SCons &buildversion;

%version; %scons; %builders-mod; %functions-mod; %tools-mod; %variables-mod; ]> SCons &buildversion; MAN page Steven Knight Steven Knight 2004, 2005, 2006, 2007, 2008, 2009, 2010 2004, 2005, 2006, 2007, 2008, 2009, 2010 Steven Knight version &buildversion; SCons &buildversion; MAN page SCONS 1 __MONTH_YEAR__ scons a software construction tool scons options name=val targets DESCRIPTION The scons utility builds software (or other files) by determining which component pieces must be rebuilt and executing the necessary commands to rebuild them. By default, scons searches for a file named SConstruct, Sconstruct, or sconstruct (in that order) in the current directory and reads its configuration from the first file found. An alternate file name may be specified via the option. The SConstruct file can specify subsidiary configuration files using the SConscript() function. By convention, these subsidiary files are named SConscript, although any name may be used. (Because of this naming convention, the term "SConscript files" is sometimes used to refer generically to all scons configuration files, regardless of actual file name.) The configuration files specify the target files to be built, and (optionally) the rules to build those targets. Reasonable default rules exist for building common software components (executable programs, object files, libraries), so that for most software projects, only the target and input files need be specified. Before reading the SConstruct file, scons looks for a directory named site_scons in various system directories (see below) and the directory containing the SConstruct file; for each of those dirs which exists, site_scons is prepended to sys.path, the file site_scons/site_init.py, is evaluated if it exists, and the directory site_scons/site_tools is prepended to the default toolpath if it exists. See the and options for more details. scons reads and executes the SConscript files as Python scripts, so you may use normal Python scripting capabilities (such as flow control, data manipulation, and imported Python libraries) to handle complicated build situations. scons, however, reads and executes all of the SConscript files before it begins building any targets. To make this obvious, scons prints the following messages about what it is doing: $ scons foo.out scons: Reading SConscript files ... scons: done reading SConscript files. scons: Building targets ... cp foo.in foo.out scons: done building targets. $ The status messages (everything except the line that reads "cp foo.in foo.out") may be suppressed using the option. scons does not automatically propagate the external environment used to execute scons to the commands used to build target files. This is so that builds will be guaranteed repeatable regardless of the environment variables set at the time scons is invoked. This also means that if the compiler or other commands that you want to use to build your target files are not in standard system locations, scons will not find them unless you explicitly set the PATH to include those locations. Whenever you create an scons construction environment, you can propagate the value of PATH from your external environment as follows: import os env = Environment(ENV = {'PATH' : os.environ['PATH']}) Similarly, if the commands use external environment variables like $PATH, $HOME, $JAVA_HOME, $LANG, $SHELL, $TERM, etc., these variables can also be explicitly propagated: import os env = Environment(ENV = {'PATH' : os.environ['PATH'], 'HOME' : os.environ['HOME']}) Or you may explicitly propagate the invoking user's complete external environment: import os env = Environment(ENV = os.environ) This comes at the expense of making your build dependent on the user's environment being set correctly, but it may be more convenient for many configurations. scons can scan known input files automatically for dependency information (for example, #include statements in C or C++ files) and will rebuild dependent files appropriately whenever any "included" input file changes. scons supports the ability to define new scanners for unknown input file types. scons knows how to fetch files automatically from SCCS or RCS subdirectories using SCCS, RCS or BitKeeper. scons is normally executed in a top-level directory containing a SConstruct file, optionally specifying as command-line arguments the target file or files to be built. By default, the command scons will build all target files in or below the current directory. Explicit default targets (to be built when no targets are specified on the command line) may be defined the SConscript file(s) using the Default() function, described below. Even when Default() targets are specified in the SConscript file(s), all target files in or below the current directory may be built by explicitly specifying the current directory (.) as a command-line target: scons . Building all target files, including any files outside of the current directory, may be specified by supplying a command-line target of the root directory (on POSIX systems): scons / or the path name(s) of the volume(s) in which all the targets should be built (on Windows systems): scons C:\ D:\ To build only specific targets, supply them as command-line arguments: scons foo bar in which case only the specified targets will be built (along with any derived files on which they depend). Specifying "cleanup" targets in SConscript files is not usually necessary. The flag removes all files necessary to build the specified target: scons -c . to remove all target files, or: scons -c build export to remove target files under build and export. Additional files or directories to remove can be specified using the Clean() function. Conversely, targets that would normally be removed by the invocation can be prevented from being removed by using the NoClean() function. A subset of a hierarchical tree may be built by remaining at the top-level directory (where the SConstruct file lives) and specifying the subdirectory as the target to be built: scons src/subdir or by changing directory and invoking scons with the option, which traverses up the directory hierarchy until it finds the SConstruct file, and then builds targets relatively to the current subdirectory: cd src/subdir scons -u . scons supports building multiple targets in parallel via a option that takes, as its argument, the number of simultaneous tasks that may be spawned: scons -j 4 builds four targets in parallel, for example. scons can maintain a cache of target (derived) files that can be shared between multiple builds. When caching is enabled in a SConscript file, any target files built by scons will be copied to the cache. If an up-to-date target file is found in the cache, it will be retrieved from the cache instead of being rebuilt locally. Caching behavior may be disabled and controlled in other ways by the , , and command-line options. The option is useful to prevent multiple builds from trying to update the cache simultaneously. Values of variables to be passed to the SConscript file(s) may be specified on the command line: scons debug=1 . These variables are available in SConscript files through the ARGUMENTS dictionary, and can be used in the SConscript file(s) to modify the build in any way: if ARGUMENTS.get('debug', 0): env = Environment(CCFLAGS = '-g') else: env = Environment() The command-line variable arguments are also available in the ARGLIST list, indexed by their order on the command line. This allows you to process them in order rather than by name, if necessary. ARGLIST[0] returns a tuple containing (argname, argvalue). A Python exception is thrown if you try to access a list member that does not exist. scons requires Python version 2.4 or later. There should be no other dependencies or requirements to run scons. By default, scons knows how to search for available programming tools on various systems. On Windows systems, scons searches in order for the Microsoft Visual C++ tools, the MinGW tool chain, the Intel compiler tools, and the PharLap ETS compiler. On OS/2 systems, scons searches in order for the OS/2 compiler, the GCC tool chain, and the Microsoft Visual C++ tools, On SGI IRIX, IBM AIX, Hewlett Packard HP-UX, and Sun Solaris systems, scons searches for the native compiler tools (MIPSpro, Visual Age, aCC, and Forte tools respectively) and the GCC tool chain. On all other platforms, including POSIX (Linux and UNIX) platforms, scons searches in order for the GCC tool chain, the Microsoft Visual C++ tools, and the Intel compiler tools. You may, of course, override these default values by appropriate configuration of Environment construction variables. OPTIONS In general, scons supports the same command-line options as GNU make, and many of those supported by cons. -b Ignored for compatibility with non-GNU versions of make. -c, --clean, --remove Clean up by removing all target files for which a construction command is specified. Also remove any files or directories associated to the construction command using the Clean() function. Will not remove any targets specified by the NoClean() function. --cache-debug=file Print debug information about the CacheDir() derived-file caching to the specified file. If file is - (a hyphen), the debug information are printed to the standard output. The printed messages describe what signature file names are being looked for in, retrieved from, or written to the CacheDir() directory tree. --cache-disable, --no-cache Disable the derived-file caching specified by CacheDir(). scons will neither retrieve files from the cache nor copy files to the cache. --cache-force, --cache-populate When using CacheDir(), populate a cache by copying any already-existing, up-to-date derived files to the cache, in addition to files built by this invocation. This is useful to populate a new cache with all the current derived files, or to add to the cache any derived files recently built with caching disabled via the option. --cache-show When using CacheDir() and retrieving a derived file from the cache, show the command that would have been executed to build the file, instead of the usual report, "Retrieved `file' from cache." This will produce consistent output for build logs, regardless of whether a target file was rebuilt or retrieved from the cache. --config=mode This specifies how the Configure call should use or generate the results of configuration tests. The option should be specified from among the following choices: --config=auto scons will use its normal dependency mechanisms to decide if a test must be rebuilt or not. This saves time by not running the same configuration tests every time you invoke scons, but will overlook changes in system header files or external commands (such as compilers) if you don't specify those dependecies explicitly. This is the default behavior. --config=force If this option is specified, all configuration tests will be re-run regardless of whether the cached results are out of date. This can be used to explicitly force the configuration tests to be updated in response to an otherwise unconfigured change in a system header file or compiler. --config=cache If this option is specified, no configuration tests will be rerun and all results will be taken from cache. Note that scons will still consider it an error if --config=cache is specified and a necessary test does not yet have any results in the cache. -C directory, --directory=directory Change to the specified directory before searching for the SConstruct, Sconstruct, or sconstruct file, or doing anything else. Multiple options are interpreted relative to the previous one, and the right-most option wins. (This option is nearly equivalent to , except that it will search for SConstruct, Sconstruct, or sconstruct in the specified directory.) -D Works exactly the same way as the option except for the way default targets are handled. When this option is used and no targets are specified on the command line, all default targets are built, whether or not they are below the current directory. --debug=type Debug the build process. type specifies what type of debugging: --debug=count Print how many objects are created of the various classes used internally by SCons before and after reading the SConscript files and before and after building targets. This is not supported when SCons is executed with the Python (optimized) option or when the SCons modules have been compiled with optimization (that is, when executing from *.pyo files). --debug=duplicate Print a line for each unlink/relink (or copy) of a variant file from its source file. Includes debugging info for unlinking stale variant files, as well as unlinking old targets before building them. --debug=dtree A synonym for the newer option. This will be deprecated in some future release and ultimately removed. --debug=explain Print an explanation of precisely why scons is deciding to (re-)build any targets. (Note: this does not print anything for targets that are not rebuilt.) --debug=findlibs Instruct the scanner that searches for libraries to print a message about each potential library name it is searching for, and about the actual libraries it finds. --debug=includes Print the include tree after each top-level target is built. This is generally used to find out what files are included by the sources of a given derived file: $ scons --debug=includes foo.o --debug=memoizer Prints a summary of hits and misses using the Memoizer, an internal subsystem that counts how often SCons uses cached values in memory instead of recomputing them each time they're needed. --debug=memory Prints how much memory SCons uses before and after reading the SConscript files and before and after building targets. --debug=nomemoizer A deprecated option preserved for backwards compatibility. --debug=objects Prints a list of the various objects of the various classes used internally by SCons. --debug=pdb Re-run SCons under the control of the pdb Python debugger. --debug=prepare Print a line each time any target (internal or external) is prepared for building. scons prints this for each target it considers, even if that target is up to date (see also --debug=explain). This can help debug problems with targets that aren't being built; it shows whether scons is at least considering them or not. --debug=presub Print the raw command line used to build each target before the construction environment variables are substituted. Also shows which targets are being built by this command. Output looks something like this: $ scons --debug=presub Building myprog.o with action(s): $SHCC $SHCFLAGS $SHCCFLAGS $CPPFLAGS $_CPPINCFLAGS -c -o $TARGET $SOURCES ... --debug=stacktrace Prints an internal Python stack trace when encountering an otherwise unexplained error. --debug=stree A synonym for the newer option. This will be deprecated in some future release and ultimately removed. --debug=time Prints various time profiling information: the time spent executing each individual build command; the total build time (time SCons ran from beginning to end); the total time spent reading and executing SConscript files; the total time spent SCons itself spend running (that is, not counting reading and executing SConscript files); and both the total time spent executing all build commands and the elapsed wall-clock time spent executing those build commands. (When scons is executed without the option, the elapsed wall-clock time will typically be slightly longer than the total time spent executing all the build commands, due to the SCons processing that takes place in between executing each command. When scons is executed with the option, and your build configuration allows good parallelization, the elapsed wall-clock time should be significantly smaller than the total time spent executing all the build commands, since multiple build commands and intervening SCons processing should take place in parallel.) --debug=tree A synonym for the newer option. This will be deprecated in some future release and ultimately removed. --diskcheck=types Enable specific checks for whether or not there is a file on disk where the SCons configuration expects a directory (or vice versa), and whether or not RCS or SCCS sources exist when searching for source and include files. The types argument can be set to: all, to enable all checks explicitly (the default behavior); none, to disable all such checks; match, to check that files and directories on disk match SCons' expected configuration; rcs, to check for the existence of an RCS source for any missing source or include files; sccs, to check for the existence of an SCCS source for any missing source or include files. Multiple checks can be specified separated by commas; for example, would still check for SCCS and RCS sources, but disable the check for on-disk matches of files and directories. Disabling some or all of these checks can provide a performance boost for large configurations, or when the configuration will check for files and/or directories across networked or shared file systems, at the slight increased risk of an incorrect build or of not handling errors gracefully (if include files really should be found in SCCS or RCS, for example, or if a file really does exist where the SCons configuration expects a directory). --duplicate=ORDER There are three ways to duplicate files in a build tree: hard links, soft (symbolic) links and copies. The default behaviour of SCons is to prefer hard links to soft links to copies. You can specify different behaviours with this option. ORDER must be one of hard-soft-copy (the default), soft-hard-copy, hard-copy, soft-copy or copy. SCons will attempt to duplicate files using the mechanisms in the specified order. -f file, --file=file, --makefile=file, --sconstruct=file Use file as the initial SConscript file. Multiple options may be specified, in which case scons will read all of the specified files. -h, --help Print a local help message for this build, if one is defined in the SConscript file(s), plus a line that describes the option for command-line option help. If no local help message is defined, prints the standard help message about command-line options. Exits after displaying the appropriate message. -H, --help-options Print the standard help message about command-line options and exit. -i, --ignore-errors Ignore all errors from commands executed to rebuild files. -I directory, --include-dir=directory Specifies a directory to search for imported Python modules. If several options are used, the directories are searched in the order specified. --implicit-cache Cache implicit dependencies. This causes scons to use the implicit (scanned) dependencies from the last time it was run instead of scanning the files for implicit dependencies. This can significantly speed up SCons, but with the following limitations: scons will not detect changes to implicit dependency search paths (e.g. CPPPATH, LIBPATH) that would ordinarily cause different versions of same-named files to be used. scons will miss changes in the implicit dependencies in cases where a new implicit dependency is added earlier in the implicit dependency search path (e.g. CPPPATH, LIBPATH) than a current implicit dependency with the same name. --implicit-deps-changed Forces SCons to ignore the cached implicit dependencies. This causes the implicit dependencies to be rescanned and recached. This implies . --implicit-deps-unchanged Force SCons to ignore changes in the implicit dependencies. This causes cached implicit dependencies to always be used. This implies . --interactive Starts SCons in interactive mode. The SConscript files are read once and a scons>>> prompt is printed. Targets may now be rebuilt by typing commands at interactive prompt without having to re-read the SConscript files and re-initialize the dependency graph from scratch. SCons interactive mode supports the following commands:

build[OPTIONS] [TARGETS] ... Builds the specified TARGETS (and their dependencies) with the specified SCons command-line OPTIONS. b and scons are synonyms. The following SCons command-line options affect the build command: --cache-debug=FILE --cache-disable, --no-cache --cache-force, --cache-populate --cache-show --debug=TYPE -i, --ignore-errors -j N, --jobs=N -k, --keep-going -n, --no-exec, --just-print, --dry-run, --recon -Q -s, --silent, --quiet --taskmastertrace=FILE --tree=OPTIONS Any other SCons command-line options that are specified do not cause errors but have no effect on the build command (mainly because they affect how the SConscript files are read, which only happens once at the beginning of interactive mode). clean[OPTIONS] [TARGETS] ... Cleans the specified TARGETS (and their dependencies) with the specified options. c is a synonym. This command is itself a synonym for build --clean exit Exits SCons interactive mode. You can also exit by terminating input (CTRL+D on UNIX or Linux systems, CTRL+Z on Windows systems). help[COMMAND] Provides a help message about the commands available in SCons interactive mode. If COMMAND is specified, h and ? are synonyms. shell[COMMANDLINE] Executes the specified COMMANDLINE in a subshell. If no COMMANDLINE is specified, executes the interactive command interpreter specified in the SHELL environment variable (on UNIX and Linux systems) or the COMSPEC environment variable (on Windows systems). sh and ! are synonyms. version Prints SCons version information.

