SCons Principles

There are a few overriding principles we try to live up to in designing and implementing SCons:

In a nutshell, we try hard to make SCons just "do the right thing" and build software correctly, with a minimum of hassles.

Acknowledgements

SCons would not exist without a lot of help from a lot of people, many of whom may not even be aware that they helped or served as inspiration. So in no particular order, and at the risk of leaving out someone:

First and foremost, SCons owes a tremendous debt to Bob Sidebotham, the original author of the classic Perl-based Cons tool which Bob first released to the world back around 1996. Bob's work on Cons classic provided the underlying architecture and model of specifying a build configuration using a real scripting language. My real-world experience working on Cons informed many of the design decisions in SCons, including the improved parallel build support, making Builder objects easily definable by users, and separating the build engine from the wrapping interface.

Greg Wilson was instrumental in getting SCons started as a real project when he initiated the Software Carpentry design competition in February 2000. Without that nudge, marrying the advantages of the Cons classic architecture with the readability of Python might have just stayed no more than a nice idea.

The entire SCons team have been absolutely wonderful to work with, and SCons would be nowhere near as useful a tool without the energy, enthusiasm and time people have contributed over the past few years. The "core team" of Chad Austin, Anthony Roach, Charles Crain, Steve Leblanc, Gary Oerbrunner, Greg Spencer and Christoph Wiedemann have been great about reviewing my (and other) changes and catching problems before they get in the code base. Of particular technical note: Anthony's outstanding and innovative work on the tasking engine has given SCons a vastly superior parallel build model; Charles has been the master of the crucial Node infrastructure; Christoph's work on the Configure infrastructure has added crucial Autoconf-like functionality; and Greg has provided excellent support for Microsoft Visual Studio.

Special thanks to David Snopek for contributing his underlying "Autoscons" code that formed the basis of Christoph's work with the Configure functionality. David was extremely generous in making this code available to SCons, given that he initially released it under the GPL and SCons is released under a less-restrictive MIT-style license.

Thanks to Peter Miller for his splendid change management system, Aegis, which has provided the SCons project with a robust development methodology from day one, and which showed me how you could integrate incremental regression tests into a practical development cycle (years before eXtreme Programming arrived on the scene).

And last, thanks to Guido van Rossum for his elegant scripting language, which is the basis not only for the SCons implementation, but for the interface itself.

Contact

The best way to contact people involved with SCons, including the author, is through the SCons mailing lists.

If you want to ask general questions about how to use SCons send email to scons-users@lists.sourceforge.net.

If you want to contact the SCons development community directly, send email to scons-devel@lists.sourceforge.net.

If you want to receive announcements about SCons, join the low-volume scons-announce@lists.sourceforge.net mailing list.

Cleaning Up After a Build

When using SCons, it is unnecessary to add special commands or target names to clean up after a build. Instead, you simply use the -c or --clean option when you invoke SCons, and SCons removes the appropriate built files. So if we build our example above and then invoke scons -c afterwards, the output on POSIX looks like:

      % scons
      scons: Reading SConscript files ...
      scons: done reading SConscript files.
      scons: Building targets ...
      cc -c -o hello.o hello.c
      cc -o hello hello.o
      scons: done building targets.
      % scons -c
      scons: Reading SConscript files ...
      scons: done reading SConscript files.
      scons: Cleaning targets ...
      Removed hello.o
      Removed hello
      scons: done cleaning targets.
   

And the output on Windows looks like:

      C:\>scons
      scons: Reading SConscript files ...
      scons: done reading SConscript files.
      scons: Building targets ...
      cl /nologo /c hello.c /Fohello.obj
      link /nologo /OUT:hello.exe hello.obj
      scons: done building targets.
      C:\>scons -c
      scons: Reading SConscript files ...
      scons: done reading SConscript files.
      scons: Cleaning targets ...
      Removed hello.obj
      Removed hello.exe
      scons: done cleaning targets.
   

Notice that SCons changes its output to tell you that it is Cleaning targets ... and done cleaning targets.

The SConstruct File

If you're used to build systems like Make you've already figured out that the SConstruct file is the SCons equivalent of a Makefile. That is, the SConstruct file is the input file that SCons reads to control the build.

There is, however, an important difference between an SConstruct file and a Makefile: the SConstruct file is actually a Python script. If you're not already familiar with Python, don't worry. This User's Guide will introduce you step-by-step to the relatively small amount of Python you'll need to know to be able to use SCons effectively. And Python is very easy to learn.

One aspect of using Python as the scripting language is that you can put comments in your SConstruct file using Python's commenting convention; that is, everything between a '#' and the end of the line will be ignored:

      # Arrange to build the "hello" program.
      Program('hello.c')    # "hello.c" is the source file.
   

You'll see throughout the remainder of this Guide that being able to use the power of a real scripting language can greatly simplify the solutions to complex requirements of real-world builds.

Making the Output Less Verbose

You've already seen how SCons prints some messages about what it's doing, surrounding the actual commands used to build the software:

      C:\>scons
      scons: Reading SConscript files ...
      scons: done reading SConscript files.
      scons: Building targets ...
      cl /nologo /c hello.c /Fohello.obj
      link /nologo /OUT:hello.exe hello.obj
      scons: done building targets.
   

These messages emphasize the order in which SCons does its work: the configuration files (generically referred to as SConscript files) are read and executed first, and only then are the target files built. Among other benefits, these messages help to distinguish between errors that occur while the configuration files are read, and errors that occur while targets are being built.

The drawback, of course, is that these messages clutter the output. Fortunately, they're easily disabled by using the -Q option when invoking SCons:

      C:\>scons -Q
      cl /nologo /c hello.c /Fohello.obj
      link /nologo /OUT:hello.exe hello.obj
   

Because we want this User's Guide to focus on what SCons is actually doing, we're going use the -Q option to remove these messages from the output of all the remaining examples in this Guide.

Compiling Multiple Source Files

You've just seen how to configure SCons to compile a program from a single source file. It's more common, of course, that you'll need to build a program from many input source files, not just one. To do this, you need to put the source files in a Python list (enclosed in square brackets), like so:

      Program(['prog.c', 'file1.c', 'file2.c'])
   

A build of the above example would look like:

      % scons -Q
      cc -c -o file1.o file1.c
      cc -c -o file2.o file2.c
      cc -c -o prog.o prog.c
      cc -o prog prog.o file1.o file2.o
   

Notice that SCons deduces the output program name from the first source file specified in the list--that is, because the first source file was prog.c, SCons will name the resulting program prog (or prog.exe on a Windows system). If you want to specify a different program name, then you slide the list of source files over to the right to make room for the output program file name. (SCons puts the output file name to the left of the source file names so that the order mimics that of an assignment statement: "program = source files".) This makes our example:

      Program('program', ['main.c', 'file1.c', 'file2.c'])
   

On Linux, a build of this example would look like:

      % scons -Q
      cc -c -o file1.o file1.c
      cc -c -o file2.o file2.c
      cc -c -o main.o main.c
      cc -o program main.o file1.o file2.o
   

Or on Windows:

      C:\>scons -Q
      cl /nologo /c file1.c /Fofile1.obj
      cl /nologo /c file2.c /Fofile2.obj
      cl /nologo /c main.c /Fomain.obj
      link /nologo /OUT:program.exe main.obj file1.obj file2.obj
   

Keeping SConstruct Files Easy to Read

One drawback to the use of a Python list for source files is that each file name must be enclosed in quotes (either single quotes or double quotes). This can get cumbersome and difficult to read when the list of file names is long. Fortunately, SCons and Python provide a number of ways to make sure that the SConstruct file stays easy to read.

