_cpu
: Canonicalizingbuild_os
: Canonicalizingbuild_vendor
: Canonicalizingbuilddir
: Preset Output VariablesCC
: System ServicesCC
: C CompilerCFLAGS
: C CompilerCFLAGS
: Preset Output Variablesconfigure_input
: Preset Output VariablesCPP
: C CompilerCPPFLAGS
: Preset Output Variablescross_compiling
: RuntimeCXX
: C++ CompilerCXXCPP
: C++ CompilerCXXFLAGS
: C++ CompilerCXXFLAGS
: Preset Output Variablesdatadir
: Installation Directory Variablesdatarootdir
: Installation Directory VariablesDEFS
: Preset Output Variablesdocdir
: Installation Directory Variablesdvidir
: Installation Directory VariablesECHO_C
: Preset Output VariablesECHO_N
: Preset Output VariablesECHO_T
: Preset Output VariablesEGREP
: Particular ProgramsERL
: Running the CompilerERL
: Language ChoiceERL
: Erlang Compiler and InterpreterERLANG_INSTALL_LIB_DIR
: Erlang LibrariesERLANG_INSTALL_LIB_DIR
: Installation Directory VariablesERLANG_INSTALL_LIB_DIR_
library: Erlang LibrariesERLANG_INSTALL_LIB_DIR_
library: Installation Directory VariablesERLANG_LIB_DIR
: Erlang LibrariesERLANG_LIB_DIR_
library: Erlang LibrariesERLANG_ROOT_DIR
: Erlang LibrariesERLC
: Language ChoiceERLC
: Erlang Compiler and InterpreterERLCFLAGS
: Language ChoiceERLCFLAGS
: Erlang Compiler and InterpreterERLCFLAGS
: Preset Output Variablesexec_prefix
: Installation Directory VariablesEXEEXT
: Obsolete MacrosEXEEXT
: Compilers and PreprocessorsF77
: Fortran CompilerFC
: Fortran CompilerFCFLAGS
: Fortran CompilerFCFLAGS
: Preset Output VariablesFCLIBS
: Fortran CompilerFFLAGS
: Fortran CompilerFFLAGS
: Preset Output VariablesFGREP
: Particular ProgramsFLIBS
: Fortran CompilerGETGROUPS_LIBS
: Particular FunctionsGETLOADAVG_LIBS
: Particular FunctionsGREP
: Particular Programshost
: Canonicalizinghost_alias
: Canonicalizinghost_cpu
: Canonicalizinghost_os
: Canonicalizinghost_vendor
: Canonicalizinghtmldir
: Installation Directory Variablesincludedir
: Installation Directory Variablesinfodir
: Installation Directory VariablesINSTALL
: Particular ProgramsINSTALL_DATA
: Particular ProgramsINSTALL_PROGRAM
: Particular ProgramsINSTALL_SCRIPT
: Particular ProgramsKMEM_GROUP
: Particular FunctionsLDFLAGS
: Preset Output VariablesLEX
: Particular ProgramsLEX_OUTPUT_ROOT
: Particular ProgramsLEXLIB
: Particular Programslibdir
: Installation Directory Variableslibexecdir
: Installation Directory VariablesLIBOBJDIR
: AC_LIBOBJ vs LIBOBJSLIBOBJS
: Particular StructuresLIBOBJS
: Generic FunctionsLIBOBJS
: Particular FunctionsLIBS
: Obsolete MacrosLIBS
: Posix VariantsLIBS
: Preset Output VariablesLN_S
: Particular Programslocaledir
: Installation Directory Variableslocalstatedir
: Installation Directory Variablesmandir
: Installation Directory VariablesMKDIR_P
: Particular ProgramsNEED_SETGID
: Particular FunctionsOBJC
: Objective C CompilerOBJCCPP
: Objective C CompilerOBJCFLAGS
: Objective C CompilerOBJCFLAGS
: Preset Output VariablesOBJEXT
: Obsolete MacrosOBJEXT
: Compilers and Preprocessorsoldincludedir
: Installation Directory VariablesPACKAGE_BUGREPORT
: Initializing configurePACKAGE_NAME
: Initializing configurePACKAGE_STRING
: Initializing configurePACKAGE_TARNAME
: Initializing configurePACKAGE_VERSION
: Initializing configurepdfdir
: Installation Directory VariablesPOW_LIB
: Particular Functionsprefix
: Installation Directory Variablesprogram_transform_name
: Transforming Namespsdir
: Installation Directory VariablesRANLIB
: Particular Programssbindir
: Installation Directory VariablesSED
: Particular ProgramsSET_MAKE
: Outputsharedstatedir
: Installation Directory Variablessrcdir
: Preset Output Variablessubdirs
: Subdirectoriessysconfdir
: Installation Directory Variablestarget
: Canonicalizingtarget_alias
: Canonicalizingtarget_cpu
: Canonicalizingtarget_os
: Canonicalizingtarget_vendor
: Canonicalizingtop_builddir
: Preset Output Variablestop_srcdir
: Preset Output VariablesX_CFLAGS
: System ServicesX_EXTRA_LIBS
: System ServicesX_LIBS
: System ServicesX_PRE_LIBS
: System ServicesYACC
: Particular ProgramsThis is an alphabetical list of the C preprocessor symbols that the
Autoconf macros define. To work with Autoconf, C source code needs to
use these names in #if
directives.