An empty line repeats the last typed command. Command-line editing can be used if the readline module is available. $ scons --interactive scons: Reading SConscript files ... scons: done reading SConscript files. scons>>> build -n prog scons>>> exit -j N, --jobs=N Specifies the number of jobs (commands) to run simultaneously. If there is more than one option, the last one is effective. -k, --keep-going Continue as much as possible after an error. The target that failed and those that depend on it will not be remade, but other targets specified on the command line will still be processed. -m Ignored for compatibility with non-GNU versions of make. --max-drift=SECONDS Set the maximum expected drift in the modification time of files to SECONDS. This value determines how long a file must be unmodified before its cached content signature will be used instead of calculating a new content signature (MD5 checksum) of the file's contents. The default value is 2 days, which means a file must have a modification time of at least two days ago in order to have its cached content signature used. A negative value means to never cache the content signature and to ignore the cached value if there already is one. A value of 0 means to always use the cached signature, no matter how old the file is. --md5-chunksize=KILOBYTES Set the block size used to compute MD5 signatures to KILOBYTES. This value determines the size of the chunks which are read in at once when computing MD5 signatures. Files below that size are fully stored in memory before performing the signature computation while bigger files are read in block-by-block. A huge block-size leads to high memory consumption while a very small block-size slows down the build considerably. The default value is to use a chunk size of 64 kilobytes, which should be appropriate for most uses. -n, --just-print, --dry-run, --recon No execute. Print the commands that would be executed to build any out-of-date target files, but do not execute the commands. --no-site-dir Prevents the automatic addition of the standard site_scons dirs to sys.path. Also prevents loading the site_scons/site_init.py modules if they exist, and prevents adding their site_scons/site_tools dirs to the toolpath. --profile=file Run SCons under the Python profiler and save the results in the specified file. The results may be analyzed using the Python pstats module. -q, --question Do not run any commands, or print anything. Just return an exit status that is zero if the specified targets are already up to date, non-zero otherwise. -Q Quiets SCons status messages about reading SConscript files, building targets and entering directories. Commands that are executed to rebuild target files are still printed. --random Build dependencies in a random order. This is useful when building multiple trees simultaneously with caching enabled, to prevent multiple builds from simultaneously trying to build or retrieve the same target files. -s, --silent, --quiet Silent. Do not print commands that are executed to rebuild target files. Also suppresses SCons status messages. -S, --no-keep-going, --stop Ignored for compatibility with GNU make. --site-dir=dir Uses the named dir as the site dir rather than the default site_scons dirs. This dir will get prepended to sys.path, the module dir/site_init.py will get loaded if it exists, and dir/site_tools will get added to the default toolpath. The default set of site_scons dirs used when is not specified depends on the system platform, as follows. Note that the directories are examined in the order given, from most generic to most specific, so the last-executed site_init.py file is the most specific one (which gives it the chance to override everything else), and the dirs are prepended to the paths, again so the last dir examined comes first in the resulting path. Windows: %ALLUSERSPROFILE/Application Data/scons/site_scons %USERPROFILE%/Local Settings/Application Data/scons/site_scons %APPDATA%/scons/site_scons %HOME%/.scons/site_scons ./site_scons Mac OS X: /Library/Application Support/SCons/site_scons /opt/local/share/scons/site_scons (for MacPorts) /sw/share/scons/site_scons (for Fink) $HOME/Library/Application Support/SCons/site_scons $HOME/.scons/site_scons ./site_scons Solaris: /opt/sfw/scons/site_scons /usr/share/scons/site_scons $HOME/.scons/site_scons ./site_scons Linux, HPUX, and other Posix-like systems: /usr/share/scons/site_scons $HOME/.scons/site_scons ./site_scons --stack-size=KILOBYTES Set the size stack used to run threads to KILOBYTES. This value determines the stack size of the threads used to run jobs. These are the threads that execute the actions of the builders for the nodes that are out-of-date. Note that this option has no effect unless the num_jobs option, which corresponds to -j and --jobs, is larger than one. Using a stack size that is too small may cause stack overflow errors. This usually shows up as segmentation faults that cause scons to abort before building anything. Using a stack size that is too large will cause scons to use more memory than required and may slow down the entire build process. The default value is to use a stack size of 256 kilobytes, which should be appropriate for most uses. You should not need to increase this value unless you encounter stack overflow errors. -t, --touch Ignored for compatibility with GNU make. (Touching a file to make it appear up-to-date is unnecessary when using scons.) --taskmastertrace=file Prints trace information to the specified file about how the internal Taskmaster object evaluates and controls the order in which Nodes are built. A file name of - may be used to specify the standard output. -tree=options Prints a tree of the dependencies after each top-level target is built. This prints out some or all of the tree, in various formats, depending on the options specified: --tree=all Print the entire dependency tree after each top-level target is built. This prints out the complete dependency tree, including implicit dependencies and ignored dependencies. --tree=derived Restricts the tree output to only derived (target) files, not source files. --tree=status Prints status information for each displayed node. --tree=prune Prunes the tree to avoid repeating dependency information for nodes that have already been displayed. Any node that has already been displayed will have its name printed in [square brackets], as an indication that the dependencies for that node can be found by searching for the relevant output higher up in the tree. Multiple options may be specified, separated by commas: # Prints only derived files, with status information: scons --tree=derived,status # Prints all dependencies of target, with status information # and pruning dependencies of already-visited Nodes: scons --tree=all,prune,status target -u, --up, --search-up Walks up the directory structure until an SConstruct , Sconstruct or sconstruct file is found, and uses that as the top of the directory tree. If no targets are specified on the command line, only targets at or below the current directory will be built. -U Works exactly the same way as the option except for the way default targets are handled. When this option is used and no targets are specified on the command line, all default targets that are defined in the SConscript(s) in the current directory are built, regardless of what directory the resultant targets end up in. -v, --version Print the scons version, copyright information, list of authors, and any other relevant information. Then exit. -w, --print-directory Print a message containing the working directory before and after other processing. --no-print-directory Turn off -w, even if it was turned on implicitly. --warn=type, --warn=no-type Enable or disable warnings. type specifies the type of warnings to be enabled or disabled: --warn=all, --warn=no-all Enables or disables all warnings. --warn=cache-write-error, --warn=no-cache-write-error Enables or disables warnings about errors trying to write a copy of a built file to a specified CacheDir(). These warnings are disabled by default. --warn=corrupt-sconsign, --warn=no-corrupt-sconsign Enables or disables warnings about unfamiliar signature data in .sconsign files. These warnings are enabled by default. --warn=dependency, --warn=no-dependency Enables or disables warnings about dependencies. These warnings are disabled by default. --warn=deprecated, --warn=no-deprecated Enables or disables all warnings about use of currently deprecated features. These warnings are enabled by default. Note that the option does not disable warnings about absolutely all deprecated features. Warnings for some deprecated features that have already been through several releases with deprecation warnings may be mandatory for a release or two before they are officially no longer supported by SCons. Warnings for some specific deprecated features may be enabled or disabled individually; see below.

--warn=deprecated-copy, --warn=no-deprecated-copy Enables or disables warnings about use of the deprecated env.Copy() method. --warn=deprecated-source-signatures, --warn=no-deprecated-source-signatures Enables or disables warnings about use of the deprecated SourceSignatures() function or env.SourceSignatures() method. --warn=deprecated-target-signatures, --warn=no-deprecated-target-signatures Enables or disables warnings about use of the deprecated TargetSignatures() function or env.TargetSignatures() method.