To make long lists of file names easier to deal with, SCons provides a Split function that takes a quoted list of file names, with the names separated by spaces or other white-space characters, and turns it into a list of separate file names. Using the Split function turns the previous example into:

      Program('program', Split('main.c file1.c file2.c'))
   

(If you're already familiar with Python, you'll have realized that this is similar to the split() method in the Python standard string module. Unlike the string.split() method, however, the Split function does not require a string as input and will wrap up a single non-string object in a list, or return its argument untouched if it's already a list. This comes in handy as a way to make sure arbitrary values can be passed to SCons functions without having to check the type of the variable by hand.)

Putting the call to the Split function inside the Program call can also be a little unwieldy. A more readable alternative is to assign the output from the Split call to a variable name, and then use the variable when calling the Program function:

      list = Split('main.c file1.c file2.c')
      Program('program', list)
   

Lastly, the Split function doesn't care how much white space separates the file names in the quoted string. This allows you to create lists of file names that span multiple lines, which often makes for easier editing:

      list = Split('main.c
                    file1.c
                    file2.c')
      Program('program', list)
   

Keyword Arguments

SCons also allows you to identify the output file and input source files using Python keyword arguments. The output file is known as the target, and the source file(s) are known (logically enough) as the source. The Python syntax for this is:

      list = Split('main.c file1.c file2.c')
      Program(target = 'program', source = list)
   

Because the keywords explicitly identify what each argument is, you can actually reverse the order if you prefer:

      list = Split('main.c file1.c file2.c')
      Program(source = list, target = 'program')
   

Whether or not you choose to use keyword arguments to identify the target and source files, and the order in which you specify them when using keywords, are purely personal choices; SCons functions the same regardless.

Compiling Multiple Programs

In order to compile multiple programs within the same SConstruct file, simply call the Program method multiple times, once for each program you need to build:

      Program('foo.c')
      Program('bar', ['bar1.c', 'bar2.c'])
   

SCons would then build the programs as follows:

      % scons -Q
      cc -c -o bar1.o bar1.c
      cc -c -o bar2.o bar2.c
      cc -o bar bar1.o bar2.o
      cc -c -o foo.o foo.c
      cc -o foo foo.o
   

Notice that SCons does not necessarily build the programs in the same order in which you specify them in the SConstruct file. SCons does, however, recognize that the individual object files must be built before the resulting program can be built. We'll discuss this in greater detail in the "Dependencies" section, below.

Sharing Source Files Between Multiple Programs

It's common to re-use code by sharing source files between multiple programs. One way to do this is to create a library from the common source files, which can then be linked into resulting programs. (Creating libraries is discussed in section XXX, below.)

A more straightforward, but perhaps less convenient, way to share source files between multiple programs is simply to include the common files in the lists of source files for each program:

      Program(Split('foo.c common1.c common2.c'))
      Program('bar', Split('bar1.c bar2.c common1.c common2.c'))
   

SCons recognizes that the object files for the common1.c and common2.c source files each only need to be built once, even though the resulting object files are each linked in to both of the resulting executable programs:

      % scons -Q
      cc -c -o bar1.o bar1.c
      cc -c -o bar2.o bar2.c
      cc -c -o common1.o common1.c
      cc -c -o common2.o common2.c
      cc -o bar bar1.o bar2.o common1.o common2.o
      cc -c -o foo.o foo.c
      cc -o foo foo.o common1.o common2.o
   

If two or more programs share a lot of common source files, repeating the common files in the list for each program can be a maintenance problem when you need to change the list of common files. You can simplify this by creating a separate Python list to hold the common file names, and concatenating it with other lists using the Python + operator:

      common = ['common1.c', 'common2.c']
      foo_files = ['foo.c'] + common
      bar_files = ['bar1.c', 'bar2.c'] + common
      Program('foo', foo_files)
      Program('bar', bar_files)
   

This is functionally equivalent to the previous example.

Building Libraries

You build your own libraries by specifying Library instead of Program:

      Library('foo', ['f1.c', 'f2.c', 'f3.c'])
    

SCons uses the appropriate library prefix and suffix for your system. So on POSIX or Linux systems, the above example would build as follows (although ranlib may not be called on all systems):

      % scons -Q
      cc -c -o f1.o f1.c
      cc -c -o f2.o f2.c
      cc -c -o f3.o f3.c
      ar r libfoo.a f1.o f2.o f3.o
      ranlib libfoo.a
    

On a Windows system, a build of the above example would look like:

      C:\>scons -Q
      cl /nologo /c f1.c /Fof1.obj
      cl /nologo /c f2.c /Fof2.obj
      cl /nologo /c f3.c /Fof3.obj
      lib /nologo /OUT:foo.lib f1.obj f2.obj f3.obj
    

The rules for the target name of the library are similar to those for programs: if you don't explicitly specify a target library name, SCons will deduce one from the name of the first source file specified, and SCons will add an appropriate file prefix and suffix if you leave them off.

        StaticLibrary('foo', ['f1.c', 'f2.c', 'f3.c'])
      

        SharedLibrary('foo', ['f1.c', 'f2.c', 'f3.c'])
      

Linking with Libraries

Usually, you build a library because you want to link it with one or more programs. You link libraries with a program by specifying the libraries in the LIBS construction variable, and by specifying the directory in which the library will be found in the LIBPATH construction variable:

      Library('foo', ['f1.c', 'f2.c', 'f3.c'])
      Program('prog.c', LIBS='foo', LIBPATH='.')
    

Notice, of course, that you don't need to specify a library prefix (like lib) or suffix (like .a or .lib). SCons uses the correct prefix or suffix for the current system.

On a POSIX or Linux system, a build of the above example would look like:

      % scons -Q
      cc -c -o f1.o f1.c
      cc -c -o f2.o f2.c
      cc -c -o f3.o f3.c
      ar r libfoo.a f1.o f2.o f3.o
      ranlib libfoo.a
      cc -c -o prog.o prog.c
      cc -o prog prog.o -L. -lfoo
    

On a Windows system, a build of the above example would look like:

      C:\>scons -Q
      cl /nologo /c f1.c /Fof1.obj
      cl /nologo /c f2.c /Fof2.obj
      cl /nologo /c f3.c /Fof3.obj
      lib /nologo /OUT:foo.lib f1.obj f2.obj f3.obj
      cl /nologo /c prog.c /Foprog.obj
      link /nologo /OUT:prog.exe /LIBPATH:. foo.lib prog.obj
    

As usual, notice that SCons has taken care of constructing the correct command lines to link with the specified library on each system.