__CHAR_UNSIGNED__
: C Compiler__EXTENSIONS__
: Posix Variants__PROTOTYPES
: C Compiler_ALL_SOURCE
: Posix Variants_FILE_OFFSET_BITS
: System Services_GNU_SOURCE
: Posix Variants_LARGE_FILES
: System Services_LARGEFILE_SOURCE
: Particular Functions_MINIX
: Posix Variants_POSIX_1_SOURCE
: Posix Variants_POSIX_PTHREAD_SEMANTICS
: Posix Variants_POSIX_SOURCE
: Posix Variants_POSIX_VERSION
: Particular HeadersC_ALLOCA
: Particular FunctionsC_GETLOADAVG
: Particular FunctionsCLOSEDIR_VOID
: Particular Functionsconst
: C CompilerCXX_NO_MINUS_C_MINUS_O
: C++ CompilerDGUX
: Particular FunctionsDIRENT
: Obsolete MacrosF77_DUMMY_MAIN
: Fortran CompilerF77_FUNC
: Fortran CompilerF77_FUNC_
: Fortran CompilerF77_MAIN
: Fortran CompilerF77_NO_MINUS_C_MINUS_O
: Fortran CompilerFC_FUNC
: Fortran CompilerFC_FUNC_
: Fortran CompilerFC_MAIN
: Fortran CompilerFC_NO_MINUS_C_MINUS_O
: Fortran CompilerGETGROUPS_T
: Particular TypesGETLODAVG_PRIVILEGED
: Particular FunctionsGETPGRP_VOID
: Particular Functionsgid_t
: Particular TypesGWINSZ_IN_SYS_IOCTL
: Particular HeadersHAVE__BOOL
: Particular HeadersHAVE_ALLOCA_H
: Particular FunctionsHAVE_CONFIG_H
: Configuration HeadersHAVE_DECL_STRERROR_R
: Particular FunctionsHAVE_DECL_
symbol: Generic DeclarationsHAVE_DIRENT_H
: Particular HeadersHAVE_DOPRNT
: Particular FunctionsHAVE_
function: Generic FunctionsHAVE_GETMNTENT
: Particular FunctionsHAVE_
header: Generic HeadersHAVE_INT16_T
: Particular TypesHAVE_INT32_T
: Particular TypesHAVE_INT64_T
: Particular TypesHAVE_INT8_T
: Particular TypesHAVE_INTMAX_T
: Particular TypesHAVE_INTPTR_T
: Particular TypesHAVE_LONG_DOUBLE
: Obsolete MacrosHAVE_LONG_DOUBLE
: Particular TypesHAVE_LONG_DOUBLE_WIDER
: Particular TypesHAVE_LONG_FILE_NAMES
: System ServicesHAVE_LONG_LONG_INT
: Particular TypesHAVE_LSTAT_EMPTY_STRING_BUG
: Particular FunctionsHAVE_MALLOC
: Particular FunctionsHAVE_MBRTOWC
: Particular FunctionsHAVE_MMAP
: Particular FunctionsHAVE_NDIR_H
: Particular HeadersHAVE_NLIST_H
: Particular FunctionsHAVE_OBSTACK
: Particular FunctionsHAVE_REALLOC
: Particular FunctionsHAVE_RESOLV_H
: Particular HeadersHAVE_RESTARTABLE_SYSCALLS
: Obsolete MacrosHAVE_ST_BLKSIZE
: Particular StructuresHAVE_ST_BLOCKS
: Particular StructuresHAVE_ST_RDEV
: Particular StructuresHAVE_STAT_EMPTY_STRING_BUG
: Particular FunctionsHAVE_STDBOOL_H
: Particular HeadersHAVE_STRCOLL
: Particular FunctionsHAVE_STRERROR_R
: Particular FunctionsHAVE_STRFTIME
: Particular FunctionsHAVE_STRINGIZE
: C CompilerHAVE_STRNLEN
: Particular FunctionsHAVE_STRUCT_DIRENT_D_INO
: Particular StructuresHAVE_STRUCT_DIRENT_D_TYPE
: Particular StructuresHAVE_STRUCT_STAT_ST_BLKSIZE
: Particular StructuresHAVE_STRUCT_STAT_ST_BLOCKS
: Particular StructuresHAVE_STRUCT_STAT_ST_RDEV
: Particular StructuresHAVE_SYS_DIR_H
: Particular HeadersHAVE_SYS_NDIR_H
: Particular HeadersHAVE_SYS_WAIT_H
: Particular HeadersHAVE_TM_ZONE
: Particular StructuresHAVE_TYPEOF
: C CompilerHAVE_TZNAME
: Particular StructuresHAVE_UINT16_T
: Particular TypesHAVE_UINT32_T
: Particular TypesHAVE_UINT64_T
: Particular TypesHAVE_UINT8_T
: Particular TypesHAVE_UINTMAX_T
: Particular TypesHAVE_UINTPTR_T
: Particular TypesHAVE_UNSIGNED_LONG_LONG_INT
: Particular TypesHAVE_UTIME_NULL
: Particular FunctionsHAVE_VFORK_H
: Particular FunctionsHAVE_VPRINTF
: Particular FunctionsHAVE_WAIT3