--warn=duplicate-environment, --warn=no-duplicate-environment Enables or disables warnings about attempts to specify a build of a target with two different construction environments that use the same action. These warnings are enabled by default. --warn=fortran-cxx-mix, --warn=no-fortran-cxx-mix Enables or disables the specific warning about linking Fortran and C++ object files in a single executable, which can yield unpredictable behavior with some compilers. --warn=future-deprecated, --warn=no-future-deprecated Enables or disables warnings about features that will be deprecated in the future. These warnings are disabled by default. Enabling this warning is especially recommended for projects that redistribute SCons configurations for other users to build, so that the project can be warned as soon as possible about to-be-deprecated features that may require changes to the configuration. --warn=link, --warn=no-link Enables or disables warnings about link steps. --warn=misleading-keywords, --warn=no-misleading-keywords Enables or disables warnings about use of the misspelled keywords targets and sources when calling Builders. (Note the last s characters, the correct spellings are target and source.) These warnings are enabled by default. --warn=missing-sconscript, --warn=no-missing-sconscript Enables or disables warnings about missing SConscript files. These warnings are enabled by default. --warn=no-md5-module, --warn=no-no-md5-module Enables or disables warnings about the version of Python not having an MD5 checksum module available. These warnings are enabled by default. --warn=no-metaclass-support, --warn=no-no-metaclass-support Enables or disables warnings about the version of Python not supporting metaclasses when the option is used. These warnings are enabled by default. --warn=no-object-count, --warn=no-no-object-count Enables or disables warnings about the feature not working when scons is run with the python option or from optimized Python (.pyo) modules. --warn=no-parallel-support, --warn=no-no-parallel-support Enables or disables warnings about the version of Python not being able to support parallel builds when the option is used. These warnings are enabled by default. --warn=python-version, --warn=no-python-version Enables or disables the warning about running SCons with a deprecated version of Python. These warnings are enabled by default. --warn=reserved-variable, --warn=no-reserved-variable Enables or disables warnings about attempts to set the reserved construction variable names CHANGED_SOURCES, CHANGED_TARGETS, TARGET, TARGETS, SOURCE, SOURCES, UNCHANGED_SOURCES or UNCHANGED_TARGETS. These warnings are disabled by default. --warn=stack-size, --warn=no-stack-size Enables or disables warnings about requests to set the stack size that could not be honored. These warnings are enabled by default. -Y repository, --repository=repository, --srcdir=repository Search the specified repository for any input and target files not found in the local directory hierarchy. Multiple options may be specified, in which case the repositories are searched in the order specified. CONFIGURATION FILE REFERENCE Construction Environments A construction environment is the basic means by which the SConscript files communicate build information to scons. A new construction environment is created using the Environment function: env = Environment() Variables, called construction variables, may be set in a construction environment either by specifying them as keywords when the object is created or by assigning them a value after the object is created: env = Environment(FOO = 'foo') env['BAR'] = 'bar' As a convenience, construction variables may also be set or modified by the parse_flags keyword argument, which applies the ParseFlags method (described below) to the argument value after all other processing is completed. This is useful either if the exact content of the flags is unknown (for example, read from a control file) or if the flags are distributed to a number of construction variables. env = Environment(parse_flags = '-Iinclude -DEBUG -lm') This example adds 'include' to CPPPATH, 'EBUG' to CPPDEFINES, and 'm' to LIBS. By default, a new construction environment is initialized with a set of builder methods and construction variables that are appropriate for the current platform. An optional platform keyword argument may be used to specify that an environment should be initialized for a different platform: env = Environment(platform = 'cygwin') env = Environment(platform = 'os2') env = Environment(platform = 'posix') env = Environment(platform = 'win32') Specifying a platform initializes the appropriate construction variables in the environment to use and generate file names with prefixes and suffixes appropriate for the platform. Note that the win32 platform adds the SystemDrive and SystemRoot variables from the user's external environment to the construction environment's ENV dictionary. This is so that any executed commands that use sockets to connect with other systems (such as fetching source files from external CVS repository specifications like :pserver:anonymous@cvs.sourceforge.net:/cvsroot/scons) will work on Windows systems. The platform argument may be function or callable object, in which case the Environment() method will call the specified argument to update the new construction environment: def my_platform(env): env['VAR'] = 'xyzzy' env = Environment(platform = my_platform) Additionally, a specific set of tools with which to initialize the environment may be specified as an optional keyword argument: env = Environment(tools = ['msvc', 'lex']) Non-built-in tools may be specified using the toolpath argument: env = Environment(tools = ['default', 'foo'], toolpath = ['tools']) This looks for a tool specification in tools/foo.py (as well as using the ordinary default tools for the platform). foo.py should have two functions: generate(env, **kw) and exists(env). The generate() function modifies the passed-in environment to set up variables so that the tool can be executed; it may use any keyword arguments that the user supplies (see below) to vary its initialization. The exists() function should return a true value if the tool is available. Tools in the toolpath are used before any of the built-in ones. For example, adding gcc.py to the toolpath would override the built-in gcc tool. Also note that the toolpath is stored in the environment for use by later calls to Clone() and Tool() methods: base = Environment(toolpath=['custom_path']) derived = base.Clone(tools=['custom_tool']) derived.CustomBuilder() The elements of the tools list may also be functions or callable objects, in which case the Environment() method will call the specified elements to update the new construction environment: def my_tool(env): env['XYZZY'] = 'xyzzy' env = Environment(tools = [my_tool]) The individual elements of the tools list may also themselves be two-element lists of the form (toolname, kw_dict). SCons searches for the toolname specification file as described above, and passes kw_dict, which must be a dictionary, as keyword arguments to the tool's generate function. The generate function can use the arguments to modify the tool's behavior by setting up the environment in different ways or otherwise changing its initialization. # in tools/my_tool.py: def generate(env, **kw): # Sets MY_TOOL to the value of keyword argument 'arg1' or 1. env['MY_TOOL'] = kw.get('arg1', '1') def exists(env): return 1 # in SConstruct: env = Environment(tools = ['default', ('my_tool', {'arg1': 'abc'})], toolpath=['tools']) The tool definition (i.e. my_tool()) can use the PLATFORM variable from the environment it receives to customize the tool for different platforms. If no tool list is specified, then SCons will auto-detect the installed tools using the PATH variable in the ENV construction variable and the platform name when the Environment is constructed. Changing the PATH variable after the Environment is constructed will not cause the tools to be redetected. SCons supports the following tool specifications out of the box: Additionally, there is a "tool" named default which configures the environment with a default set of tools for the current platform. On posix and cygwin platforms the GNU tools (e.g. gcc) are preferred by SCons, on Windows the Microsoft tools (e.g. msvc) followed by MinGW are preferred by SCons, and in OS/2 the IBM tools (e.g. icc) are preferred by SCons. Builder Methods Build rules are specified by calling a construction environment's builder methods. The arguments to the builder methods are target (a list of targets to be built, usually file names) and source (a list of sources to be built, usually file names). Because long lists of file names can lead to a lot of quoting, scons supplies a Split() global function and a same-named environment method that split a single string into a list, separated on strings of white-space characters. (These are similar to the split() member function of Python strings but work even if the input isn't a string.) Like all Python arguments, the target and source arguments to a builder method can be specified either with or without the "target" and "source" keywords. When the keywords are omitted, the target is first, followed by the source. The following are equivalent examples of calling the Program builder method: env.Program('bar', ['bar.c', 'foo.c']) env.Program('bar', Split('bar.c foo.c')) env.Program('bar', env.Split('bar.c foo.c')) env.Program(source = ['bar.c', 'foo.c'], target = 'bar') env.Program(target = 'bar', Split('bar.c foo.c')) env.Program(target = 'bar', env.Split('bar.c foo.c')) env.Program('bar', source = 'bar.c foo.c'.split()) Target and source file names that are not absolute path names (that is, do not begin with / on POSIX systems or \fR on Windows systems, with or without an optional drive letter) are interpreted relative to the directory containing the SConscript file being read. An initial # (hash mark) on a path name means that the rest of the file name is interpreted relative to the directory containing the top-level SConstruct file, even if the # is followed by a directory separator character (slash or backslash). Examples: # The comments describing the targets that will be built # assume these calls are in a SConscript file in the # a subdirectory named "subdir". # Builds the program "subdir/foo" from "subdir/foo.c": env.Program('foo', 'foo.c') # Builds the program "/tmp/bar" from "subdir/bar.c": env.Program('/tmp/bar', 'bar.c') # An initial '#' or '#/' are equivalent; the following # calls build the programs "foo" and "bar" (in the # top-level SConstruct directory) from "subdir/foo.c" and # "subdir/bar.c", respectively: env.Program('#foo', 'foo.c') env.Program('#/bar', 'bar.c') # Builds the program "other/foo" (relative to the top-level # SConstruct directory) from "subdir/foo.c": env.Program('#other/foo', 'foo.c') When the target shares the same base name as the source and only the suffix varies, and if the builder method has a suffix defined for the target file type, then the target argument may be omitted completely, and scons will deduce the target file name from the source file name. The following examples all build the executable program bar (on POSIX systems) or bar.exe (on Windows systems) from the bar.c source file: env.Program(target = 'bar', source = 'bar.c') env.Program('bar', source = 'bar.c') env.Program(source = 'bar.c') env.Program('bar.c') As a convenience, a srcdir keyword argument may be specified when calling a Builder. When specified, all source file strings that are not absolute paths will be interpreted relative to the specified srcdir. The following example will build the build/prog (or build/prog.exe on Windows) program from the files src/f1.c and src/f2.c: env.Program('build/prog', ['f1.c', 'f2.c'], srcdir='src') It is possible to override or add construction variables when calling a builder method by passing additional keyword arguments. These overridden or added variables will only be in effect when building the target, so they will not affect other parts of the build. For example, if you want to add additional libraries for just one program: env.Program('hello', 'hello.c', LIBS=['gl', 'glut']) or generate a shared library with a non-standard suffix: env.SharedLibrary('word', 'word.cpp', SHLIBSUFFIX='.ocx', LIBSUFFIXES=['.ocx']) (Note that both the $SHLIBSUFFIX and $LIBSUFFIXES variables must be set if you want SCons to search automatically for dependencies on the non-standard library names; see the descriptions of these variables, below, for more information.) It is also possible to use the parse_flags keyword argument in an override: env = Program('hello', 'hello.c', parse_flags = '-Iinclude -DEBUG -lm') This example adds 'include' to CPPPATH, 'EBUG' to CPPDEFINES, and 'm' to LIBS. Although the builder methods defined by scons are, in fact, methods of a construction environment object, they may also be called without an explicit environment: Program('hello', 'hello.c') SharedLibrary('word', 'word.cpp') In this case, the methods are called internally using a default construction environment that consists of the tools and values that scons has determined are appropriate for the local system. Builder methods that can be called without an explicit environment may be called from custom Python modules that you import into an SConscript file by adding the following to the Python module: from SCons.Script import * All builder methods return a list-like object containing Nodes that represent the target or targets that will be built. A Node is an internal SCons object which represents build targets or sources. The returned Node-list object can be passed to other builder methods as source(s) or passed to any SCons function or method where a filename would normally be accepted. For example, if it were necessary to add a specific flag when compiling one specific object file: bar_obj_list = env.StaticObject('bar.c', CPPDEFINES='-DBAR') env.Program(source = ['foo.c', bar_obj_list, 'main.c']) Using a Node in this way makes for a more portable build by avoiding having to specify a platform-specific object suffix when calling the Program() builder method. Note that Builder calls will automatically "flatten" the source and target file lists, so it's all right to have the bar_obj list return by the StaticObject() call in the middle of the source file list. If you need to manipulate a list of lists returned by Builders directly using Python, you can either build the list by hand: foo = Object('foo.c') bar = Object('bar.c') objects = ['begin.o'] + foo + ['middle.o'] + bar + ['end.o'] for object in objects: print str(object) Or you can use the Flatten() function supplied by scons to create a list containing just the Nodes, which may be more convenient: foo = Object('foo.c') bar = Object('bar.c') objects = Flatten(['begin.o', foo, 'middle.o', bar, 'end.o']) for object in objects: print str(object) Note also that because Builder calls return a list-like object, not an actual Python list, you should not use the Python += operator to append Builder results to a Python list. Because the list and the object are different types, Python will not update the original list in place, but will instead create a new Node-list object containing the concatenation of the list elements and the Builder results. This will cause problems for any other Python variables in your SCons configuration that still hold on to a reference to the original list. Instead, use the Python .extend() method to make sure the list is updated in-place. Example: object_files = [] # Do NOT use += as follows: # # object_files += Object('bar.c') # # It will not update the object_files list in place. # # Instead, use the .extend() method: object_files.extend(Object('bar.c')) The path name for a Node's file may be used by passing the Node to the Python-builtin str() function: bar_obj_list = env.StaticObject('bar.c', CPPDEFINES='-DBAR') print "The path to bar_obj is:", str(bar_obj_list[0]) Note again that because the Builder call returns a list, we have to access the first element in the list (bar_obj_list[0]) to get at the Node that actually represents the object file. Builder calls support a chdir keyword argument that specifies that the Builder's action(s) should be executed after changing directory. If the chdir argument is a string or a directory Node, scons will change to the specified directory. If the chdir is not a string or Node and is non-zero, then scons will change to the target file's directory. # scons will change to the "sub" subdirectory # before executing the "cp" command. env.Command('sub/dir/foo.out', 'sub/dir/foo.in', "cp dir/foo.in dir/foo.out", chdir='sub') # Because chdir is not a string, scons will change to the # target's directory ("sub/dir") before executing the # "cp" command. env.Command('sub/dir/foo.out', 'sub/dir/foo.in', "cp foo.in foo.out", chdir=1) Note that scons will not automatically modify its expansion of construction variables like $TARGET and $SOURCE when using the chdir keyword argument--that is, the expanded file names will still be relative to the top-level SConstruct directory, and consequently incorrect relative to the chdir directory. If you use the chdir keyword argument, you will typically need to supply a different command line using expansions like ${TARGET.file} and ${SOURCE.file} to use just the filename portion of the targets and source. scons provides the following builder methods: All targets of builder methods automatically depend on their sources. An explicit dependency can be specified using the Depends method of a construction environment (see below). In addition, scons automatically scans source files for various programming languages, so the dependencies do not need to be specified explicitly. By default, SCons can C source files, C++ source files, Fortran source files with .F (POSIX systems only), .fpp, or .FPP file extensions, and assembly language files with .S (POSIX systems only), .spp, or .SPP files extensions for C preprocessor dependencies. SCons also has default support for scanning D source files, You can also write your own Scanners to add support for additional source file types. These can be added to the default Scanner object used by the Object(), StaticObject(), and SharedObject() Builders by adding them to the SourceFileScanner object. See