Finding Libraries: the LIBPATH Construction Variable

By default, the linker will only look in certain system-defined directories for libraries. SCons knows how to look for libraries in directories that you specify with the LIBPATH construction variable. LIBPATH consists of a list of directory names, like so:

      Program('prog.c', LIBS = 'm',
                        LIBPATH = ['/usr/lib', '/usr/local/lib'])
    

Using a Python list is preferred because it's portable across systems. Alternatively, you could put all of the directory names in a single string, separated by the system-specific path separator character: a colon on POSIX systems:

      LIBPATH = '/usr/lib:/usr/local/lib'
    

or a semi-colon on Windows systems:

      LIBPATH = 'C:\lib;D:\lib'
    

When the linker is executed, SCons will create appropriate flags so that the linker will look for libraries in the same directories as SCons. So on a POSIX or Linux system, a build of the above example would look like:

      % scons -Q
      cc -c -o prog.o prog.c
      cc -o prog prog.o -L/usr/lib -L/usr/local/lib -lm
    

On a Windows system, a build of the above example would look like:

      C:\>scons -Q
      cl /nologo /c prog.c /Foprog.obj
      link /nologo /OUT:prog.exe /LIBPATH:\usr\lib /LIBPATH:\usr\local\lib m.lib prog.obj
    

Note again that SCons has taken care of the system-specific details of creating the right command-line options.

Source File Signatures

The other side of avoiding unnecessary rebuilds is the fundamental build tool behavior of rebuilding things when a source file changes, so that the built software is up to date. SCons keeps track of this through a signature for each source file, and allows you to configure whether you want to use the source file contents or the modification time (timestamp) as the signature.

        Program('hello.c')
        SourceSignatures('MD5')
      

        Program('hello.c')
        SourceSignatures('timestamp')
      

Target File Signatures

As you've just seen, SCons uses signatures to decide whether a target file is up to date or must be rebuilt. When a target file depends on another target file, SCons allows you to separately configure how the signatures of "intermediate" target files are used when deciding if a dependent target file must be rebuilt.

        Program('hello.c')
        TargetSignatures('build')
      

        Program('hello.c')
        TargetSignatures('content')
      

Implicit Dependencies: The CPPPATH Construction Variable

Now suppose that our "Hello, World!" program actually has a #include line to include the hello.h file in the compilation:

       #include <hello.h>
       int
       main()
       {
           printf("Hello, %s!\n", string);
       }
    

And, for completeness, the hello.h file looks like this:

       #define string    "world"
      

In this case, we want SCons to recognize that, if the contents of the hello.h file change, the hello program must be recompiled. To do this, we need to modify the SConstruct file like so:

       Program('hello.c', CPPPATH = '.')
      

The CPPPATH value tells SCons to look in the current directory ('.') for any files included by C source files (.c or .h files). With this assignment in the SConstruct file:

       % scons -Q hello
       cc -I. -c -o hello.o hello.c
       cc -o hello hello.o
       % scons -Q hello
       scons: `hello' is up to date.
       % edit hello.h
           [CHANGE THE CONTENTS OF hello.h]
       % scons -Q hello
       cc -I. -c -o hello.o hello.c
       cc -o hello hello.o
    

First, notice that SCons added the -I. argument from the CPPPATH variable so that the compilation would find the hello.h file in the local directory.

Second, realize that SCons knows that the hello program must be rebuilt because it scans the contents of the hello.c file for the #include lines that indicate another file is being included in the compilation. SCons records these as implicit dependencies of the target file, Consequently, when the hello.h file changes, SCons realizes that the hello.c file includes it, and rebuilds the resulting hello program that depends on both the hello.c and hello.h files.

Like the LIBPATH variable, the CPPPATH variable may be a list of directories, or a string separated by the system-specific path separate character (':' on POSIX/Linux, ';' on Windows). Either way, SCons creates the right command-line options so that the following example:

       Program('hello.c', CPPPATH = ['include', '/home/project/inc'])
    

Will look like this on POSIX or Linux:

       % scons -Q hello
       cc -Iinclude -I/home/project/inc -c -o hello.o hello.c
       cc -o hello hello.o
    

And like this on Windows:

       C:\>scons -Q hello.exe
       cl /nologo /Iinclude /I\home\project\inc /c hello.c /Fohello.obj
       link /nologo /OUT:hello.exe hello.obj
    

Caching Implicit Dependencies

Scanning each file for #include lines does take some extra processing time. When you're doing a full build of a large system, the scanning time is usually a very small percentage of the overall time spent on the build. You're most likely to notice the scanning time, however, when you rebuild all or part of a large system: SCons will likely take some extra time to "think about" what must be built before it issues the first build command (or decides that everything is up to date and nothing must be rebuilt).

In practice, having SCons scan files saves time relative to the amount of potential time lost to tracking down subtle problems introduced by incorrect dependencies. Nevertheless, the "waiting time" while SCons scans files can annoy individual developers waiting for their builds to finish. Consequently, SCons lets you cache the implicit dependencies that its scanners find, for use by later builds. You can do this by specifying the --implicit-cache option on the command line:

       % scons -Q --implicit-cache hello
       cc -c -o hello.o hello.c
       cc -o hello hello.o
       % scons -Q hello
       scons: `hello' is up to date.
    

If you don't want to specify --implicit-cache on the command line each time, you can make it the default behavior for your build by setting the implicit_cache option in an SConscript file:

       SetOption('implicit_cache', 1)
    

The Ignore Method

Sometimes it makes sense to not rebuild a program, even if a dependency file changes. In this case, you would tell SCons specifically to ignore a dependency as follows:

      hello = Program('hello.c')
      Ignore(hello, 'hello.h')
    

      % scons -Q hello
      cc -c -o hello.o hello.c
      cc -o hello hello.o
      % scons -Q hello
      scons: `hello' is up to date.
      % edit hello.h
        [CHANGE THE CONTENTS OF hello.h]
      % scons -Q hello
      scons: `hello' is up to date.
    