: Obsolete MacrosHAVE_WORKING_FORK
: Particular FunctionsHAVE_WORKING_VFORK
: Particular Functionsinline
: C Compilerint16_t
: Particular Typesint32_t
: Particular Typesint64_t
: Particular Typesint8_t
: Particular TypesINT_16_BITS
: Obsolete Macrosintmax_t
: Particular Typesintptr_t
: Particular TypesLONG_64_BITS
: Obsolete MacrosLSTAT_FOLLOWS_SLASHED_SYMLINK
: Particular FunctionsMAJOR_IN_MKDEV
: Particular HeadersMAJOR_IN_SYSMACROS
: Particular Headersmalloc
: Particular Functionsmbstate_t
: Particular Typesmode_t
: Particular TypesNDEBUG
: Particular HeadersNDIR
: Obsolete MacrosNEED_MEMORY_H
: Obsolete MacrosNEED_SETGID
: Particular FunctionsNLIST_NAME_UNION
: Particular FunctionsNO_MINUS_C_MINUS_O
: C Compileroff_t
: Particular TypesPACKAGE_BUGREPORT
: Initializing configurePACKAGE_NAME
: Initializing configurePACKAGE_STRING
: Initializing configurePACKAGE_TARNAME
: Initializing configurePACKAGE_VERSION
: Initializing configurePARAMS
: C Compilerpid_t
: Particular TypesPROTOTYPES
: C Compilerrealloc
: Particular Functionsrestrict
: C CompilerRETSIGTYPE
: Particular TypesSELECT_TYPE_ARG1
: Particular FunctionsSELECT_TYPE_ARG234
: Particular FunctionsSELECT_TYPE_ARG5
: Particular FunctionsSETPGRP_VOID
: Particular FunctionsSETVBUF_REVERSED
: Particular Functionssize_t
: Particular Typesssize_t
: Particular TypesSTAT_MACROS_BROKEN
: Particular HeadersSTDC_HEADERS
: Particular HeadersSTRERROR_R_CHAR_P
: Particular FunctionsSVR4
: Particular FunctionsSYS_SIGLIST_DECLARED
: Obsolete MacrosSYSDIR
: Obsolete MacrosSYSNDIR
: Obsolete MacrosTIME_WITH_SYS_TIME
: Particular HeadersTM_IN_SYS_TIME
: Particular Structurestypeof
: C Compileruid_t
: Particular Typesuint16_t
: Particular Typesuint32_t
: Particular Typesuint64_t
: Particular Typesuint8_t
: Particular Typesuintmax_t
: Particular Typesuintptr_t
: Particular TypesUMAX
: Particular FunctionsUMAX4_3
: Particular FunctionsUSG
: Obsolete Macrosvfork
: Particular Functionsvolatile
: C CompilerWORDS_BIGENDIAN
: C CompilerX_DISPLAY_MISSING
: System ServicesYYTEXT_POINTER
: Particular ProgramsThis is an alphabetical list of the Autoconf macros.
AC_AC_PROG_MKDIR_P
: Particular ProgramsAC_AIX
: Posix VariantsAC_ALLOCA
: Obsolete MacrosAC_ARG_ARRAY
: Obsolete MacrosAC_ARG_ENABLE
: Package OptionsAC_ARG_PROGRAM
: Transforming NamesAC_ARG_VAR
: Setting Output VariablesAC_ARG_WITH
: External SoftwareAC_BEFORE
: Suggested OrderingAC_C_BIGENDIAN
: C CompilerAC_C_CHAR_UNSIGNED
: C CompilerAC_C_CONST
: C CompilerAC_C_CROSS
: Obsolete MacrosAC_C_INLINE
: C CompilerAC_C_LONG_DOUBLE
: Obsolete MacrosAC_C_PROTOTYPES
: C CompilerAC_C_RESTRICT
: C CompilerAC_C_STRINGIZE
: C CompilerAC_C_TYPEOF
: C CompilerAC_C_VOLATILE
: C CompilerAC_CACHE_CHECK
: Caching ResultsAC_CACHE_LOAD
: Cache CheckpointingAC_CACHE_SAVE
: Cache CheckpointingAC_CACHE_VAL
: Caching ResultsAC_CANONICAL_BUILD
: CanonicalizingAC_CANONICAL_HOST
: CanonicalizingAC_CANONICAL_SYSTEM
: Obsolete MacrosAC_CANONICAL_TARGET
: CanonicalizingAC_CHAR_UNSIGNED
: Obsolete MacrosAC_CHECK_ALIGNOF
: Generic Compiler CharacteristicsAC_CHECK_DECL
: Generic DeclarationsAC_CHECK_DECLS
: Generic DeclarationsAC_CHECK_DECLS_ONCE
: Generic DeclarationsAC_CHECK_FILE
: FilesAC_CHECK_FILES
: FilesAC_CHECK_FUNC
: Generic FunctionsAC_CHECK_FUNCS
: Generic FunctionsAC_CHECK_FUNCS_ONCE
: Generic FunctionsAC_CHECK_HEADER