the section "Scanner Objects" below, for more information about defining your own Scanner objects and using the SourceFileScanner object. Methods and Functions to Do Things In addition to Builder methods, scons provides a number of other construction environment methods and global functions to manipulate the build configuration. Usually, a construction environment method and global function with the same name both exist so that you don't have to remember whether to a specific bit of functionality must be called with or without a construction environment. In the following list, if you call something as a global function it looks like: Function(arguments) and if you call something through a construction environment it looks like: env.Function(arguments) If you can call the functionality in both ways, then both forms are listed. Global functions may be called from custom Python modules that you import into an SConscript file by adding the following to the Python module: from SCons.Script import * Except where otherwise noted, the same-named construction environment method and global function provide the exact same functionality. The only difference is that, where appropriate, calling the functionality through a construction environment will substitute construction variables into any supplied strings. For example: env = Environment(FOO = 'foo') Default('$FOO') env.Default('$FOO') In the above example, the first call to the global Default() function will actually add a target named $FOO to the list of default targets, while the second call to the env.Default() construction environment method will expand the value and add a target named foo to the list of default targets. For more on construction variable expansion, see the next section on construction variables. Construction environment methods and global functions supported by scons include: SConscript Variables In addition to the global functions and methods, scons supports a number of Python variables that can be used in SConscript files to affect how you want the build to be performed. These variables may be accessed from custom Python modules that you import into an SConscript file by adding the following to the Python module: from SCons.Script import * ARGLIST A list keyword=value arguments specified on the command line. Each element in the list is a tuple containing the (keyword,value) of the argument. The separate keyword and value elements of the tuple can be accessed by subscripting for element [0] and [1] of the tuple, respectively. Example: print "first keyword, value =", ARGLIST[0][0], ARGLIST[0][1] print "second keyword, value =", ARGLIST[1][0], ARGLIST[1][1] third_tuple = ARGLIST[2] print "third keyword, value =", third_tuple[0], third_tuple[1] for key, value in ARGLIST: # process key and value ARGUMENTS A dictionary of all the keyword=value arguments specified on the command line. The dictionary is not in order, and if a given keyword has more than one value assigned to it on the command line, the last (right-most) value is the one in the ARGUMENTS dictionary. Example: if ARGUMENTS.get('debug', 0): env = Environment(CCFLAGS = '-g') else: env = Environment() BUILD_TARGETS A list of the targets which scons will actually try to build, regardless of whether they were specified on the command line or via the Default() function or method. The elements of this list may be strings or nodes, so you should run the list through the Python str function to make sure any Node path names are converted to strings. Because this list may be taken from the list of targets specified using the Default() function or method, the contents of the list may change on each successive call to Default(). See the DEFAULT_TARGETS list, below, for additional information. Example: if 'foo' in BUILD_TARGETS: print "Don't forget to test the `foo' program!" if 'special/program' in BUILD_TARGETS: SConscript('special') Note that the BUILD_TARGETS list only contains targets expected listed on the command line or via calls to the Default() function or method. It does not contain all dependent targets that will be built as a result of making the sure the explicitly-specified targets are up to date. COMMAND_LINE_TARGETS A list of the targets explicitly specified on the command line. If there are no targets specified on the command line, the list is empty. This can be used, for example, to take specific actions only when a certain target or targets is explicitly being built. Example: if 'foo' in COMMAND_LINE_TARGETS: print "Don't forget to test the `foo' program!" if 'special/program' in COMMAND_LINE_TARGETS: SConscript('special') DEFAULT_TARGETS A list of the target nodes that have been specified using the Default() function or method. The elements of the list are nodes, so you need to run them through the Python str function to get at the path name for each Node. Example: print str(DEFAULT_TARGETS[0]) if 'foo' in map(str, DEFAULT_TARGETS): print "Don't forget to test the `foo' program!" The contents of the DEFAULT_TARGETS list change on on each successive call to the Default() function: print map(str, DEFAULT_TARGETS) # originally [] Default('foo') print map(str, DEFAULT_TARGETS) # now a node ['foo'] Default('bar') print map(str, DEFAULT_TARGETS) # now a node ['foo', 'bar'] Default(None) print map(str, DEFAULT_TARGETS) # back to [] Consequently, be sure to use DEFAULT_TARGETS only after you've made all of your Default() calls, or else simply be careful of the order of these statements in your SConscript files so that you don't look for a specific default target before it's actually been added to the list. Construction Variables A construction environment has an associated dictionary of construction variables that are used by built-in or user-supplied build rules. Construction variables must follow the same rules for Python identifiers: the initial character must be an underscore or letter, followed by any number of underscores, letters, or digits. A number of useful construction variables are automatically defined by scons for each supported platform, and additional construction variables can be defined by the user. The following is a list of the automatically defined construction variables: Construction variables can be retrieved and set using the Dictionary method of the construction environment: dict = env.Dictionary() dict["CC"] = "cc" or using the [] operator: env["CC"] = "cc" Construction variables can also be passed to the construction environment constructor: env = Environment(CC="cc") or when copying a construction environment using the Clone method: env2 = env.Clone(CC="cl.exe") Configure Contexts scons supports configure contexts, an integrated mechanism similar to the various AC_CHECK macros in GNU autoconf for testing for the existence of C header files, libraries, etc. In contrast to autoconf, scons does not maintain an explicit cache of the tested values, but uses its normal dependency tracking to keep the checked values up to date. However, users may override this behaviour with the command line option. The following methods can be used to perform checks: Configure(env, [custom_tests, conf_dir, log_file, config_h, clean, help]) env.Configure([custom_tests, conf_dir, log_file, config_h, clean, help]) This creates a configure context, which can be used to perform checks. env specifies the environment for building the tests. This environment may be modified when performing checks. custom_tests is a dictionary containing custom tests. See also the section about custom tests below. By default, no custom tests are added to the configure context. conf_dir specifies a directory where the test cases are built. Note that this directory is not used for building normal targets. The default value is the directory #/.sconf_temp. log_file specifies a file which collects the output from commands that are executed to check for the existence of header files, libraries, etc. The default is the file #/config.log. If you are using the VariantDir() method, you may want to specify a subdirectory under your variant directory. config_h specifies a C header file where the results of tests will be written, e.g. #define HAVE_STDIO_H, #define HAVE_LIBM, etc. The default is to not write a config.h file. You can specify the same config.h file in multiple calls to Configure, in which case scons will concatenate all results in the specified file. Note that SCons uses its normal dependency checking to decide if it's necessary to rebuild the specified config_h file. This means that the file is not necessarily re-built each time scons is run, but is only rebuilt if its contents will have changed and some target that depends on the config_h file is being built. The optional clean and help arguments can be used to suppress execution of the configuration tests when the or options are used, respectively. The default behavior is always to execute configure context tests, since the results of the tests may affect the list of targets to be cleaned or the help text. If the configure tests do not affect these, then you may add the clean=False or help=False arguments (or both) to avoid unnecessary test execution. A created Configure instance has the following associated methods: SConf.Finish(context) sconf.Finish() This method should be called after configuration is done. It returns the environment as modified by the configuration checks performed. After this method is called, no further checks can be performed with this configuration context. However, you can create a new Configure context to perform additional checks. Only one context should be active at a time. The following Checks are predefined. (This list will likely grow larger as time goes by and developers contribute new useful tests.) SConf.CheckHeader(context, header, [include_quotes, language]) sconf.CheckHeader(header, [include_quotes, language]) Checks if header is usable in the specified language. header may be a list, in which case the last item in the list is the header file to be checked, and the previous list items are header files whose #include lines should precede the header line being checked for. The optional argument include_quotes must be a two character string, where the first character denotes the opening quote and the second character denotes the closing quote. By default, both characters are " (double quote). The optional argument language should be either C or C++ and selects the compiler to be used for the check. Returns 1 on success and 0 on failure. SConf.CheckCHeader(context, header, [include_quotes]) sconf.CheckCHeader(header, [include_quotes]) This is a wrapper around SConf.CheckHeader which checks if header is usable in the C language. header may be a list, in which case the last item in the list is the header file to be checked, and the previous list items are header files whose #include lines should precede the header line being checked for. The optional argument include_quotes must be a two character string, where the first character denotes the opening quote and the second character denotes the closing quote (both default to \N'34'). Returns 1 on success and 0 on failure. SConf.CheckCXXHeader(context, header, [include_quotes]) sconf.CheckCXXHeader(header, [include_quotes]) This is a wrapper around SConf.CheckHeader which checks if header is usable in the C++ language. header may be a list, in which case the last item in the list is the header file to be checked, and the previous list items are header files whose #include lines should precede the header line being checked for. The optional argument include_quotes must be a two character string, where the first character denotes the opening quote and the second character denotes the closing quote (both default to \N'34'). Returns 1 on success and 0 on failure. SConf.CheckFunc(context,, function_name, [header, language]) sconf.CheckFunc(function_name, [header, language]) Checks if the specified C or C++ function is available. function_name is the name of the function to check for. The optional header argument is a string that will be placed at the top of the test file that will be compiled to check if the function exists; the default is: #ifdef __cplusplus extern "C" #endif char function_name(); The optional language argument should be C or C++ and selects the compiler to be used for the check; the default is "C". SConf.CheckLib(context, [library, symbol, header, language, autoadd=1]) sconf.CheckLib([library, symbol, header, language, autoadd=1]) Checks if library provides symbol. If the value of autoadd is 1 and the library provides the specified symbol, appends the library to the LIBS construction environment variable. library may also be None (the default), in which case symbol is checked with the current LIBS variable, or a list of library names, in which case each library in the list will be checked for symbol. If symbol is not set or is None, then SConf.CheckLib() just checks if you can link against the specified library. The optional language argument should be C or C++ and selects the compiler to be used for the check; the default is "C". The default value for autoadd is 1. This method returns 1 on success and 0 on error. SConf.CheckLibWithHeader(context, library, header, language, [call, autoadd]) sconf.CheckLibWithHeader(library, header, language, [call, autoadd]) In contrast to the SConf.CheckLib call, this call provides a more sophisticated way to check against libraries. Again, library specifies the library or a list of libraries to check. header specifies a header to check for. header may be a list, in which case the last item in the list is the header file to be checked, and the previous list items are header files whose #include lines should precede the header line being checked for. language may be one of 'C','c','CXX','cxx','C++' and 'c++'. call can be any valid expression (with a trailing ';'). If call is not set, the default simply checks that you can link against the specified library. autoadd specifies whether to add the library to the environment (only if the check succeeds). This method returns 1 on success and 0 on error. SConf.CheckType(context, type_name, [includes, language]) sconf.CheckType(type_name, [includes, language]) Checks for the existence of a type defined by typedef. type_name specifies the typedef name to check for. includes is a string containing one or more #include lines that will be inserted into the program that will be run to test for the existence of the type. The optional language argument should be C or C++ and selects the compiler to be used for the check; the default is "C". Example: sconf.CheckType('foo_type', '#include "my_types.h"', 'C++') Configure.CheckCC(self) Checks whether the C compiler (as defined by the CC construction variable) works by trying to compile a small source file. By default, SCons only detects if there is a program with the correct name, not if it is a functioning compiler. This uses the exact same command than the one used by the object builder for C source file, so it can be used to detect if a particular compiler flag works or not. Configure.CheckCXX(self) Checks whether the C++ compiler (as defined by the CXX construction variable) works by trying to compile a small source file. By default, SCons only detects if there is a program with the correct name, not if it is a functioning compiler. This uses the exact same command than the one used by the object builder for CXX source files, so it can be used to detect if a particular compiler flag works or not. Configure.CheckSHCC(self) Checks whether the C compiler (as defined by the SHCC construction variable) works by trying to compile a small source file. By default, SCons only detects if there is a program with the correct name, not if it is a functioning compiler. This uses the exact same command than the one used by the object builder for C source file, so it can be used to detect if a particular compiler flag works or not. This does not check whether the object code can be used to build a shared library, only that the compilation (not link) succeeds. Configure.CheckSHCXX(self) Checks whether the C++ compiler (as defined by the SHCXX construction variable) works by trying to compile a small source file. By default, SCons only detects if there is a program with the correct name, not if it is a functioning compiler. This uses the exact same command than the one used by the object builder for CXX source files, so it can be used to detect if a particular compiler flag works or not. This does not check whether the object code can be used to build a shared library, only that the compilation (not link) succeeds. Example of a typical Configure usage: env = Environment() conf = Configure( env ) if not conf.CheckCHeader( 'math.h' ): print 'We really need math.h!' Exit(1) if conf.CheckLibWithHeader( 'qt', 'qapp.h', 'c++', 'QApplication qapp(0,0);' ): # do stuff for qt - usage, e.g. conf.env.Append( CPPFLAGS = '-DWITH_QT' ) env = conf.Finish() SConf.CheckTypeSize(context, type_name, [header, language, expect]) sconf.CheckTypeSize(type_name, [header, language, expect]) Checks for the size of a type defined by typedef. type_name specifies the typedef name to check for. The optional header argument is a string that will be placed at the top of the test file that will be compiled to check if the function exists; the default is empty. The optional language argument should be C or C++ and selects the compiler to be used for the check; the default is "C". The optional expect argument should be an integer. If this argument is used, the function will only check whether the type given in type_name has the expected size (in bytes). For example, CheckTypeSize('short', expect = 2) will return success only if short is two bytes. SConf.CheckDeclaration(context, symbol, [includes, language]) sconf.CheckDeclaration(symbol, [includes, language]) Checks if the specified symbol is declared. includes is a string containing one or more #include lines that will be inserted into the program that will be run to test for the existence of the type. The optional language argument should be C or C++ and selects the compiler to be used for the check; the default is "C". SConf.Define(context, symbol, [value, comment]) sconf.Define(symbol, [value, comment]) This function does not check for anything, but defines a preprocessor symbol that will be added to the configuration header file. It is the equivalent of AC_DEFINE, and defines the symbol name with the optional value and the optional comment comment. Examples: env = Environment() conf = Configure( env ) # Puts the following line in the config header file: # #define A_SYMBOL conf.Define('A_SYMBOL') # Puts the following line in the config header file: # #define A_SYMBOL 1 conf.Define('A_SYMBOL', 1) Be careful about quoting string values, though: env = Environment() conf = Configure( env ) # Puts the following line in the config header file: # #define A_SYMBOL YA conf.Define('A_SYMBOL', "YA") # Puts the following line in the config header file: # #define A_SYMBOL "YA" conf.Define('A_SYMBOL', '"YA"') For comment: env = Environment() conf = Configure( env ) # Puts the following lines in the config header file: # /* Set to 1 if you have a symbol */ # #define A_SYMBOL 1 conf.Define('A_SYMBOL', 1, 'Set to 1 if you have a symbol') You can define your own custom checks. in addition to the predefined checks. These are passed in a dictionary to the Configure function. This dictionary maps the names of the checks to user defined Python callables (either Python functions or class instances implementing the __call__ method). The first argument of the call is always a CheckContext instance followed by the arguments, which must be supplied by the user of the check. These CheckContext instances define the following methods: CheckContext.Message(self, text) Usually called before the check is started. text will be displayed to the user, e.g. 