Now, the above example is a little contrived, because it's hard to imagine a real-world situation where you wouldn't to rebuild hello if the hello.h file changed. A more realistic example might be if the hello program is being built in a directory that is shared between multiple systems that have different copies of the stdio.h include file. In that case, SCons would notice the differences between the different systems' copies of stdio.h and would rebuild hello each time you change systems. You could avoid these rebuilds as follows:

       hello = Program('hello.c')
       Ignore(hello, '/usr/include/stdio.h')
    

The Depends Method

On the other hand, sometimes a file depends on another file that is not detected by an SCons scanner. For this situation, SCons allows you to specific explicitly that one file depends on another file, and must be rebuilt whenever that file changes. This is specified using the Depends method:

       hello = Program('hello.c')
       Depends(hello, 'other_file')
    

       % scons -Q hello
       cc -c hello.c -o hello.o
       cc -o hello hello.o
       % scons -Q hello
       scons: `hello' is up to date.
       % edit other_file
           [CHANGE THE CONTENTS OF other_file]
       % scons -Q hello
       cc -c hello.c -o hello.o
       cc -o hello hello.o
    

Multiple Construction Environments

The real advantage of construction environments is that you can create as many different construction environments as you need, each tailored to a different way to build some piece of software or other file. If, for example, we need to build one program with the -O2 flag and another with the -g (debug) flag, we would do this like so:

      opt = Environment(CCFLAGS = '-O2')
      dbg = Environment(CCFLAGS = '-g')

      opt.Program('foo', 'foo.c')

      dbg.Program('bar', 'bar.c')
   

      % scons -Q
      cc -g -c -o bar.o bar.c
      cc -o bar bar.o
      cc -O2 -c -o foo.o foo.c
      cc -o foo foo.o
   

We can even use multiple construction environments to build multiple versions of a single program. If you do this by simply trying to use the Program builder with both environments, though, like this:

      opt = Environment(CCFLAGS = '-O2')
      dbg = Environment(CCFLAGS = '-g')

      opt.Program('foo', 'foo.c')

      dbg.Program('foo', 'foo.c')
   

Then SCons generates the following error:

      % scons -Q
      
      scons: *** Two environments with different actions were specified for the same target: foo.o
      File "SConstruct", line 6, in ?
   

This is because the two Program calls have each implicitly told SCons to generate an object file named foo.o, one with a CCFLAGS value of -O2 and one with a CCFLAGS value of -g. SCons can't just decide that one of them should take precedence over the other, so it generates the error. To avoid this problem, we must explicitly specify that each environment compile foo.c to a separately-named object file using the Object call, like so:

      opt = Environment(CCFLAGS = '-O2')
      dbg = Environment(CCFLAGS = '-g')

      o = opt.Object('foo-opt', 'foo.c')
      opt.Program(o)

      d = dbg.Object('foo-dbg', 'foo.c')
      dbg.Program(d)
   

Notice that each call to the Object builder returns a value, an internal SCons object that represents the object file that will be built. We then use that object as input to the Program builder. This avoids having to specify explicitly the object file name in multiple places, and makes for a compact, readable SConstruct file. Our SCons output then looks like:

      % scons -Q
      cc -g -c -o foo-dbg.o foo.c
      cc -o foo-dbg foo-dbg.o
      cc -O2 -c -o foo-opt.o foo.c
      cc -o foo-opt foo-opt.o
   

Copying Construction Environments

Sometimes you want more than one construction environment to share the same values for one or more variables. Rather than always having to repeat all of the common variables when you create each construction environment, you can use the Copy method to create a copy of a construction environment.

Like the Environment call that creates a construction environment, the Copy method takes construction variable assignments, which will override the values in the copied construction environment. For example, suppose we want to use gcc to create three versions of a program, one optimized, one debug, and one with neither. We could do this by creating a "base" construction environment that sets CC to gcc, and then creating two copies, one which sets CCFLAGS for optimization and the other which sets CCFLAGS for debugging:

      env = Environment(CC = 'gcc')
      opt = env.Copy(CCFLAGS = '-O2')
      dbg = env.Copy(CCFLAGS = '-g')

      env.Program('foo', 'foo.c')

      o = opt.Object('foo-opt', 'foo.c')
      opt.Program(o)

      d = dbg.Object('foo-dbg', 'foo.c')
      dbg.Program(d)
   

Then our output would look like:

      % scons -Q
      gcc -c -o foo.o foo.c
      gcc -o foo foo.o
      gcc -g -c -o foo-dbg.o foo.c
      gcc -o foo-dbg foo-dbg.o
      gcc -O2 -c -o foo-opt.o foo.c
      gcc -o foo-opt foo-opt.o
   

Fetching Values From a Construction Environment

You can fetch individual construction variables using the normal syntax for accessing individual named items in a Python dictionary:

      env = Environment()
      print "CC is:", env['CC']
   

This example SConstruct file doesn't build anything, but because it's actually a Python script, it will print the value of CC for us:

      % scons -Q
      CC is: cc
      scons: `.' is up to date.
   

A construction environment, however, is actually a Python object with associated methods, etc. If you want to have direct access to only the dictionary of construction variables, you can fetch this using the Dictionary method:

      env = Environment(FOO = 'foo', BAR = 'bar')
      dict = env.Dictionary()
      for key in ['OBJSUFFIX', 'LIBSUFFIX', 'PROGSUFFIX']:
          print "key = %s, value = %s" % (key, dict[key])
   

This SConstruct file will print the specified dictionary items for us on POSIX systems as follows:

      % scons -Q
      key = OBJSUFFIX, value = .o
      key = LIBSUFFIX, value = .a
      key = PROGSUFFIX, value = 
      scons: `.' is up to date.
   

And on Win32:

      C:\>scons -Q
      key = OBJSUFFIX, value = .obj
      key = LIBSUFFIX, value = .lib
      key = PROGSUFFIX, value = .exe
      scons: `.' is up to date.
   

Modifying a Construction Environment

SCons provides various methods that support modifying existing values in a construction environment.

        env = Environment(CCFLAGS = '-DDEFINE1')
        env.Program('foo.c')
        env.Replace(CCFLAGS = '-DDEFINE2')
        env.Program('bar.c')
     

        env = Environment(CCFLAGS = '-DMY_VALUE')
        env.Append(CCFLAGS = ' -DLAST')
        env.Program('foo.c')
     

        env = Environment(CCFLAGS = '-DMY_VALUE')
        env.Prepend(CCFLAGS = '-DFIRST ')
        env.Program('foo.c')
     

Propagating PATH From the External Environment

You may want to propagate the external PATH to the execution environment for commands. You do this by initializing the PATH variable with the PATH value from the os.environ dictionary, which is Python's way of letting you get at the external environment:

      import os
      env = Environment(ENV = {'PATH' : os.environ['PATH']})
    

Alternatively, you may find it easier to just propagate the entire external environment to the execution environment for commands. This is simpler to code than explicity selecting the PATH value:

      import os
      env = Environment(ENV = os.environ)
    

Either of these will guarantee that SCons will be able to execute any command that you can execute from the command line. The drawback is that the build can behave differently if it's run by people with different PATH values in their environment--for example, both the /bin and /usr/local/bin directories have different cc commands, then which one will be used to compile programs will depend on which directory is listed first in the user's PATH variable.