: Generic HeadersAC_CHECK_HEADERS
: Generic HeadersAC_CHECK_HEADERS_ONCE
: Generic HeadersAC_CHECK_LIB
: LibrariesAC_CHECK_MEMBER
: Generic StructuresAC_CHECK_MEMBERS
: Generic StructuresAC_CHECK_PROG
: Generic ProgramsAC_CHECK_PROGS
: Generic ProgramsAC_CHECK_SIZEOF
: Generic Compiler CharacteristicsAC_CHECK_TARGET_TOOL
: Generic ProgramsAC_CHECK_TARGET_TOOLS
: Generic ProgramsAC_CHECK_TOOL
: Generic ProgramsAC_CHECK_TOOLS
: Generic ProgramsAC_CHECK_TYPE
: Obsolete MacrosAC_CHECK_TYPE
: Generic TypesAC_CHECK_TYPES
: Generic TypesAC_CHECKING
: Obsolete MacrosAC_COMPILE_CHECK
: Obsolete MacrosAC_COMPILE_IFELSE
: Running the CompilerAC_CONFIG_AUX_DIR
: InputAC_CONFIG_COMMANDS
: Configuration CommandsAC_CONFIG_COMMANDS_POST
: Configuration CommandsAC_CONFIG_COMMANDS_PRE
: Configuration CommandsAC_CONFIG_FILES
: Configuration FilesAC_CONFIG_HEADERS
: Configuration HeadersAC_CONFIG_LIBOBJ_DIR
: Generic FunctionsAC_CONFIG_LINKS
: Configuration LinksAC_CONFIG_MACRO_DIR
: InputAC_CONFIG_SRCDIR
: InputAC_CONFIG_SUBDIRS
: SubdirectoriesAC_CONFIG_TESTDIR
: Making testsuite ScriptsAC_CONST
: Obsolete MacrosAC_COPYRIGHT
: NoticesAC_CROSS_CHECK
: Obsolete MacrosAC_CYGWIN
: Obsolete MacrosAC_DATAROOTDIR_CHECKED
: Changed Directory VariablesAC_DECL_SYS_SIGLIST
: Obsolete MacrosAC_DECL_YYTEXT
: Obsolete MacrosAC_DEFINE
: Defining SymbolsAC_DEFINE_UNQUOTED
: Defining SymbolsAC_DEFUN
: Macro DefinitionsAC_DEFUN_ONCE
: One-Shot MacrosAC_DIAGNOSE
: Reporting MessagesAC_DIR_HEADER
: Obsolete MacrosAC_DYNIX_SEQ
: Obsolete MacrosAC_EGREP_CPP
: Running the PreprocessorAC_EGREP_HEADER
: Running the PreprocessorAC_EMXOS2
: Obsolete MacrosAC_ENABLE
: Package OptionsAC_ERLANG_CHECK_LIB
: Erlang LibrariesAC_ERLANG_NEED_ERL
: Erlang Compiler and InterpreterAC_ERLANG_NEED_ERLC
: Erlang Compiler and InterpreterAC_ERLANG_PATH_ERL
: Erlang Compiler and InterpreterAC_ERLANG_PATH_ERLC
: Erlang Compiler and InterpreterAC_ERLANG_SUBST_INSTALL_LIB_DIR
: Erlang LibrariesAC_ERLANG_SUBST_INSTALL_LIB_DIR
: Installation Directory VariablesAC_ERLANG_SUBST_INSTALL_LIB_SUBDIR
: Erlang LibrariesAC_ERLANG_SUBST_INSTALL_LIB_SUBDIR
: Installation Directory VariablesAC_ERLANG_SUBST_LIB_DIR
: Erlang LibrariesAC_ERLANG_SUBST_ROOT_DIR
: Erlang LibrariesAC_ERROR
: Obsolete MacrosAC_EXEEXT
: Obsolete MacrosAC_F77_DUMMY_MAIN
: Fortran CompilerAC_F77_FUNC
: Fortran CompilerAC_F77_LIBRARY_LDFLAGS
: Fortran CompilerAC_F77_MAIN
: Fortran CompilerAC_F77_WRAPPERS
: Fortran CompilerAC_FATAL
: Reporting MessagesAC_FC_FREEFORM
: Fortran CompilerAC_FC_FUNC
: Fortran CompilerAC_FC_LIBRARY_LDFLAGS
: Fortran CompilerAC_FC_MAIN
: Fortran CompilerAC_FC_SRCEXT
: Fortran CompilerAC_FC_WRAPPERS
: Fortran CompilerAC_FIND_X
: Obsolete MacrosAC_FIND_XTRA
: Obsolete MacrosAC_FOREACH
: Obsolete MacrosAC_FUNC_ALLOCA
: Particular FunctionsAC_FUNC_CHECK
: Obsolete MacrosAC_FUNC_CHOWN
: Particular FunctionsAC_FUNC_CLOSEDIR_VOID
: Particular FunctionsAC_FUNC_ERROR_AT_LINE
: Particular FunctionsAC_FUNC_FNMATCH
: Particular FunctionsAC_FUNC_FNMATCH_GNU
: Particular FunctionsAC_FUNC_FORK
: Particular FunctionsAC_FUNC_FSEEKO
: Particular FunctionsAC_FUNC_GETGROUPS
: Particular FunctionsAC_FUNC_GETLOADAVG
: Particular FunctionsAC_FUNC_GETMNTENT
: Particular FunctionsAC_FUNC_GETPGRP
: Particular FunctionsAC_FUNC_LSTAT