'Checking for library X...' CheckContext.Result(self,, res) Usually called after the check is done. res can be either an integer or a string. In the former case, 'yes' (res != 0) or 'no' (res == 0) is displayed to the user, in the latter case the given string is displayed. CheckContext.TryCompile(self, text, extension) Checks if a file with the specified extension (e.g. '.c') containing text can be compiled using the environment's Object builder. Returns 1 on success and 0 on failure. CheckContext.TryLink(self, text, extension) Checks, if a file with the specified extension (e.g. '.c') containing text can be compiled using the environment's Program builder. Returns 1 on success and 0 on failure. CheckContext.TryRun(self, text, extension) Checks, if a file with the specified extension (e.g. '.c') containing text can be compiled using the environment's Program builder. On success, the program is run. If the program executes successfully (that is, its return status is 0), a tuple (1, outputStr) is returned, where outputStr is the standard output of the program. If the program fails execution (its return status is non-zero), then (0, '') is returned. CheckContext.TryAction(self, action, [text, extension]) Checks if the specified action with an optional source file (contents text , extension extension = '' ) can be executed. action may be anything which can be converted to a scons Action. On success, (1, outputStr) is returned, where outputStr is the content of the target file. On failure (0, '') is returned. CheckContext.TryBuild(self, builder, [text, extension]) Low level implementation for testing specific builds; the methods above are based on this method. Given the Builder instance builder and the optional text of a source file with optional extension, this method returns 1 on success and 0 on failure. In addition, self.lastTarget is set to the build target node, if the build was successful. Example for implementing and using custom tests: def CheckQt(context, qtdir): context.Message( 'Checking for qt ...' ) lastLIBS = context.env['LIBS'] lastLIBPATH = context.env['LIBPATH'] lastCPPPATH= context.env['CPPPATH'] context.env.Append(LIBS = 'qt', LIBPATH = qtdir + '/lib', CPPPATH = qtdir + '/include' ) ret = context.TryLink(""" #include <qapp.h> int main(int argc, char **argv) { QApplication qapp(argc, argv); return 0; } """) if not ret: context.env.Replace(LIBS = lastLIBS, LIBPATH=lastLIBPATH, CPPPATH=lastCPPPATH) context.Result( ret ) return ret env = Environment() conf = Configure( env, custom_tests = { 'CheckQt' : CheckQt } ) if not conf.CheckQt('/usr/lib/qt'): print 'We really need qt!' Exit(1) env = conf.Finish() Command-Line Construction Variables Often when building software, some variables must be specified at build time. For example, libraries needed for the build may be in non-standard locations, or site-specific compiler options may need to be passed to the compiler. scons provides a Variables object to support overriding construction variables on the command line: $ scons VARIABLE=foo The variable values can also be specified in a text-based SConscript file. To create a Variables object, call the Variables() function: Variables([files], [args]) This creates a Variables object that will read construction variables from the file or list of filenames specified in files. If no files are specified, or the files argument is None, then no files will be read. The optional argument args is a dictionary of values that will override anything read from the specified files; it is primarily intended to be passed the ARGUMENTS dictionary that holds variables specified on the command line. Example: vars = Variables('custom.py') vars = Variables('overrides.py', ARGUMENTS) vars = Variables(None, {FOO:'expansion', BAR:7}) Variables objects have the following methods: Add(key, [help, default, validator, converter]) This adds a customizable construction variable to the Variables object. key is the name of the variable. help is the help text for the variable. default is the default value of the variable; if the default value is None and there is no explicit value specified, the construction variable will not be added to the construction environment. validator is called to validate the value of the variable, and should take three arguments: key, value, and environment. The recommended way to handle an invalid value is to raise an exception (see example below). converter is called to convert the value before putting it in the environment, and should take either a value, or the value and environment, as parameters. The converter must return a value, which will be converted into a string before being validated by the validator (if any) and then added to the environment. Examples: vars.Add('CC', 'The C compiler') def validate_color(key, val, env): if not val in ['red', 'blue', 'yellow']: raise Exception("Invalid color value '%s'" % val) vars.Add('COLOR', validator=valid_color) AddVariables(list) A wrapper script that adds multiple customizable construction variables to a Variables object. list is a list of tuple or list objects that contain the arguments for an individual call to the Add method. opt.AddVariables( ('debug', '', 0), ('CC', 'The C compiler'), ('VALIDATE', 'An option for testing validation', 'notset', validator, None), ) Update(env, [args]) This updates a construction environment env with the customized construction variables. Any specified variables that are not configured for the Variables object will be saved and may be retrieved with the UnknownVariables() method, below. Normally this method is not called directly, but is called indirectly by passing the Variables object to the Environment() function: env = Environment(variables=vars) The text file(s) that were specified when the Variables object was created are executed as Python scripts, and the values of (global) Python variables set in the file are added to the construction environment. Example: CC = 'my_cc' UnknownVariables() Returns a dictionary containing any variables that were specified either in the files or the dictionary with which the Variables object was initialized, but for which the Variables object was not configured. env = Environment(variables=vars) for key, value in vars.UnknownVariables(): print "unknown variable: %s=%s" % (key, value) Save(filename, env) This saves the currently set variables into a script file named filename that can be used on the next invocation to automatically load the current settings. This method combined with the Variables method can be used to support caching of variables between runs. env = Environment() vars = Variables(['variables.cache', 'custom.py']) vars.Add(...) vars.Update(env) vars.Save('variables.cache', env) GenerateHelpText(env, [sort]) This generates help text documenting the customizable construction variables suitable to passing in to the Help() function. env is the construction environment that will be used to get the actual values of customizable variables. Calling with an optional sort function will cause the output to be sorted by the specified argument. The specific sort function should take two arguments and return -1, 0 or 1 (like the standard Python cmp function). Help(vars.GenerateHelpText(env)) Help(vars.GenerateHelpText(env, sort=cmp)) FormatVariableHelpText(env, opt, help, default, actual) This method returns a formatted string containing the printable help text for one option. It is normally not called directly, but is called by the GenerateHelpText() method to create the returned help text. It may be overridden with your own function that takes the arguments specified above and returns a string of help text formatted to your liking. Note that the GenerateHelpText() will not put any blank lines or extra characters in between the entries, so you must add those characters to the returned string if you want the entries separated. def my_format(env, opt, help, default, actual): fmt = "\n%s: default=%s actual=%s (%s)\n" return fmt % (opt, default. actual, help) vars.FormatVariableHelpText = my_format To make it more convenient to work with customizable Variables, scons provides a number of functions that make it easy to set up various types of Variables: BoolVariable(key, help, default) Return a tuple of arguments to set up a Boolean option. The option will use the specified name key, have a default value of default, and display the specified help text. The option will interpret the values y, yes, t, true, 1, on and all as true, and the values n, no, f, false, 0, off and none as false. EnumVariable(key, help, default, allowed_values, [map, ignorecase]) Return a tuple of arguments to set up an option whose value may be one of a specified list of legal enumerated values. The option will use the specified name key, have a default value of default, and display the specified help text. The option will only support those values in the allowed_values list. The optional map argument is a dictionary that can be used to convert input values into specific legal values in the allowed_values list. If the value of ignore_case is 0 (the default), then the values are case-sensitive. If the value of ignore_case is 1, then values will be matched case-insensitive. If the value of ignore_case is 1, then values will be matched case-insensitive, and all input values will be converted to lower case. ListVariable(key, help, default, names, [,map]) Return a tuple of arguments to set up an option whose value may be one or more of a specified list of legal enumerated values. The option will use the specified name key, have a default value of default, and display the specified help text. The option will only support the values all, none, or the values in the names list. More than one value may be specified, with all values separated by commas. The default may be a string of comma-separated default values, or a list of the default values. The optional map argument is a dictionary that can be used to convert input values into specific legal values in the names list. PackageVariable(key, help, default) Return a tuple of arguments to set up an option whose value is a path name of a package that may be enabled, disabled or given an explicit path name. The option will use the specified name key, have a default value of default, and display the specified help text. The option will support the values yes, true, on, enable or search, in which case the specified default will be used, or the option may be set to an arbitrary string (typically the path name to a package that is being enabled). The option will also support the values no, false, off or disable to disable use of the specified option. PathVariable(key, help, default, [validator]) Return a tuple of arguments to set up an option whose value is expected to be a path name. The option will use the specified name key, have a default value of default, and display the specified help text. An additional validator may be specified that will be called to verify that the specified path is acceptable. SCons supplies the following ready-made validators: PathVariable.PathExists (the default), which verifies that the specified path exists; PathVariable.PathIsFile, which verifies that the specified path is an existing file; PathVariable.PathIsDir, which verifies that the specified path is an existing directory; PathVariable.PathIsDirCreate, which verifies that the specified path is a directory and will create the specified directory if the path does not exist; and PathVariable.PathAccept, which simply accepts the specific path name argument without validation, and which is suitable if you want your users to be able to specify a directory path that will be created as part of the build process, for example. You may supply your own validator function, which must take three arguments (key, the name of the variable to be set; val, the specified value being checked; and env, the construction environment) and should raise an exception if the specified value is not acceptable. These functions make it convenient to create a number of variables with consistent behavior in a single call to the AddVariables method: vars.AddVariables( BoolVariable('warnings', 'compilation with -Wall and similiar', 1), EnumVariable('debug', 'debug output and symbols', 'no' allowed_values=('yes', 'no', 'full'), map={}, ignorecase=0), # case sensitive ListVariable('shared', 'libraries to build as shared libraries', 'all', names = list_of_libs), PackageVariable('x11', 'use X11 installed here (yes = search some places)', 'yes'), PathVariable('qtdir', 'where the root of Qt is installed', qtdir), PathVariable('foopath', 'where the foo library is installed', foopath, PathVariable.PathIsDir), ) File and Directory Nodes The File() and Dir() functions return File and Dir Nodes, respectively. python objects, respectively. Those objects have several user-visible attributes and methods that are often useful: path The build path of the given file or directory. This path is relative to the top-level directory (where the SConstruct file is found). The build path is the same as the source path if variant_dir is not being used. abspath The absolute build path of the given file or directory. srcnode() The srcnode() method returns another File or Dir object representing the source path of the given File or Dir. The # Get the current build dir's path, relative to top. Dir('.').path # Current dir's absolute path Dir('.').abspath # Next line is always '.', because it is the top dir's path relative to itself. Dir('#.').path File('foo.c').srcnode().path # source path of the given source file. # Builders also return File objects: foo = env.Program('foo.c') print "foo will be built in %s"%foo.path A Dir Node or File Node can also be used to create file and subdirectory Nodes relative to the generating Node. A Dir Node will place the new Nodes within the directory it represents. A File node will place the new Nodes within its parent directory (that is, "beside" the file in question). If d is a Dir (directory) Node and f is a File (file) Node, then these methods are available: d.Dir(name) Returns a directory Node for a subdirectory of d named name. d.File(name) Returns a file Node for a file within d named name. d.Entry(name) Returns an unresolved Node within d named name. f.Dir(name) Returns a directory named name within the parent directory of f. f.File(name) Returns a file named name within the parent directory of f. f.Entry(name) Returns an unresolved Node named name within the parent directory of f. For example: # Get a Node for a file within a directory incl = Dir('include') f = incl.File('header.h') # Get a Node for a subdirectory within a directory dist = Dir('project-3.2.1) src = dist.Dir('src') # Get a Node for a file in the same directory cfile = File('sample.c') hfile = cfile.File('sample.h') # Combined example docs = Dir('docs') html = docs.Dir('html') index = html.File('index.html') css = index.File('app.css') EXTENDING SCONS Builder Objects scons can be extended to build different types of targets by adding new Builder objects to a construction environment. In general, you should only need to add a new Builder object when you want to build a new type of file or other external target. If you just want to invoke a different compiler or other tool to build a Program, Object, Library, or any other type of output file for which scons already has an existing Builder, it is generally much easier to use those existing Builders in a construction environment that sets the appropriate construction variables (CC, LINK, etc.). Builder objects are created using the Builder function. The Builder function accepts the following arguments: action The command line string used to build the target from the source. action can also be: a list of strings representing the command to be executed and its arguments (suitable for enclosing white space in an argument), a dictionary mapping source file name suffixes to any combination of command line strings (if the builder should accept multiple source file extensions), a Python function; an Action object (see the next section); or a list of any of the above. An action function takes three arguments: source - a list of source nodes, target - a list of target nodes, env - the construction environment. prefix The prefix that will be prepended to the target file name. This may be specified as a:

* string, * callable object - a function or other callable that takes two arguments (a construction environment and a list of sources) and returns a prefix, * dictionary - specifies a mapping from a specific source suffix (of the first source specified) to a corresponding target prefix. Both the source suffix and target prefix specifications may use environment variable substitution, and the target prefix (the 'value' entries in the dictionary) may also be a callable object. The default target prefix may be indicated by a dictionary entry with a key value of None.