Not Having to Specify Command-Line Options Each Time: the SCONSFLAGS Environment Variable

Users may find themselves supplying the same command-line options every time they run SCons. For example, a user might find that it saves time to specify a value of -j 2 to run the builds in parallel. To avoid having to type -j 2 by hand every time, you can set the external environment variable SCONSFLAGS to a string containing command-line options that you want SCons to use.

If, for example, and you're using a POSIX shell that's compatible with the Bourne shell, and you always want SCons to use the -Q option, you can set the SCONSFLAGS environment as follows:

      % scons
      scons: Reading SConscript files ...
      scons: done reading SConscript files.
      scons: Building targets ...
          ... [build output] ...
      scons: done building targets.
      % export SCONSFLAGS="-Q"
      % scons
          ... [build output] ...
    

Users of csh-style shells on POSIX systems can set the SCONSFLAGS environment as follows:

      $ setenv SCONSFLAGS "-Q"
    

Windows users may typically want to set this SCONSFLAGS in the appropriate tab of the System Properties window.

Getting at Command-Line Targets

SCons supports a COMMAND_LINE_TARGETS variable that lets you get at the list of targets that the user specified on the command line. You can use the targets to manipulate the build in any way you wish. As a simple example, suppose that you want to print a reminder to the user whenever a specific program is built. You can do this by checking for the target in the COMMAND_LINE_TARGETS list:

      if 'bar' in COMMAND_LINE_TARGETS:
          print "Don't forget to copy `bar' to the archive!"
      Default(Program('foo.c'))
      Program('bar.c')
    

Then, running SCons with the default target works as it always does, but explicity specifying the bar target on the command line generates the warning message:

      % scons -Q
      cc -c -o foo.o foo.c
      cc -o foo foo.o
      % scons -Q bar
      Don't forget to copy `bar' to the archive!
      cc -c -o bar.o bar.c
      cc -o bar bar.o
    

Another practical use for the COMMAND_LINE_TARGETS variable might be to speed up a build by only reading certain subsidiary SConscript files if a specific target is requested.

Controlling the Default Targets

One of the most basic things you can control is which targets SCons will build by default--that is, when there are no targets specified on the command line. As mentioned previously, SCons will normally build every target in or below the current directory by default--that is, when you don't explicitly specify one or more targets on the command line. Sometimes, however, you may want to specify explicitly that only certain programs, or programs in certain directories, should be built by default. You do this with the Default function:

       env = Environment()
       hello = env.Program('hello.c')
       env.Program('goodbye.c')
       Default(hello)
    

This SConstruct file knows how to build two programs, hello and goodbye, but only builds the hello program by default:

       % scons -Q
       cc -c -o hello.o hello.c
       cc -o hello hello.o
       % scons -Q
       scons: `hello' is up to date.
       % scons -Q goodbye
       cc -c -o goodbye.o goodbye.c
       cc -o goodbye goodbye.o
    

Note that, even when you use the Default function in your SConstruct file, you can still explicitly specify the current directory (.) on the command line to tell SCons to build everything in (or below) the current directory:

       % scons -Q .
       cc -c -o goodbye.o goodbye.c
       cc -o goodbye goodbye.o
       cc -c -o hello.o hello.c
       cc -o hello hello.o
    

You can also call the Default function more than once, in which case each call adds to the list of targets to be built by default:

       env = Environment()
       prog1 = env.Program('prog1.c')
       Default(prog1)
       prog2 = env.Program('prog2.c')
       prog3 = env.Program('prog3.c')
       Default(prog3)
    

Or you can specify more than one target in a single call to the Default function:

       env = Environment()
       prog1 = env.Program('prog1.c')
       prog2 = env.Program('prog2.c')
       prog3 = env.Program('prog3.c')
       Default(prog1, prog3)
    

Either of these last two examples will build only the prog1 and prog3 programs by default:

       % scons -Q
       cc -c -o prog1.o prog1.c
       cc -o prog1 prog1.o
       cc -c -o prog3.o prog3.c
       cc -o prog3 prog3.o
       % scons -Q .
       cc -c -o prog2.o prog2.c
       cc -o prog2 prog2.o
    

You can list a directory as an argument to Default:

       env = Environment()
       env.Program(['prog1/main.c', 'prog1/foo.c'])
       env.Program(['prog2/main.c', 'prog2/bar.c'])
       Default('prog1')
    

In which case only the target(s) in that directory will be built by default:

       % scons -Q
       cc -c -o prog1/foo.o prog1/foo.c
       cc -c -o prog1/main.o prog1/main.c
       cc -o prog1/main prog1/main.o prog1/foo.o
       % scons -Q
       scons: `prog1' is up to date.
       % scons -Q .
       cc -c -o prog2/bar.o prog2/bar.c
       cc -c -o prog2/main.o prog2/main.c
       cc -o prog2/main prog2/main.o prog2/bar.o
    

Lastly, if for some reason you don't want any targets built by default, you can use the Python None variable:

       env = Environment()
       prog1 = env.Program('prog1.c')
       prog2 = env.Program('prog2.c')
       Default(None)
    

Which would produce build output like:

       % scons -Q
       scons: *** No targets specified and no Default() targets found.  Stop.
       % scons -Q .
       cc -c -o prog1.o prog1.c
       cc -o prog1 prog1.o
       cc -c -o prog2.o prog2.c
       cc -o prog2 prog2.o
    

         prog1 = Program('prog1.c')
         Default(prog1)
         print "DEFAULT_TARGETS is", map(str, DEFAULT_TARGETS)
      

         prog1 = Program('prog1.c')
         Default(prog1)
         print "DEFAULT_TARGETS is now", map(str, DEFAULT_TARGETS)
         prog2 = Program('prog2.c')
         Default(prog2)
         print "DEFAULT_TARGETS is now", map(str, DEFAULT_TARGETS)
      

Getting at the List of Build Targets, Regardless of Origin

We've already been introduced to the COMMAND_LINE_TARGETS variable, which contains a list of targets specified on the command line, and the DEFAULT_TARGETS variable, which contains a list of targets specified via calls to the Default method or function. Sometimes, however, you want a list of whatever targets SCons will try to build, regardless of whether the targets came from the command line or a Default call. You could code this up by hand, as follows:

      if COMMAND_LINE_TARGETS:
          targets = COMMAND_LINE_TARGETS
      else:
          targets = DEFAULT_TARGETS
    

SCons, however, provides a convenient BUILD_TARGETS variable that eliminates the need for this by-hand manipulation. Essentially, the BUILD_TARGETS variable contains a list of the command-line targets, if any were specified, and if no command-line targets were specified, it contains a list of the targets specified via the Default method or function.