: Particular FunctionsAC_FUNC_LSTAT_FOLLOWS_SLASHED_SYMLINK
: Particular FunctionsAC_FUNC_MALLOC
: Particular FunctionsAC_FUNC_MBRTOWC
: Particular FunctionsAC_FUNC_MEMCMP
: Particular FunctionsAC_FUNC_MKTIME
: Particular FunctionsAC_FUNC_MMAP
: Particular FunctionsAC_FUNC_OBSTACK
: Particular FunctionsAC_FUNC_REALLOC
: Particular FunctionsAC_FUNC_SELECT_ARGTYPES
: Particular FunctionsAC_FUNC_SETPGRP
: Particular FunctionsAC_FUNC_SETVBUF_REVERSED
: Particular FunctionsAC_FUNC_STAT
: Particular FunctionsAC_FUNC_STRCOLL
: Particular FunctionsAC_FUNC_STRERROR_R
: Particular FunctionsAC_FUNC_STRFTIME
: Particular FunctionsAC_FUNC_STRNLEN
: Particular FunctionsAC_FUNC_STRTOD
: Particular FunctionsAC_FUNC_STRTOLD
: Particular FunctionsAC_FUNC_UTIME_NULL
: Particular FunctionsAC_FUNC_VPRINTF
: Particular FunctionsAC_FUNC_WAIT3
: Obsolete MacrosAC_GCC_TRADITIONAL
: Obsolete MacrosAC_GETGROUPS_T
: Obsolete MacrosAC_GETLOADAVG
: Obsolete MacrosAC_GNU_SOURCE
: Posix VariantsAC_HAVE_C_BACKSLASH_A
: C CompilerAC_HAVE_FUNCS
: Obsolete MacrosAC_HAVE_HEADERS
: Obsolete MacrosAC_HAVE_LIBRARY
: Obsolete MacrosAC_HAVE_POUNDBANG
: Obsolete MacrosAC_HEADER_ASSERT
: Particular HeadersAC_HEADER_CHECK
: Obsolete MacrosAC_HEADER_DIRENT
: Particular HeadersAC_HEADER_EGREP
: Obsolete MacrosAC_HEADER_MAJOR
: Particular HeadersAC_HEADER_RESOLV
: Particular HeadersAC_HEADER_STAT
: Particular HeadersAC_HEADER_STDBOOL
: Particular HeadersAC_HEADER_STDC
: Particular HeadersAC_HEADER_SYS_WAIT
: Particular HeadersAC_HEADER_TIME
: Particular HeadersAC_HEADER_TIOCGWINSZ
: Particular HeadersAC_HELP_STRING
: Obsolete MacrosAC_HELP_STRING
: Pretty Help StringsAC_INCLUDES_DEFAULT
: Default IncludesAC_INIT
: Obsolete MacrosAC_INIT
: Initializing configureAC_INLINE
: Obsolete MacrosAC_INT_16_BITS
: Obsolete MacrosAC_IRIX_SUN
: Obsolete MacrosAC_ISC_POSIX
: Posix VariantsAC_LANG_ASSERT
: Language ChoiceAC_LANG_C
: Obsolete MacrosAC_LANG_CALL
: Generating SourcesAC_LANG_CONFTEST
: Generating SourcesAC_LANG_CPLUSPLUS
: Obsolete MacrosAC_LANG_FORTRAN77
: Obsolete MacrosAC_LANG_FUNC_LINK_TRY
: Generating SourcesAC_LANG_POP
: Language ChoiceAC_LANG_PROGRAM
: Generating SourcesAC_LANG_PUSH
: Language ChoiceAC_LANG_RESTORE
: Obsolete MacrosAC_LANG_SAVE
: Obsolete MacrosAC_LANG_SOURCE
: Generating SourcesAC_LANG_WERROR
: Generic Compiler CharacteristicsAC_LIBOBJ
: Generic FunctionsAC_LIBSOURCE
: Generic FunctionsAC_LIBSOURCES
: Generic FunctionsAC_LINK_FILES
: Obsolete MacrosAC_LINK_IFELSE
: Running the LinkerAC_LN_S
: Obsolete MacrosAC_LONG_64_BITS
: Obsolete MacrosAC_LONG_DOUBLE
: Obsolete MacrosAC_LONG_FILE_NAMES
: Obsolete MacrosAC_MAJOR_HEADER
: Obsolete MacrosAC_MEMORY_H
: Obsolete MacrosAC_MINGW32
: Obsolete MacrosAC_MINIX
: Posix VariantsAC_MINUS_C_MINUS_O
: Obsolete MacrosAC_MMAP
: Obsolete MacrosAC_MODE_T
: Obsolete MacrosAC_MSG_CHECKING
: Printing MessagesAC_MSG_ERROR
: Printing MessagesAC_MSG_FAILURE
: Printing MessagesAC_MSG_NOTICE
: Printing MessagesAC_MSG_RESULT
: Printing MessagesAC_MSG_WARN
: Printing MessagesAC_OBJEXT
: Obsolete MacrosAC_OBSOLETE
: Obsolete MacrosAC_OFF_T
: Obsolete MacrosAC_OUTPUT
: Obsolete MacrosAC_OUTPUT
: OutputAC_OUTPUT_COMMANDS
: Obsolete MacrosAC_PACKAGE_BUGREPORT
: Initializing configureAC_PACKAGE_NAME
: Initializing configureAC_PACKAGE_STRING
: Initializing configureAC_PACKAGE_TARNAME