b = Builder("build_it < $SOURCE > $TARGET", prefix = "file-") def gen_prefix(env, sources): return "file-" + env['PLATFORM'] + '-' b = Builder("build_it < $SOURCE > $TARGET", prefix = gen_prefix) b = Builder("build_it < $SOURCE > $TARGET", suffix = { None: "file-", "$SRC_SFX_A": gen_prefix }) suffix The suffix that will be appended to the target file name. This may be specified in the same manner as the prefix above. If the suffix is a string, then scons will append a '.' to the beginning of the suffix if it's not already there. The string returned by callable object (or obtained from the dictionary) is untouched and must append its own '.' to the beginning if one is desired. b = Builder("build_it < $SOURCE > $TARGET" suffix = "-file") def gen_suffix(env, sources): return "." + env['PLATFORM'] + "-file" b = Builder("build_it < $SOURCE > $TARGET", suffix = gen_suffix) b = Builder("build_it < $SOURCE > $TARGET", suffix = { None: ".sfx1", "$SRC_SFX_A": gen_suffix }) ensure_suffix When set to any true value, causes scons to add the target suffix specified by the suffix keyword to any target strings that have a different suffix. (The default behavior is to leave untouched any target file name that looks like it already has any suffix.) b1 = Builder("build_it < $SOURCE > $TARGET" suffix = ".out") b2 = Builder("build_it < $SOURCE > $TARGET" suffix = ".out", ensure_suffix) env = Environment() env['BUILDERS']['B1'] = b1 env['BUILDERS']['B2'] = b2 # Builds "foo.txt" because ensure_suffix is not set. env.B1('foo.txt', 'foo.in') # Builds "bar.txt.out" because ensure_suffix is set. env.B2('bar.txt', 'bar.in') src_suffix The expected source file name suffix. This may be a string or a list of strings. target_scanner A Scanner object that will be invoked to find implicit dependencies for this target file. This keyword argument should be used for Scanner objects that find implicit dependencies based only on the target file and the construction environment, not for implicit dependencies based on source files. (See the section "Scanner Objects" below, for information about creating Scanner objects.) source_scanner A Scanner object that will be invoked to find implicit dependencies in any source files used to build this target file. This is where you would specify a scanner to find things like #include lines in source files. The pre-built DirScanner Scanner object may be used to indicate that this Builder should scan directory trees for on-disk changes to files that scons does not know about from other Builder or function calls. (See the section "Scanner Objects" below, for information about creating your own Scanner objects.) target_factory A factory function that the Builder will use to turn any targets specified as strings into SCons Nodes. By default, SCons assumes that all targets are files. Other useful target_factory values include Dir, for when a Builder creates a directory target, and Entry, for when a Builder can create either a file or directory target. Example: MakeDirectoryBuilder = Builder(action=my_mkdir, target_factory=Dir) env = Environment() env.Append(BUILDERS = {'MakeDirectory':MakeDirectoryBuilder}) env.MakeDirectory('new_directory', []) Note that the call to the MakeDirectory Builder needs to specify an empty source list to make the string represent the builder's target; without that, it would assume the argument is the source, and would try to deduce the target name from it, which in the absence of an automatically-added prefix or suffix would lead to a matching target and source name and a circular dependency. source_factory A factory function that the Builder will use to turn any sources specified as strings into SCons Nodes. By default, SCons assumes that all source are files. Other useful source_factory values include Dir, for when a Builder uses a directory as a source, and Entry, for when a Builder can use files or directories (or both) as sources. Example: CollectBuilder = Builder(action=my_mkdir, source_factory=Entry) env = Environment() env.Append(BUILDERS = {'Collect':CollectBuilder}) env.Collect('archive', ['directory_name', 'file_name']) emitter A function or list of functions to manipulate the target and source lists before dependencies are established and the target(s) are actually built. emitter can also be a string containing a construction variable to expand to an emitter function or list of functions, or a dictionary mapping source file suffixes to emitter functions. (Only the suffix of the first source file is used to select the actual emitter function from an emitter dictionary.) An emitter function takes three arguments: source - a list of source nodes, target - a list of target nodes, env - the construction environment. An emitter must return a tuple containing two lists, the list of targets to be built by this builder, and the list of sources for this builder. Example: def e(target, source, env): return (target + ['foo.foo'], source + ['foo.src']) # Simple association of an emitter function with a Builder. b = Builder("my_build < $TARGET > $SOURCE", emitter = e) def e2(target, source, env): return (target + ['bar.foo'], source + ['bar.src']) # Simple association of a list of emitter functions with a Builder. b = Builder("my_build < $TARGET > $SOURCE", emitter = [e, e2]) # Calling an emitter function through a construction variable. env = Environment(MY_EMITTER = e) b = Builder("my_build < $TARGET > $SOURCE", emitter = '$MY_EMITTER') # Calling a list of emitter functions through a construction variable. env = Environment(EMITTER_LIST = [e, e2]) b = Builder("my_build < $TARGET > $SOURCE", emitter = '$EMITTER_LIST') # Associating multiple emitters with different file # suffixes using a dictionary. def e_suf1(target, source, env): return (target + ['another_target_file'], source) def e_suf2(target, source, env): return (target, source + ['another_source_file']) b = Builder("my_build < $TARGET > $SOURCE", emitter = {'.suf1' : e_suf1, '.suf2' : e_suf2}) multi Specifies whether this builder is allowed to be called multiple times for the same target file(s). The default is 0, which means the builder can not be called multiple times for the same target file(s). Calling a builder multiple times for the same target simply adds additional source files to the target; it is not allowed to change the environment associated with the target, specify addition environment overrides, or associate a different builder with the target. env A construction environment that can be used to fetch source code using this Builder. (Note that this environment is not used for normal builds of normal target files, which use the environment that was used to call the Builder for the target file.) generator A function that returns a list of actions that will be executed to build the target(s) from the source(s). The returned action(s) may be an Action object, or anything that can be converted into an Action object (see the next section). The generator function takes four arguments: source - a list of source nodes, target - a list of target nodes, env - the construction environment, for_signature - a Boolean value that specifies whether the generator is being called for generating a build signature (as opposed to actually executing the command). Example: def g(source, target, env, for_signature): return [["gcc", "-c", "-o"] + target + source] b = Builder(generator=g) The generator and action arguments must not both be used for the same Builder. src_builder Specifies a builder to use when a source file name suffix does not match any of the suffixes of the builder. Using this argument produces a multi-stage builder. single_source Specifies that this builder expects exactly one source file per call. Giving more than one source file without target files results in implicitely calling the builder multiple times (once for each source given). Giving multiple source files together with target files results in a UserError exception. The generator and action arguments must not both be used for the same Builder. source_ext_match When the specified action argument is a dictionary, the default behavior when a builder is passed multiple source files is to make sure that the extensions of all the source files match. If it is legal for this builder to be called with a list of source files with different extensions, this check can be suppressed by setting source_ext_match to None or some other non-true value. When source_ext_match is disable, scons will use the suffix of the first specified source file to select the appropriate action from the action dictionary. In the following example, the setting of source_ext_match prevents scons from exiting with an error due to the mismatched suffixes of foo.in and foo.extra. b = Builder(action={'.in' : 'build $SOURCES > $TARGET'}, source_ext_match = None) env = Environment(BUILDERS = {'MyBuild':b}) env.MyBuild('foo.out', ['foo.in', 'foo.extra']) env A construction environment that can be used to fetch source code using this Builder. (Note that this environment is not used for normal builds of normal target files, which use the environment that was used to call the Builder for the target file.) b = Builder(action="build < $SOURCE > $TARGET") env = Environment(BUILDERS = {'MyBuild' : b}) env.MyBuild('foo.out', 'foo.in', my_arg = 'xyzzy') chdir A directory from which scons will execute the action(s) specified for this Builder. If the chdir argument is a string or a directory Node, scons will change to the specified directory. If the chdir is not a string or Node and is non-zero, then scons will change to the target file's directory. Note that scons will not automatically modify its expansion of construction variables like $TARGET and $SOURCE when using the chdir keyword argument--that is, the expanded file names will still be relative to the top-level SConstruct directory, and consequently incorrect relative to the chdir directory. Builders created using chdir keyword argument, will need to use construction variable expansions like ${TARGET.file} and ${SOURCE.file} to use just the filename portion of the targets and source. b = Builder(action="build < ${SOURCE.file} > ${TARGET.file}", chdir=1) env = Environment(BUILDERS = {'MyBuild' : b}) env.MyBuild('sub/dir/foo.out', 'sub/dir/foo.in') WARNING: Python only keeps one current directory location for all of the threads. This means that use of the chdir argument will not work with the SCons option, because individual worker threads spawned by SCons interfere with each other when they start changing directory. Any additional keyword arguments supplied when a Builder object is created (that is, when the Builder() function is called) will be set in the executing construction environment when the Builder object is called. The canonical example here would be to set a construction variable to the repository of a source code system. Any additional keyword arguments supplied when a Builder object is called will only be associated with the target created by that particular Builder call (and any other files built as a result of the call). These extra keyword arguments are passed to the following functions: command generator functions, function Actions, and emitter functions. Action Objects The Builder() function will turn its action keyword argument into an appropriate internal Action object. You can also explicity create Action objects using the Action() global function, which can then be passed to the Builder() function. This can be used to configure an Action object more flexibly, or it may simply be more efficient than letting each separate Builder object create a separate Action when multiple Builder objects need to do the same thing. The Action() global function returns an appropriate object for the action represented by the type of the first argument: Action If the first argument is already an Action object, the object is simply returned. String If the first argument is a string, a command-line Action is returned. Note that the command-line string may be preceded by an @ (at-sign) to suppress printing of the specified command line, or by a - (hyphen) to ignore the exit status from the specified command: Action('$CC -c -o $TARGET $SOURCES') # Doesn't print the line being executed. Action('@build $TARGET $SOURCES') # Ignores return value Action('-build $TARGET $SOURCES') List If the first argument is a list, then a list of Action objects is returned. An Action object is created as necessary for each element in the list. If an element within the list is itself a list, the internal list is the command and arguments to be executed via the command line. This allows white space to be enclosed in an argument by defining a command in a list within a list: Action([['cc', '-c', '-DWHITE SPACE', '-o', '$TARGET', '$SOURCES']]) Function If the first argument is a Python function, a function Action is returned. The Python function must take three keyword arguments, target (a Node object representing the target file), source (a Node object representing the source file) and env (the construction environment used for building the target file). The target and source arguments may be lists of Node objects if there is more than one target file or source file. The actual target and source file name(s) may be retrieved from their Node objects via the built-in Python str() function: target_file_name = str(target) source_file_names = map(lambda x: str(x), source) The function should return 0 or None to indicate a successful build of the target file(s). The function may raise an exception or return a non-zero exit status to indicate an unsuccessful build. def build_it(target = None, source = None, env = None): # build the target from the source return 0 a = Action(build_it) If the action argument is not one of the above, None is returned. The second argument is optional and is used to define the output which is printed when the Action is actually performed. In the absence of this parameter, or if it's an empty string, a default output depending on the type of the action is used. For example, a command-line action will print the executed command. The argument must be either a Python function or a string. In the first case, it's a function that returns a string to be printed to describe the action being executed. The function may also be specified by the strfunction= keyword argument. Like a function to build a file, this function must take three keyword arguments: target (a Node object representing the target file), source (a Node object representing the source file) and env (a construction environment). The target and source arguments may be lists of Node objects if there is more than one target file or source file. In the second case, you provide the string itself. The string may also be specified by the cmdstr= keyword argument. The string typically contains variables, notably $TARGET(S) and $SOURCE(S), or consists of just a single variable, which is optionally defined somewhere else. SCons itself heavily uses the latter variant. Examples: def build_it(target, source, env): # build the target from the source return 0 def string_it(target, source, env): return "building '%s' from '%s'" % (target[0], source[0]) # Use a positional argument. f = Action(build_it, string_it) s = Action(build_it, "building '$TARGET' from '$SOURCE'") # Alternatively, use a keyword argument. f = Action(build_it, strfunction=string_it) s = Action(build_it, cmdstr="building '$TARGET' from '$SOURCE'") # You can provide a configurable variable. l = Action(build_it, '$STRINGIT') The third and succeeding arguments, if present, may either be a construction variable or a list of construction variables whose values will be included in the signature of the Action when deciding whether a target should be rebuilt because the action changed. The variables may also be specified by a varlist= keyword parameter; if both are present, they are combined. This is necessary whenever you want a target to be rebuilt when a specific construction variable changes. This is not often needed for a string action, as the expanded variables will normally be part of the command line, but may be needed if a Python function action uses the value of a construction variable when generating the command line. def build_it(target, source, env): # build the target from the 'XXX' construction variable open(target[0], 'w').write(env['XXX']) return 0 # Use positional arguments. a = Action(build_it, '$STRINGIT', ['XXX']) # Alternatively, use a keyword argument. a = Action(build_it, varlist=['XXX']) The Action() global function can be passed the following