Because BUILD_TARGETS may contain a list of SCons nodes, you must convert the list elements to strings if you want to print them or look for a specific target name, just like the DEFAULT_TARGETS list:

      prog1 = Program('prog1.c')
      Program('prog2.c')
      Default(prog1)
      print "BUILD_TARGETS is", map(str, BUILD_TARGETS)
    

Notice how the value of BUILD_TARGETS changes depending on whether a target is specified on the command line:

      % scons -Q
      BUILD_TARGETS is ['prog1']
      cc -c -o prog1.o prog1.c
      cc -o prog1 prog1.o
      % scons -Q prog2
      BUILD_TARGETS is ['prog2']
      cc -c -o prog2.o prog2.c
      cc -o prog2 prog2.o
      % scons -Q -c .
      BUILD_TARGETS is ['.']
      Removed prog1.o
      Removed prog1
      Removed prog2.o
      Removed prog2
    

Command-Line variable=value Build Options

You may want to control various aspects of your build by allowing the user to specify variable=value values on the command line. For example, suppose you want users to be able to build a debug version of a program by running SCons as follows:

      % scons -Q debug=1
    

SCons provides an ARGUMENTS dictionary that stores all of the variable=value assignments from the command line. This allows you to modify aspects of your build in response to specifications on the command line. (Note that unless you want to require that users always specify an option, you probably want to use the Python ARGUMENTS.get() function, which allows you to specify a default value to be used if there is no specification on the command line.)

The following code sets the CCFLAGS construction variable in response to the debug flag being set in the ARGUMENTS dictionary:

       env = Environment()
       debug = ARGUMENTS.get('debug', 0)
       if int(debug):
           env.Append(CCFLAGS = '-g')
       env.Program('prog.c')
    

This results in the -g compiler option being used when debug=1 is used on the command line:

       % scons -Q debug=0
       cc -c -o prog.o prog.c
       cc -o prog prog.o
       % scons -Q debug=0
       scons: `.' is up to date.
       % scons -Q debug=1
       cc -g -c -o prog.o prog.c
       cc -o prog prog.o
       % scons -Q debug=1
       scons: `.' is up to date.
    

Notice that SCons keeps track of the last values used to build the object files, and as a result correctly rebuilds the object and executable files only when the value of the debug argument has changed.

Controlling Command-Line Build Options

Being able to use a command-line build option like debug=1 is handy, but it can be a chore to write specific Python code to recognize each such option and apply the values to a construction variable. To help with this, SCons supports a class to define such build options easily, and a mechanism to apply the build options to a construction environment. This allows you to control how the build options affect construction environments.

For example, suppose that you want users to set a RELEASE construction variable on the command line whenever the time comes to build a program for release, and that the value of this variable should be added to the command line with the appropriate -D option (or other command line option) to pass the value to the C compiler. Here's how you might do that by setting the appropriate value in a dictionary for the CPPDEFINES construction variable:

         opts = Options()
         opts.Add('RELEASE', 'Set to 1 to build for release', 0)
         env = Environment(options = opts,
                           CPPDEFINES={'RELEASE_BUILD' : '${RELEASE}'})
         env.Program(['foo.c', 'bar.c'])
    

This SConstruct file first creates an Options object (the opts = Options() call), and then uses the object's Add method to indicate that the RELEASE option can be set on the command line, and that it's default value will be 0 (the third argument to the Add method). The second argument is a line of help text; we'll learn how to use it in the next section.

We then pass the created Options object as an options keyword argument to the Environment call used to create the construction environment. This then allows a user to set the RELEASE build option on the command line and have the variable show up in the command line used to build each object from a C source file:

      % scons -Q RELEASE=1
      cc -DRELEASE_BUILD=1 -c -o bar.o bar.c
      cc -DRELEASE_BUILD=1 -c -o foo.o foo.c
      cc -o foo foo.o bar.o
    

Providing Help for Command-Line Build Options

To make command-line build options most useful, you ideally want to provide some help text that will describe the available options when the user runs scons -h. You could write this text by hand, but SCons provides an easier way. Options objects support a GenerateHelpText method that will, as its name indicates, generate text that describes the various options that have been added to it. You then pass the output from this method to the Help function:

         opts = Options('custom.py')
         opts.Add('RELEASE', 'Set to 1 to build for release', 0)
         env = Environment(options = opts)
         Help(opts.GenerateHelpText(env))
    

SCons will now display some useful text when the -h option is used:

      % scons -Q -h
      
      RELEASE: Set to 1 to build for release
          default: 0
          actual: 0
      
      Use scons -H for help about command-line options.
    

Notice that the help output shows the default value, and the current actual value of the build option.

Reading Build Options From a File

Being able to use a command-line build option like debug=1 is handy, but it can be a chore to write specific Python code to recognize each such option and apply the values to a construction variable. To help with this, SCons supports a class to define such build options easily and to read build option values from a file. This allows you to control how the build options affect construction environments. The way you do this is by specifying a file name when you call Options, like custom.py in the following example:

         opts = Options('custom.py')
         opts.Add('RELEASE', 'Set to 1 to build for release', 0)
         env = Environment(options = opts,
                           CPPDEFINES={'RELEASE_BUILD' : '${RELEASE}'})
         env.Program(['foo.c', 'bar.c'])
         Help(opts.GenerateHelpText(env))
    

This then allows us to control the RELEASE variable by setting it in the custom.py file:

      RELEASE = 1
      

Note that this file is actually executed like a Python script. Now when we run SCons:

      % scons -Q
      cc -DRELEASE_BUILD=1 -c -o bar.o bar.c
      cc -DRELEASE_BUILD=1 -c -o foo.o foo.c
      cc -o foo foo.o bar.o
    

And if we change the contents of custom.py to:

      RELEASE = 0
    

The object files are rebuilt appropriately with the new option:

      % scons -Q
      cc -DRELEASE_BUILD=0 -c -o bar.o bar.c
      cc -DRELEASE_BUILD=0 -c -o foo.o foo.c
      cc -o foo foo.o bar.o
    

Canned Build Options

SCons provides a number of functions that provide ready-made behaviors for various types of command-line build options.