: Initializing configureAC_PACKAGE_VERSION
: Initializing configureAC_PATH_PROG
: Generic ProgramsAC_PATH_PROGS
: Generic ProgramsAC_PATH_TARGET_TOOL
: Generic ProgramsAC_PATH_TOOL
: Generic ProgramsAC_PATH_X
: System ServicesAC_PATH_XTRA
: System ServicesAC_PID_T
: Obsolete MacrosAC_PREFIX
: Obsolete MacrosAC_PREFIX_DEFAULT
: Default PrefixAC_PREFIX_PROGRAM
: Default PrefixAC_PREPROC_IFELSE
: Running the PreprocessorAC_PREREQ
: NoticesAC_PRESERVE_HELP_ORDER
: Help FormattingAC_PROG_AWK
: Particular ProgramsAC_PROG_CC
: C CompilerAC_PROG_CC_C89
: C CompilerAC_PROG_CC_C99
: C CompilerAC_PROG_CC_C_O
: C CompilerAC_PROG_CC_STDC
: C CompilerAC_PROG_CPP
: C CompilerAC_PROG_CPP_WERROR
: C CompilerAC_PROG_CXX
: C++ CompilerAC_PROG_CXX_C_O
: C++ CompilerAC_PROG_CXXCPP
: C++ CompilerAC_PROG_EGREP
: Particular ProgramsAC_PROG_F77
: Fortran CompilerAC_PROG_F77_C_O
: Fortran CompilerAC_PROG_FC
: Fortran CompilerAC_PROG_FC_C_O
: Fortran CompilerAC_PROG_FGREP
: Particular ProgramsAC_PROG_GCC_TRADITIONAL
: C CompilerAC_PROG_GREP
: Particular ProgramsAC_PROG_INSTALL
: Particular ProgramsAC_PROG_LEX
: Particular ProgramsAC_PROG_LN_S
: Particular ProgramsAC_PROG_MAKE_SET
: OutputAC_PROG_OBJC
: Objective C CompilerAC_PROG_OBJCCPP
: Objective C CompilerAC_PROG_RANLIB
: Particular ProgramsAC_PROG_SED
: Particular ProgramsAC_PROG_YACC
: Particular ProgramsAC_PROGRAM_CHECK
: Obsolete MacrosAC_PROGRAM_EGREP
: Obsolete MacrosAC_PROGRAM_PATH
: Obsolete MacrosAC_PROGRAMS_CHECK
: Obsolete MacrosAC_PROGRAMS_PATH
: Obsolete MacrosAC_REMOTE_TAPE
: Obsolete MacrosAC_REPLACE_FNMATCH
: Particular FunctionsAC_REPLACE_FUNCS
: Generic FunctionsAC_REQUIRE
: Prerequisite MacrosAC_REQUIRE_AUX_FILE
: InputAC_REQUIRE_CPP
: Language ChoiceAC_RESTARTABLE_SYSCALLS
: Obsolete MacrosAC_RETSIGTYPE
: Obsolete MacrosAC_REVISION
: NoticesAC_RSH
: Obsolete MacrosAC_RUN_IFELSE
: RuntimeAC_SCO_INTL
: Obsolete MacrosAC_SEARCH_LIBS
: LibrariesAC_SET_MAKE
: Obsolete MacrosAC_SETVBUF_REVERSED
: Obsolete MacrosAC_SIZE_T
: Obsolete MacrosAC_SIZEOF_TYPE
: Obsolete MacrosAC_ST_BLKSIZE
: Obsolete MacrosAC_ST_BLOCKS
: Obsolete MacrosAC_ST_RDEV
: Obsolete MacrosAC_STAT_MACROS_BROKEN
: Obsolete MacrosAC_STDC_HEADERS
: Obsolete MacrosAC_STRCOLL
: Obsolete MacrosAC_STRUCT_DIRENT_D_INO
: Particular StructuresAC_STRUCT_DIRENT_D_TYPE
: Particular StructuresAC_STRUCT_ST_BLKSIZE
: Particular StructuresAC_STRUCT_ST_BLOCKS
: Particular StructuresAC_STRUCT_ST_RDEV
: Particular StructuresAC_STRUCT_TIMEZONE
: Particular StructuresAC_STRUCT_TM
: Particular StructuresAC_SUBST
: Setting Output VariablesAC_SUBST_FILE
: Setting Output VariablesAC_SYS_INTERPRETER
: System ServicesAC_SYS_LARGEFILE
: System ServicesAC_SYS_LONG_FILE_NAMES
: System ServicesAC_SYS_POSIX_TERMIOS
: System ServicesAC_SYS_RESTARTABLE_SYSCALLS
: Obsolete MacrosAC_SYS_SIGLIST_DECLARED
: Obsolete MacrosAC_TEST_CPP
: Obsolete MacrosAC_TEST_PROGRAM
: Obsolete MacrosAC_TIME_WITH_SYS_TIME
: Obsolete MacrosAC_TIMEZONE
: Obsolete MacrosAC_TRY_COMPILE
: Obsolete MacrosAC_TRY_CPP
: Obsolete MacrosAC_TRY_LINK
: Obsolete MacrosAC_TRY_LINK_FUNC
: Obsolete MacrosAC_TRY_RUN
: Obsolete MacrosAC_TYPE_GETGROUPS
: Particular TypesAC_TYPE_INT16_T
: Particular TypesAC_TYPE_INT32_T
: Particular TypesAC_TYPE_INT64_T
: Particular TypesAC_TYPE_INT8_T
: Particular TypesAC_TYPE_INTMAX_T
: Particular TypesAC_TYPE_INTPTR_T