optional keyword arguments to modify the Action object's behavior: chdir The chdir keyword argument specifies that scons will execute the action after changing to the specified directory. If the chdir argument is a string or a directory Node, scons will change to the specified directory. If the chdir argument is not a string or Node and is non-zero, then scons will change to the target file's directory. Note that scons will not automatically modify its expansion of construction variables like $TARGET and $SOURCE when using the chdir keyword argument--that is, the expanded file names will still be relative to the top-level SConstruct directory, and consequently incorrect relative to the chdir directory. Builders created using chdir keyword argument, will need to use construction variable expansions like ${TARGET.file} and ${SOURCE.file} to use just the filename portion of the targets and source. a = Action("build < ${SOURCE.file} > ${TARGET.file}", chdir=1) exitstatfunc The Action() global function also takes an exitstatfunc keyword argument which specifies a function that is passed the exit status (or return value) from the specified action and can return an arbitrary or modified value. This can be used, for example, to specify that an Action object's return value should be ignored under special conditions and SCons should, therefore, consider that the action always suceeds: def always_succeed(s): # Always return 0, which indicates success. return 0 a = Action("build < ${SOURCE.file} > ${TARGET.file}", exitstatfunc=always_succeed) batch_key The batch_key keyword argument can be used to specify that the Action can create multiple target files by processing multiple independent source files simultaneously. (The canonical example is "batch compilation" of multiple object files by passing multiple source files to a single invocation of a compiler such as Microsoft's Visual C / C++ compiler.) If the batch_key argument is any non-False, non-callable Python value, the configured Action object will cause scons to collect all targets built with the Action object and configured with the same construction environment into single invocations of the Action object's command line or function. Command lines will typically want to use the CHANGED_SOURCES construction variable (and possibly CHANGED_TARGETS as well) to only pass to the command line those sources that have actually changed since their targets were built. Example: a = Action('build $CHANGED_SOURCES', batch_key=True) The batch_key argument may also be a callable function that returns a key that will be used to identify different "batches" of target files to be collected for batch building. A batch_key function must take the following arguments: action The action object. env The construction environment configured for the target. target The list of targets for a particular configured action. source The list of source for a particular configured action. The returned key should typically be a tuple of values derived from the arguments, using any appropriate logic to decide how multiple invocations should be batched. For example, a batch_key function may decide to return the value of a specific construction variable from the env argument which will cause scons to batch-build targets with matching values of that variable, or perhaps return the id() of the entire construction environment, in which case scons will batch-build all targets configured with the same construction environment. Returning None indicates that the particular target should not be part of any batched build, but instead will be built by a separate invocation of action's command or function. Example: def batch_key(action, env, target, source): tdir = target[0].dir if tdir.name == 'special': # Don't batch-build any target # in the special/ subdirectory. return None return (id(action), id(env), tdir) a = Action('build $CHANGED_SOURCES', batch_key=batch_key) Miscellaneous Action Functions scons supplies a number of functions that arrange for various common file and directory manipulations to be performed. These are similar in concept to "tasks" in the Ant build tool, although the implementation is slightly different. These functions do not actually perform the specified action at the time the function is called, but instead return an Action object that can be executed at the appropriate time. (In Object-Oriented terminology, these are actually Action Factory functions that return Action objects.) In practice, there are two natural ways that these Action Functions are intended to be used. First, if you need to perform the action at the time the SConscript file is being read, you can use the Execute global function to do so: Execute(Touch('file')) Second, you can use these functions to supply Actions in a list for use by the Command method. This can allow you to perform more complicated sequences of file manipulation without relying on platform-specific external commands: that env = Environment(TMPBUILD = '/tmp/builddir') env.Command('foo.out', 'foo.in', [Mkdir('$TMPBUILD'), Copy('$TMPBUILD', '${SOURCE.dir}'), "cd $TMPBUILD && make", Delete('$TMPBUILD')]) Chmod(dest, mode) Returns an Action object that changes the permissions on the specified dest file or directory to the specified mode. Examples: Execute(Chmod('file', 0755)) env.Command('foo.out', 'foo.in', [Copy('$TARGET', '$SOURCE'), Chmod('$TARGET', 0755)]) Copy(dest, src) Returns an Action object that will copy the src source file or directory to the dest destination file or directory. Examples: Execute(Copy('foo.output', 'foo.input')) env.Command('bar.out', 'bar.in', Copy('$TARGET', '$SOURCE')) Delete(entry, [must_exist]) Returns an Action that deletes the specified entry, which may be a file or a directory tree. If a directory is specified, the entire directory tree will be removed. If the must_exist flag is set, then a Python error will be thrown if the specified entry does not exist; the default is must_exist=0, that is, the Action will silently do nothing if the entry does not exist. Examples: Execute(Delete('/tmp/buildroot')) env.Command('foo.out', 'foo.in', [Delete('${TARGET.dir}'), MyBuildAction]) Execute(Delete('file_that_must_exist', must_exist=1)) Mkdir(dir) Returns an Action that creates the specified directory dir . Examples: Execute(Mkdir('/tmp/outputdir')) env.Command('foo.out', 'foo.in', [Mkdir('/tmp/builddir'), Copy('/tmp/builddir/foo.in', '$SOURCE'), "cd /tmp/builddir && make", Copy('$TARGET', '/tmp/builddir/foo.out')]) Move(dest, src) Returns an Action that moves the specified src file or directory to the specified dest file or directory. Examples: Execute(Move('file.destination', 'file.source')) env.Command('output_file', 'input_file', [MyBuildAction, Move('$TARGET', 'file_created_by_MyBuildAction')]) Touch(file) Returns an Action that updates the modification time on the specified file. Examples: Execute(Touch('file_to_be_touched')) env.Command('marker', 'input_file', [MyBuildAction, Touch('$TARGET')]) Variable Substitution Before executing a command, scons performs construction variable interpolation on the strings that make up the command line of builders. Variables are introduced by a $ prefix. Besides construction variables, scons provides the following variables for each command execution: CHANGED_SOURCES The file names of all sources of the build command that have changed since the target was last built. CHANGED_TARGETS The file names of all targets that would be built from sources that have changed since the target was last built. SOURCE The file name of the source of the build command, or the file name of the first source if multiple sources are being built. SOURCES The file names of the sources of the build command. TARGET The file name of the target being built, or the file name of the first target if multiple targets are being built. TARGETS The file names of all targets being built. UNCHANGED_SOURCES The file names of all sources of the build command that have not changed since the target was last built. UNCHANGED_TARGETS The file names of all targets that would be built from sources that have not changed since the target was last built. (Note that the above variables are reserved and may not be set in a construction environment.) For example, given the construction variable CC='cc', targets=['foo'], and sources=['foo.c', 'bar.c']: action='$CC -c -o $TARGET $SOURCES' would produce the command line: cc -c -o foo foo.c bar.c Variable names may be surrounded by curly braces ({}) to separate the name from the trailing characters. Within the curly braces, a variable name may have a Python slice subscript appended to select one or more items from a list. In the previous example, the string: ${SOURCES[1]} would produce: bar.c Additionally, a variable name may have the following special modifiers appended within the enclosing curly braces to modify the interpolated string: base The base path of the file name, including the directory path but excluding any suffix. dir The name of the directory in which the file exists. file The file name, minus any directory portion. filebase Just the basename of the file, minus any suffix and minus the directory. suffix Just the file suffix. abspath The absolute path name of the file. posix The POSIX form of the path, with directories separated by / (forward slashes) not backslashes. This is sometimes necessary on Windows systems when a path references a file on other (POSIX) systems. srcpath The directory and file name to the source file linked to this file through VariantDir(). If this file isn't linked, it just returns the directory and filename unchanged. srcdir The directory containing the source file linked to this file through VariantDir(). If this file isn't linked, it just returns the directory part of the filename. rsrcpath The directory and file name to the source file linked to this file through VariantDir(). If the file does not exist locally but exists in a Repository, the path in the Repository is returned. If this file isn't linked, it just returns the directory and filename unchanged. rsrcdir The Repository directory containing the source file linked to this file through VariantDir(). If this file isn't linked, it just returns the directory part of the filename. For example, the specified target will expand as follows for the corresponding modifiers: $TARGET => sub/dir/file.x ${TARGET.base} => sub/dir/file ${TARGET.dir} => sub/dir ${TARGET.file} => file.x ${TARGET.filebase} => file ${TARGET.suffix} => .x ${TARGET.abspath} => /top/dir/sub/dir/file.x SConscript('src/SConscript', variant_dir='sub/dir') $SOURCE => sub/dir/file.x ${SOURCE.srcpath} => src/file.x ${SOURCE.srcdir} => src Repository('/usr/repository') $SOURCE => sub/dir/file.x ${SOURCE.rsrcpath} => /usr/repository/src/file.x ${SOURCE.rsrcdir} => /usr/repository/src Note that curly braces braces may also be used to enclose arbitrary Python code to be evaluated. (In fact, this is how the above modifiers are substituted, they are simply attributes of the Python objects that represent TARGET, SOURCES, etc.) See the section "Python Code Substitution" below, for more thorough examples of how this can be used. Lastly, a variable name may be a callable Python function associated with a construction variable in the environment. The function should take four arguments: target - a list of target nodes, source - a list of source nodes, env - the construction environment, for_signature - a Boolean value that specifies whether the function is being called for generating a build signature. SCons will insert whatever the called function returns into the expanded string: def foo(target, source, env, for_signature): return "bar" # Will expand $BAR to "bar baz" env=Environment(FOO=foo, BAR="$FOO baz") You can use this feature to pass arguments to a Python function by creating a callable class that stores one or more arguments in an object, and then uses them when the __call__() method is called. Note that in this case, the entire variable expansion must be enclosed by curly braces so that the arguments will be associated with the instantiation of the class: class foo(object): def __init__(self, arg): self.arg = arg def __call__(self, target, source, env, for_signature): return self.arg + " bar" # Will expand $BAR to "my argument bar baz" env=Environment(FOO=foo, BAR="${FOO('my argument')} baz") The special pseudo-variables $( and $) may be used to surround parts of a command line that may change without causing a rebuild--that is, which are not included in the signature of target files built with this command. All text between $( and $) will be removed from the command line before it is added to file signatures, and the $( and $) will be removed before the command is executed. For example, the command line: echo Last build occurred $( $TODAY $). > $TARGET would execute the command: echo Last build occurred $TODAY. > $TARGET but the command signature added to any target files would be: echo Last build occurred . > $TARGET Python Code Substitution Any python code within ${-} pairs gets evaluated by python 'eval', with the python globals set to the current environment's set of construction variables. So in the following case: env['COND'] = 0 env.Command('foo.out', 'foo.in', the command executed will be either echo FOO > foo.out or echo BAR > foo.out according to the current value of env['COND'] when the command is executed. The evaluation occurs when the target is being built, not when the SConscript is being read. So if env['COND'] is changed later in the SConscript, the final value will be used. Here's a more interesting example. Note that all of COND, FOO, and BAR are environment variables, and their values are substituted into the final command. FOO is a list, so its elements are interpolated separated by spaces. env=Environment() env['COND'] = 0 env['FOO'] = ['foo1', 'foo2'] env['BAR'] = 'barbar' env.Command('foo.out', 'foo.in', 'echo ${COND==1 and FOO or BAR} > $TARGET') # Will execute this: # echo foo1 foo2 > foo.out SCons uses the following rules when converting construction variables into command lines: String When the value is a string it is interpreted as a space delimited list of command line arguments. List When the value is a list it is interpreted as a list of command line arguments. Each element of the list is converted to a string. Other Anything that is not a list or string is converted to a string and interpreted as a single command line argument. Newline Newline characters (\n) delimit lines. The newline parsing is done after all other parsing, so it is not possible for arguments (e.g. file names) to contain embedded newline characters. This limitation will likely go away in a future version of SCons. Scanner Objects You can use the Scanner function to define objects to scan new file types for implicit dependencies. The Scanner function accepts the following arguments: function This can be either: 1) a Python function that will process the Node (file) and return a list of File Nodes representing the implicit dependencies (file names) found in the contents; or: 2) a dictionary that maps keys (typically the file suffix, but see below for more discussion) to other Scanners that should be called. If the argument is actually a Python function, the function must take three or four arguments: def scanner_function(node, env, path): def scanner_function(node, env, path, arg=None): The node argument is the internal SCons node representing the file. Use str(node) to fetch the name of the file, and node.get_contents() to fetch contents of the file. Note that the file is not guaranteed to exist before the scanner is called, so the scanner function should check that if there's any chance that the scanned file might not exist (for example, if it's built from other files). The env argument is the construction environment for the scan. Fetch values from it using the env.Dictionary() method. The path argument is a tuple (or list) of directories that can be searched for files. This will usually be the tuple returned by the path_function argument (see below). The arg argument is the argument supplied when the scanner was created, if any. name The name of the Scanner. This is mainly used to identify the Scanner internally. argument An optional argument that, if specified, will be passed to the scanner function (described above) and the path function (specified below). skeys An optional list that can be used to determine which scanner should be used for a given Node. In the usual case of scanning for file names, this argument will be a list of