           opts = Options('custom.py')
           opts.Add(BoolOption('RELEASE', 0, 'Set to build for release'))
           env = Environment(options = opts,
                             CPPDEFINES={'RELEASE_BUILD' : '${RELEASE}'})
           env.Program('foo.c')
      

           opts = Options('custom.py')
           opts.Add(EnumOption('COLOR', 'red', 'Set background color',
                               allowed_values=('red', 'green', 'blue')))
           env = Environment(options = opts,
                             CPPDEFINES={'COLOR' : '"${COLOR}"'})
           env.Program('foo.c')
      

           opts = Options('custom.py')
           opts.Add(EnumOption('COLOR', 'red', 'Set background color',
                               allowed_values=('red', 'green', 'blue'),
                               map={'navy':'blue'}))
           env = Environment(options = opts,
                             CPPDEFINES={'COLOR' : '"${COLOR}"'})
           env.Program('foo.c')
      

           opts = Options('custom.py')
           opts.Add(EnumOption('COLOR', 'red', 'Set background color',
                               allowed_values=('red', 'green', 'blue'),
                               map={'navy':'blue'},
                               ignorecase=1))
           env = Environment(options = opts,
                             CPPDEFINES={'COLOR' : '"${COLOR}"'})
           env.Program('foo.c')
      

           opts = Options('custom.py')
           opts.Add(EnumOption('COLOR', 'red', 'Set background color',
                               allowed_values=('red', 'green', 'blue'),
                               map={'navy':'blue'},
                               ignorecase=2))
           env = Environment(options = opts,
                             CPPDEFINES={'COLOR' : '"${COLOR}"'})
           env.Program('foo.c')
      

           opts = Options('custom.py')
           opts.Add(ListOption('COLORS', 0, 'List of colors',
                               ['red', 'green', 'blue']))
           env = Environment(options = opts,
                             CPPDEFINES={'COLORS' : '"${COLORS}"'})
           env.Program('foo.c')
      

           opts = Options('custom.py')
           opts.Add(PathOption('CONFIG', '/etc/my_config', 'Path to configuration file'))
           env = Environment(options = opts,
                             CPPDEFINES={'CONFIG_FILE' : '"$CONFIG"'})
           env.Program('foo.c')
      

           opts = Options('custom.py')
           opts.Add(PackageOption('PACKAGE', '/opt/location', 'Location package'))
           env = Environment(options = opts,
                             CPPDEFINES={'PACKAGE' : '"$PACKAGE"'})
           env.Program('foo.c')
      

Adding Multiple Command-Line Build Options at Once

Lastly, SCons provides a way to add multiple build options to an Options object at once. Instead of having to call the Add method multiple times, you can call the AddOptions method with a list of build options to be added to the object. Each build option is specified as either a tuple of arguments, just like you'd pass to the Add method itself, or as a call to one of the canned functions for pre-packaged command-line build options. in any order:

        opts = Options()
        opts.AddOptions(
            ('RELEASE', 'Set to 1 to build for release', 0),
            ('CONFIG', 'Configuration file', '/etc/my_config'),
            BoolOption('warnings', 'compilation with -Wall and similiar', 1),
            EnumOption('debug', 'debug output and symbols', 'no',
                       allowed_values=('yes', 'no', 'full'),
                       map={}, ignorecase=0),  # case sensitive
            ListOption('shared',
                       'libraries to build as shared libraries',
                       'all',
                       names = list_of_libs),
            PackageOption('x11',
                          'use X11 installed here (yes = search some places)',
                          'yes'),
            PathOption('qtdir', 'where the root of Qt is installed', qtdir),
        )
    

Installing Multiple Files in a Directory

You can install multiple files into a directory simply by calling the Install function multiple times:

       env = Environment()
       hello = env.Program('hello.c')
       goodbye = env.Program('goodbye.c')
       env.Install('/usr/bin', hello)
       env.Install('/usr/bin', goodbye)
       env.Alias('install', '/usr/bin')
    

Or, more succinctly, listing the multiple input files in a list (just like you can do with any other builder):

       env = Environment()
       hello = env.Program('hello.c')
       goodbye = env.Program('goodbye.c')
       env.Install('/usr/bin', [hello, goodbye])
       env.Alias('install', '/usr/bin')
    

Either of these two examples yields:

       % scons -Q install
       cc -c -o goodbye.o goodbye.c
       cc -o goodbye goodbye.o
       Install file: "goodbye" as "/usr/bin/goodbye"
       cc -c -o hello.o hello.c
       cc -o hello hello.o
       Install file: "hello" as "/usr/bin/hello"
    

Installing a File Under a Different Name

The Install method preserves the name of the file when it is copied into the destination directory. If you need to change the name of the file when you copy it, use the InstallAs function:

       env = Environment()
       hello = env.Program('hello.c')
       env.InstallAs('/usr/bin/hello-new', hello)
       env.Alias('install', '/usr/bin')
    

This installs the hello program with the name hello-new as follows:

       % scons -Q install
       cc -c -o hello.o hello.c
       cc -o hello hello.o
       Install file: "hello" as "/usr/bin/hello-new"
    

Installing Multiple Files Under Different Names

Lastly, if you have multiple files that all need to be installed with different file names, you can either call the InstallAs function multiple times, or as a shorthand, you can supply same-length lists for the both the target and source arguments:

       env = Environment()
       hello = env.Program('hello.c')
       goodbye = env.Program('goodbye.c')
       env.InstallAs(['/usr/bin/hello-new',
                      '/usr/bin/goodbye-new'],
                     [hello, goodbye])
       env.Alias('install', '/usr/bin')
    

In this case, the InstallAs function loops through both lists simultaneously, and copies each source file into its corresponding target file name:

       % scons -Q install
       cc -c -o goodbye.o goodbye.c
       cc -o goodbye goodbye.o
       Install file: "goodbye" as "/usr/bin/goodbye-new"
       cc -c -o hello.o hello.c
       cc -o hello hello.o
       Install file: "hello" as "/usr/bin/hello-new"
    

SConscript Files

As we've already seen, the build script at the top of the tree is called SConstruct. The top-level SConstruct file can use the SConscript function to include other subsidiary scripts in the build. These subsidiary scripts can, in turn, use the SConscript function to include still other scripts in the build. By convention, these subsidiary scripts are usually named SConscript. For example, a top-level SConstruct file might arrange for four subsidiary scripts to be included in the build as follows:

      SConscript(['drivers/display/SConscript',
                  'drivers/mouse/SConscript',
                  'parser/SConscript',
                  'utilities/SConscript'])
    

In this case, the SConstruct file lists all of the SConscript files in the build explicitly. (Note, however, that not every directory in the tree necessarily has an SConscript file.) Alternatively, the drivers subdirectory might contain an intermediate SConscript file, in which case the SConscript call in the top-level SConstruct file would look like:

      SConscript(['drivers/SConscript',
                  'parser/SConscript',
                  'utilities/SConscript'])
    

And the subsidiary SConscript file in the drivers subdirectory would look like:

      SConscript(['display/SConscript',
                  'mouse/SConscript'])
    

Whether you list all of the SConscript files in the top-level SConstruct file, or place a subsidiary SConscript file in intervening directories, or use some mix of the two schemes, is up to you and the needs of your software.

Path Names Are Relative to the SConscript Directory

Subsidiary SConscript files make it easy to create a build hierarchy because all of the file and directory names in a subsidiary SConscript files are interpreted relative to the directory in which the SConscript file lives. Typically, this allows the SConscript file containing the instructions to build a target file to live in the same directory as the source files from which the target will be built, making it easy to update how the software is built whenever files are added or deleted (or other changes are made).