: Particular TypesAC_TYPE_LONG_DOUBLE
: Particular TypesAC_TYPE_LONG_DOUBLE_WIDER
: Particular TypesAC_TYPE_LONG_LONG_INT
: Particular TypesAC_TYPE_MBSTATE_T
: Particular TypesAC_TYPE_MODE_T
: Particular TypesAC_TYPE_OFF_T
: Particular TypesAC_TYPE_PID_T
: Particular TypesAC_TYPE_SIGNAL
: Particular TypesAC_TYPE_SIZE_T
: Particular TypesAC_TYPE_SSIZE_T
: Particular TypesAC_TYPE_UID_T
: Particular TypesAC_TYPE_UINT16_T
: Particular TypesAC_TYPE_UINT32_T
: Particular TypesAC_TYPE_UINT64_T
: Particular TypesAC_TYPE_UINT8_T
: Particular TypesAC_TYPE_UINTMAX_T
: Particular TypesAC_TYPE_UINTPTR_T
: Particular TypesAC_TYPE_UNSIGNED_LONG_LONG_INT
: Particular TypesAC_UID_T
: Obsolete MacrosAC_UNISTD_H
: Obsolete MacrosAC_USE_SYSTEM_EXTENSIONS
: Posix VariantsAC_USG
: Obsolete MacrosAC_UTIME_NULL
: Obsolete MacrosAC_VALIDATE_CACHED_SYSTEM_TUPLE
: Obsolete MacrosAC_VERBOSE
: Obsolete MacrosAC_VFORK
: Obsolete MacrosAC_VPRINTF
: Obsolete MacrosAC_WAIT3
: Obsolete MacrosAC_WARN
: Obsolete MacrosAC_WARNING
: Reporting MessagesAC_WITH
: External SoftwareAC_WORDS_BIGENDIAN
: Obsolete MacrosAC_XENIX_DIR
: Obsolete MacrosAC_YYTEXT_POINTER
: Obsolete MacrosAH_BOTTOM
: Autoheader MacrosAH_HEADER
: Configuration HeadersAH_TEMPLATE
: Autoheader MacrosAH_TOP
: Autoheader MacrosAH_VERBATIM
: Autoheader MacrosAU_ALIAS
: Obsoleting MacrosAU_DEFUN
: Obsoleting MacrosThis is an alphabetical list of the M4, M4sugar, and M4sh macros.
AS_BOURNE_COMPATIBLE
: Programming in M4shAS_CASE
: Programming in M4shAS_DIRNAME
: Programming in M4shAS_IF
: Programming in M4shAS_MESSAGE_FD
: File Descriptor MacrosAS_MESSAGE_LOG_FD
: File Descriptor MacrosAS_MKDIR_P
: Programming in M4shAS_ORIGINAL_STDIN_FD
: File Descriptor MacrosAS_SET_CATFILE
: Programming in M4shAS_SHELL_SANITIZE
: Programming in M4shAS_TR_CPP
: Programming in M4shAS_TR_SH
: Programming in M4shm4_append
: Text processing Macrosm4_append_uniq
: Text processing Macrosm4_bpatsubst
: Redefined M4 Macrosm4_bregexp
: Redefined M4 Macrosm4_builtin
: Redefined M4 Macrosm4_decr
: Redefined M4 Macrosm4_define
: Redefined M4 Macrosm4_defn
: Redefined M4 Macrosm4_dnl
: Redefined M4 Macrosm4_dquote
: Evaluation Macrosm4_dumpdef
: Redefined M4 Macrosm4_errprint
: Redefined M4 Macrosm4_esyscmd
: Redefined M4 Macrosm4_eval
: Redefined M4 Macrosm4_exit
: Redefined M4 Macrosm4_for
: Looping constructsm4_foreach
: Looping constructsm4_foreach_w
: Looping constructsm4_format
: Redefined M4 Macrosm4_if
: Redefined M4 Macrosm4_ifdef
: Redefined M4 Macrosm4_include
: Redefined M4 Macrosm4_incr
: Redefined M4 Macrosm4_index
: Redefined M4 Macrosm4_indir
: Redefined M4 Macrosm4_len
: Redefined M4 Macrosm4_maketemp
: Redefined M4 Macrosm4_normalize
: Text processing Macrosm4_pattern_allow
: Forbidden Patternsm4_pattern_forbid
: Forbidden Patternsm4_popdef
: Redefined M4 Macrosm4_pushdef
: Redefined M4 Macrosm4_quote
: Evaluation Macrosm4_re_escape
: Text processing Macrosm4_shift
: Redefined M4 Macrosm4_sinclude
: Redefined M4 Macrosm4_split
: Text processing Macrosm4_substr
: Redefined M4 Macrosm4_syscmd
: Redefined M4 Macrosm4_sysval
: Redefined M4 Macrosm4_tolower
: Text processing Macrosm4_toupper
: Text processing Macrosm4_translit
: Redefined M4 Macrosm4_undefine
: Redefined M4 Macrosm4_wrap
: Redefined M4 MacrosThis is an alphabetical list of the Autotest macros.