suffixes for the different file types that this Scanner knows how to scan. If the argument is a string, then it will be expanded into a list by the current environment. path_function A Python function that takes four or five arguments: a construction environment, a Node for the directory containing the SConscript file in which the first target was defined, a list of target nodes, a list of source nodes, and an optional argument supplied when the scanner was created. The path_function returns a tuple of directories that can be searched for files to be returned by this Scanner object. (Note that the FindPathDirs() function can be used to return a ready-made path_function for a given construction variable name, instead of having to write your own function from scratch.) node_class The class of Node that should be returned by this Scanner object. Any strings or other objects returned by the scanner function that are not of this class will be run through the node_factory function. node_factory A Python function that will take a string or other object and turn it into the appropriate class of Node to be returned by this Scanner object. scan_check An optional Python function that takes two arguments, a Node (file) and a construction environment, and returns whether the Node should, in fact, be scanned for dependencies. This check can be used to eliminate unnecessary calls to the scanner function when, for example, the underlying file represented by a Node does not yet exist. recursive An optional flag that specifies whether this scanner should be re-invoked on the dependency files returned by the scanner. When this flag is not set, the Node subsystem will only invoke the scanner on the file being scanned, and not (for example) also on the files specified by the #include lines in the file being scanned. recursive may be a callable function, in which case it will be called with a list of Nodes found and should return a list of Nodes that should be scanned recursively; this can be used to select a specific subset of Nodes for additional scanning. Note that scons has a global SourceFileScanner object that is used by the Object(), SharedObject(), and StaticObject() builders to decide which scanner should be used for different file extensions. You can using the SourceFileScanner.add_scanner() method to add your own Scanner object to the scons infrastructure that builds target programs or libraries from a list of source files of different types: def xyz_scan(node, env, path): contents = node.get_text_contents() # Scan the contents and return the included files. XYZScanner = Scanner(xyz_scan) SourceFileScanner.add_scanner('.xyz', XYZScanner) env.Program('my_prog', ['file1.c', 'file2.f', 'file3.xyz']) SYSTEM-SPECIFIC BEHAVIOR SCons and its configuration files are very portable, due largely to its implementation in Python. There are, however, a few portability issues waiting to trap the unwary. .C file suffix SCons handles the upper-case .C file suffix differently, depending on the capabilities of the underlying system. On a case-sensitive system such as Linux or UNIX, SCons treats a file with a .C suffix as a C++ source file. On a case-insensitive system such as Windows, SCons treats a file with a .C suffix as a C source file. .F file suffix SCons handles the upper-case .F file suffix differently, depending on the capabilities of the underlying system. On a case-sensitive system such as Linux or UNIX, SCons treats a file with a .F suffix as a Fortran source file that is to be first run through the standard C preprocessor. On a case-insensitive system such as Windows, SCons treats a file with a .F suffix as a Fortran source file that should not be run through the C preprocessor. Windows: Cygwin Tools and Cygwin Python vs. Windows Pythons Cygwin supplies a set of tools and utilities that let users work on a Windows system using a more POSIX-like environment. The Cygwin tools, including Cygwin Python, do this, in part, by sharing an ability to interpret UNIX-like path names. For example, the Cygwin tools will internally translate a Cygwin path name like /cygdrive/c/mydir to an equivalent Windows pathname of C:/mydir (equivalent to C:\mydir). Versions of Python that are built for native Windows execution, such as the python.org and ActiveState versions, do not have the Cygwin path name semantics. This means that using a native Windows version of Python to build compiled programs using Cygwin tools (such as gcc, bison, and flex) may yield unpredictable results. "Mixing and matching" in this way can be made to work, but it requires careful attention to the use of path names in your SConscript files. In practice, users can sidestep the issue by adopting the following rules: When using gcc, use the Cygwin-supplied Python interpreter to run SCons; when using Microsoft Visual C/C++ (or some other Windows compiler) use the python.org or ActiveState version of Python to run SCons. Windows: scons.bat file On Windows systems, SCons is executed via a wrapper scons.bat file. This has (at least) two ramifications: First, Windows command-line users that want to use variable assignment on the command line may have to put double quotes around the assignments: scons "FOO=BAR" "BAZ=BLEH" Second, the Cygwin shell does not recognize this file as being the same as an scons command issued at the command-line prompt. You can work around this either by executing scons.bat from the Cygwin command line, or by creating a wrapper shell script named scons . MinGW The MinGW bin directory must be in your PATH environment variable or the PATH variable under the ENV construction variable for SCons to detect and use the MinGW tools. When running under the native Windows Python interpreter, SCons will prefer the MinGW tools over the Cygwin tools, if they are both installed, regardless of the order of the bin directories in the PATH variable. If you have both MSVC and MinGW installed and you want to use MinGW instead of MSVC, then you must explictly tell SCons to use MinGW by passing tools=['mingw'] to the Environment() function, because SCons will prefer the MSVC tools over the MinGW tools. EXAMPLES To help you get started using SCons, this section contains a brief overview of some common tasks. Basic Compilation From a Single Source File env = Environment() env.Program(target = 'foo', source = 'foo.c') Note: Build the file by specifying the target as an argument ("scons foo" or "scons foo.exe"). or by specifying a dot ("scons ."). Basic Compilation From Multiple Source Files env = Environment() env.Program(target = 'foo', source = Split('f1.c f2.c f3.c')) Setting a Compilation Flag env = Environment(CCFLAGS = '-g') env.Program(target = 'foo', source = 'foo.c') Search The Local Directory For .h Files Note: You do not need to set CCFLAGS to specify -I options by hand. SCons will construct the right -I options from CPPPATH. env = Environment(CPPPATH = ['.']) env.Program(target = 'foo', source = 'foo.c') Search Multiple Directories For .h Files env = Environment(CPPPATH = ['include1', 'include2']) env.Program(target = 'foo', source = 'foo.c') Building a Static Library env = Environment() env.StaticLibrary(target = 'foo', source = Split('l1.c l2.c')) env.StaticLibrary(target = 'bar', source = ['l3.c', 'l4.c']) Building a Shared Library env = Environment() env.SharedLibrary(target = 'foo', source = ['l5.c', 'l6.c']) env.SharedLibrary(target = 'bar', source = Split('l7.c l8.c')) Linking a Local Library Into a Program env = Environment(LIBS = 'mylib', LIBPATH = ['.']) env.Library(target = 'mylib', source = Split('l1.c l2.c')) env.Program(target = 'prog', source = ['p1.c', 'p2.c']) Defining Your Own Builder Object Notice that when you invoke the Builder, you can leave off the target file suffix, and SCons will add it automatically. bld = Builder(action = 'pdftex < $SOURCES > $TARGET' suffix = '.pdf', src_suffix = '.tex') env = Environment(BUILDERS = {'PDFBuilder' : bld}) env.PDFBuilder(target = 'foo.pdf', source = 'foo.tex') # The following creates "bar.pdf" from "bar.tex" env.PDFBuilder(target = 'bar', source = 'bar') Note also that the above initialization overwrites the default Builder objects, so the Environment created above can not be used call Builders like env.Program(), env.Object(), env.StaticLibrary(), etc. Adding Your Own Builder Object to an Environment bld = Builder(action = 'pdftex < $SOURCES > $TARGET' suffix = '.pdf', src_suffix = '.tex') env = Environment() env.Append(BUILDERS = {'PDFBuilder' : bld}) env.PDFBuilder(target = 'foo.pdf', source = 'foo.tex') env.Program(target = 'bar', source = 'bar.c') You also can use other Pythonic techniques to add to the BUILDERS construction variable, such as: env = Environment() env['BUILDERS]['PDFBuilder'] = bld Defining Your Own Scanner Object The following example shows an extremely simple scanner (the kfile_scan() function) that doesn't use a search path at all and simply returns the file names present on any include lines in the scanned file. This would implicitly assume that all included files live in the top-level directory: import re include_re = re.compile(r'^include\s+(\S+)$', re.M) def kfile_scan(node, env, path, arg): contents = node.get_text_contents() includes = include_re.findall(contents) return env.File(includes) kscan = Scanner(name = 'kfile', function = kfile_scan, argument = None, skeys = ['.k']) scanners = Environment().Dictionary('SCANNERS') env = Environment(SCANNERS = scanners + [kscan]) env.Command('foo', 'foo.k', 'kprocess < $SOURCES > $TARGET') bar_in = File('bar.in') env.Command('bar', bar_in, 'kprocess $SOURCES > $TARGET') bar_in.target_scanner = kscan It is important to note that you have to return a list of File nodes from the scan function, simple strings for the file names won't do. As in the examples we are showing here, you can use the File() function of your current Environment in order to create nodes on the fly from a sequence of file names with relative paths. Here is a similar but more complete example that searches a path of directories (specified as the MYPATH construction variable) for files that actually exist: import re import os include_re = re.compile(r'^include\s+(\S+)$', re.M) def my_scan(node, env, path, arg): contents = node.get_text_contents() includes = include_re.findall(contents) if includes == []: return [] results = [] for inc in includes: for dir in path: file = str(dir) + os.sep + inc if os.path.exists(file): results.append(file) break return env.File(results) scanner = Scanner(name = 'myscanner', function = my_scan, argument = None, skeys = ['.x'], path_function = FindPathDirs('MYPATH') ) scanners = Environment().Dictionary('SCANNERS') env = Environment(SCANNERS = scanners + [scanner], MYPATH = ['incs']) env.Command('foo', 'foo.x', 'xprocess < $SOURCES > $TARGET') The FindPathDirs() function used in the previous example returns a function (actually a callable Python object) that will return a list of directories specified in the $MYPATH construction variable. It lets SCons detect the file incs/foo.inc , even if foo.x contains the line include foo.inc only. If you need to customize how the search path is derived, you would provide your own path_function argument when creating the Scanner object, as follows: # MYPATH is a list of directories to search for files in def pf(env, dir, target, source, arg): top_dir = Dir('#').abspath results = [] if 'MYPATH' in env: for p in env['MYPATH']: results.append(top_dir + os.sep + p) return results scanner = Scanner(name = 'myscanner', function = my_scan, argument = None, skeys = ['.x'], path_function = pf ) Creating a Hierarchical Build Notice that the file names specified in a subdirectory's SConscript file are relative to that subdirectory. SConstruct: env = Environment() env.Program(target = 'foo', source = 'foo.c') SConscript('sub/SConscript') sub/SConscript: env = Environment() # Builds sub/foo from sub/foo.c env.Program(target = 'foo', source = 'foo.c') SConscript('dir/SConscript') sub/dir/SConscript: env = Environment() # Builds sub/dir/foo from sub/dir/foo.c env.Program(target = 'foo', source = 'foo.c') Sharing Variables Between SConscript Files You must explicitly Export() and Import() variables that you want to share between SConscript files. SConstruct: env = Environment() env.Program(target = 'foo', source = 'foo.c') Export("env") SConscript('subdirectory/SConscript') subdirectory/SConscript: Import("env") env.Program(target = 'foo', source = 'foo.c') Building Multiple Variants From the Same Source Use the variant_dir keyword argument to the SConscript function to establish one or more separate variant build directory trees for a given source directory: SConstruct: cppdefines = ['FOO'] Export("cppdefines") SConscript('src/SConscript', variant_dir='foo') cppdefines = ['BAR'] Export("cppdefines") SConscript('src/SConscript', variant_dir='bar') src/SConscript: Import("cppdefines") env = Environment(CPPDEFINES = cppdefines) env.Program(target = 'src', source = 'src.c') Note the use of the Export() method to set the "cppdefines" variable to a different value each time we call the SConscript function. Hierarchical Build of Two Libraries Linked With a Program SConstruct: env = Environment(LIBPATH = ['#libA', '#libB']) Export('env') SConscript('libA/SConscript') SConscript('libB/SConscript') SConscript('Main/SConscript') libA/SConscript: Import('env') env.Library('a', Split('a1.c a2.c a3.c')) libB/SConscript: Import('env') env.Library('b', Split('b1.c b2.c b3.c')) Main/SConscript: Import('env') e = env.Copy(LIBS = ['a', 'b']) e.Program('foo', Split('m1.c m2.c m3.c')) The '#' in the LIBPATH directories specify that they're relative to the top-level directory, so they don't turn into "Main/libA" when they're used in Main/SConscript. Specifying only 'a' and 'b' for the library names allows SCons to append the appropriate library prefix and suffix for the current platform (for example, 'liba.a' on POSIX systems, 'a.lib' on Windows). Customizing construction variables from the command line. The following would allow the C compiler to be specified on the command line or in the file custom.py. vars = Variables('custom.py') vars.Add('CC', 'The C compiler.') env = Environment(variables=vars) Help(vars.GenerateHelpText(env)) The user could specify the C compiler on the command line: scons "CC=my_cc" or in the custom.py file: CC = 'my_cc' or get documentation on the options: $ scons -h CC: The C compiler. default: None actual: cc Using Microsoft Visual C++ precompiled headers Since windows.h includes everything and the kitchen sink, it can take quite some time to compile it over and over again for a bunch of object files, so Microsoft provides a mechanism to compile a set of headers once and then include the previously compiled headers in any object file. This technology is called precompiled headers. The general recipe is to create a file named "StdAfx.cpp" that includes a single header named "StdAfx.h", and then include every header you want to precompile in "StdAfx.h", and finally include "StdAfx.h" as the first header in all the source files you are compiling to object files. For example: StdAfx.h: #include <windows.h> #include <my_big_header.h> StdAfx.cpp: #include <StdAfx.h> Foo.cpp: #include <StdAfx.h> /* do some stuff */ Bar.cpp: #include <StdAfx.h> /* do some other stuff */ SConstruct: env=Environment() env['PCHSTOP'] = 'StdAfx.h' env['PCH'] = env.PCH('StdAfx.cpp')[0] env.Program('MyApp', ['Foo.cpp', 'Bar.cpp']) For more information see the document for the PCH builder, and the PCH and PCHSTOP construction variables. To learn about the details of precompiled headers consult the MSDN documention for /Yc, /Yu, and /Yp. Using Microsoft Visual C++ external debugging information Since including debugging information in programs and shared libraries can cause their size to increase significantly, Microsoft provides a mechanism for including the debugging information in an external file called a PDB file. SCons supports PDB files through the PDB construction variable. SConstruct: env=Environment() env['PDB'] = 'MyApp.pdb' env.Program('MyApp', ['Foo.cpp', 'Bar.cpp']) For more information see the document for the PDB construction variable. ENVIRONMENT SCONS_LIB_DIR Specifies the directory that contains the SCons Python module directory (e.g. /home/aroach/scons-src-0.01/src/engine). SCONSFLAGS A string of options that will be used by scons in addition to those passed on the command line. SEE ALSO scons User Manual, scons Design Document, scons source code. AUTHORS Steven Knight <knight@baldmt.com> Anthony Roach <aroach@electriceyeball.com>