AT_CAPTURE_FILE
: Writing testsuite.atAT_CHECK
: Writing testsuite.atAT_CLEANUP
: Writing testsuite.atAT_COPYRIGHT
: Writing testsuite.atAT_DATA
: Writing testsuite.atAT_INIT
: Writing testsuite.atAT_KEYWORDS
: Writing testsuite.atAT_SETUP
: Writing testsuite.atAT_TESTED
: Writing testsuite.atAT_XFAIL_IF
: Writing testsuite.atThis is an alphabetical list of the programs and functions which portability is discussed in this document.
alloca
: Particular Functionschown
: Particular Functionsclosedir
: Particular Functionserror_at_line
: Particular Functionsexit
: Function Portabilityfnmatch
: Particular Functionsfork
: Particular Functionsfree
: Function Portabilityfseeko
: Particular Functionsgetgroups
: Particular Functionsgetloadavg
: Particular Functionsgetmntent
: Particular Functionsgetpgid
: Particular Functionsgetpgrp
: Particular Functionsisinf
: Function Portabilityisnan
: Function Portabilitylstat
: Particular Functionsmalloc
: Particular Functionsmalloc
: Function Portabilitymbrtowc
: Particular Functionsmemcmp
: Particular Functionsmktime
: Particular Functionsmmap
: Particular Functionsputenv
: Function Portabilityrealloc
: Particular Functionsrealloc
: Function Portabilityselect
: Particular Functionssetpgrp
: Particular Functionssetvbuf
: Particular Functionssignal
: Function Portabilitysnprintf
: Function Portabilitysprintf
: Function Portabilitysscanf
: Function Portabilitystat
: Particular Functionsstrcoll
: Particular Functionsstrerror_r
: Particular Functionsstrerror_r
: Function Portabilitystrftime
: Particular Functionsstrnlen
: Particular Functionsstrnlen
: Function Portabilitystrtod
: Particular Functionsstrtold
: Particular Functionssysconf
: Function Portabilityunlink
: Function Portabilityunsetenv
: Function Portabilityutime
: Particular Functionsva_copy
: Function Portabilityva_list
: Function Portabilityvfork
: Particular Functionsvprintf
: Particular Functionsvsnprintf
: Function Portabilityvsprintf
: Function PortabilityThis is an alphabetical list of the files, tools, and concepts introduced in this document.
$<
, explicit rules, and VPATH
: $< in Explicit Rules_m4_divert_diversion
: New MacrosVPATH
: Automatic Rule RewritingAUTOTEST_PATH
: testsuite Invocationchangequote
: Changequote is Evildnl
: Coding Stylednl
: Macro DefinitionsVPATH
: VPATH and Double-colon$<
, and VPATH
: $< in Explicit Rulesmake -k
: make -k StatusMAKEFLAGS
: The Make Macro MAKEFLAGSSHELL
: The Make Macro SHELLMAKEFLAGS
and make: The Make Macro MAKEFLAGSVPATH
: OSF/Tru64 Directory MagicSHELL
and make: The Make Macro SHELLVPATH
: VPATH and MakeVPATH
and automatic rule rewriting: Automatic Rule RewritingVPATH
and double-colon rules: VPATH and Double-colonVPATH
and prerequisite directories: OSF/Tru64 Directory MagicVPATH
, explicit rules, and $<
: $< in Explicit RulesVPATH
, resolving target pathnames: Make Target Lookup[1] GNU Autoconf, Automake and Libtool, by G. V. Vaughan, B. Elliston, T. Tromey, and I. L. Taylor. SAMS (originally New Riders), 2000, ISBN 1578701902.
[2] Because M4 is not aware of Sh code, especially conditionals, some optimizations that look nice statically may produce incorrect results at runtime.
[3] Using
defn
.
[4] Yet another great name from Lars J. Aas.