For example, suppose we want to build two programs prog1 and prog2 in two separate directories with the same names as the programs. One typical way to do this would be with a top-level SConstruct file like this:

      SConscript(['prog1/SConscript',
                  'prog2/SConscript'])
    

And subsidiary SConscript files that look like this:

      env = Environment()
      env.Program('prog1', ['main.c', 'foo1.c', 'foo2.c'])
      

And this:

      env = Environment()
      env.Program('prog2', ['main.c', 'bar1.c', 'bar2.c'])
      

Then, when we run SCons in the top-level directory, our build looks like:

       % scons -Q
       cc -c -o prog1/foo1.o prog1/foo1.c
       cc -c -o prog1/foo2.o prog1/foo2.c
       cc -c -o prog1/main.o prog1/main.c
       cc -o prog1/prog1 prog1/main.o prog1/foo1.o prog1/foo2.o
       cc -c -o prog2/bar1.o prog2/bar1.c
       cc -c -o prog2/bar2.o prog2/bar2.c
       cc -c -o prog2/main.o prog2/main.c
       cc -o prog2/prog2 prog2/main.o prog2/bar1.o prog2/bar2.o
    

Notice the following: First, you can have files with the same names in multiple directories, like main.c in the above example. Second, unlike standard recursive use of Make, SCons stays in the top-level directory (where the SConstruct file lives) and issues commands that use the path names from the top-level directory to the target and source files within the hierarchy.

Top-Level Path Names in Subsidiary SConscript Files

If you need to use a file from another directory, it's sometimes more convenient to specify the path to a file in another directory from the top-level SConstruct directory, even when you're using that file in a subsidiary SConscript file in a subdirectory. You can tell SCons to interpret a path name as relative to the top-level SConstruct directory, not the local directory of the SConscript file, by appending a # (hash mark) to the beginning of the path name:

       env = Environment()
       env.Program('prog', ['main.c', '#lib/foo1.c', 'foo2.c'])
    

In this example, the lib directory is directly underneath the top-level SConstruct directory. If the above SConscript file is in a subdirectory named src/prog, the output would look like:

       % scons -Q
       cc -c -o lib/foo1.o lib/foo1.c
       cc -c -o src/prog/foo2.o src/prog/foo2.c
       cc -c -o src/prog/main.o src/prog/main.c
       cc -o src/prog/prog src/prog/main.o lib/foo1.o src/prog/foo2.o
    

(Notice that the lib/foo1.o object file is built in the same directory as its source file. See section XXX, below, for information about how to build the object file in a different subdirectory.)

Absolute Path Names

Of course, you can always specify an absolute path name for a file--for example:

       env = Environment()
       env.Program('prog', ['main.c', '/usr/joe/lib/foo1.c', 'foo2.c'])
    

Which, when executed, would yield:

       % scons -Q
       cc -c -o src/prog/foo2.o src/prog/foo2.c
       cc -c -o src/prog/main.o src/prog/main.c
       cc -c -o /usr/joe/lib/foo1.o /usr/joe/lib/foo1.c
       cc -o src/prog/prog src/prog/main.o /usr/joe/lib/foo1.o src/prog/foo2.o
    

(As was the case with top-relative path names, notice that the /usr/joe/lib/foo1.o object file is built in the same directory as its source file. See section XXX, below, for information about how to build the object file in a different subdirectory.)

Sharing Environments (and Other Variables) Between SConscript Files

In the previous example, each of the subsidiary SConscript files created its own construction environment by calling Environment separately. This obviously works fine, but if each program must be built with the same construction variables, it's cumbersome and error-prone to initialize separate construction environments in the same way over and over in each subsidiary SConscript file.

SCons supports the ability to export variables from a parent SConscript file to its subsidiary SConscript files, which allows you to share common initialized values throughout your build hierarchy.

        env = Environment()
        Export('env')
      

        env = Environment()
        debug = ARGUMENTS['debug']
        Export('env', 'debug')
      

        Export('env debug')
      

        SConscript('src/SConscript', 'env')
      

        SConscript('src/SConscript', exports='env')
      

        SConscript(['src1/SConscript',
                    'src2/SConscript'], exports='env')
      

        Import('env')
        env.Program('prog', ['prog.c'])
      

        Import('env', 'debug')
        env = env.Copy(DEBUG = debug)
        env.Program('prog', ['prog.c'])
      

        Import('env debug')
        env = env.Copy(DEBUG = debug)
        env.Program('prog', ['prog.c'])
      

        Import('*')
        env = env.Copy(DEBUG = debug)
        env.Program('prog', ['prog.c'])
      

          env = Environment()
          Export('env')
          objs = []
          for subdir in ['foo', 'bar']:
              o = SConscript('%s/SConscript' % subdir)
              objs.append(o)
          env.Library('prog', objs)
      

          Import('env')
          obj = env.Object('foo.c')
          Return('obj')
        

Specifying a Build Directory as Part of an SConscript Call

The most straightforward way to establish a build directory uses the fact that the usual way to set up a build hierarchy is to have an SConscript file in the source subdirectory. If you then pass a build_dir argument to the SConscript function call:

      SConscript('src/SConscript', build_dir='build')
    

SCons will then build all of the files in the build subdirectory:

      % ls src
      SConscript  hello.c
      % scons -Q
      cc -c -o build/hello.o build/hello.c
      cc -o build/hello build/hello.o
      % ls build
      SConscript  hello  hello.c  hello.o
    

But wait a minute--what's going on here? SCons created the object file build/hello.o in the build subdirectory, as expected. But even though our hello.c file lives in the src subdirectory, SCons has actually compiled a build/hello.c file to create the object file.

What's happened is that SCons has duplicated the hello.c file from the src subdirectory to the build subdirectory, and built the program from there. The next section explains why SCons does this.

Why SCons Duplicates Source Files in a Build Directory

SCons duplicates source files in build directories because it's the most straightforward way to guarantee a correct build regardless of include-file directory paths, relative references between files, or tool support for putting files in different locations, and the SCons philosophy is to, by default, guarantee a correct build in all cases.

The most direct reason to duplicate source files in build directories is simply that some tools (mostly older vesions) are written to only build their output files in the same directory as the source files. In this case, the choices are either to build the output file in the source directory and move it to the build directory, or to duplicate the source files in the build directory.

Additionally, relative references between files can cause problems if we don't just duplicate the hierarchy of source files in the build directory. You can see this at work in use of the C preprocessor #include mechanism with double quotes, not angle brackets:

      #include "file.h"
    

The de facto standard behavior for most C compilers in this case is to first look in the same directory as the source file that contains the #include line, then to look in the directories in the preprocessor search path. Add to this that the SCons implementation of support for code repositories (described below) means not all of the files will be found in the same directory hierarchy, and the simplest way to make sure that the right include file is found is to duplicate the source files into the build directory, which provides a correct build regardless of the original location(s) of the source files.

Although source-file duplication guarantees a correct build even in these end-cases, it can usually be safely disabled. The next section describes how you can disable the duplication of source files in the build directory.

Telling SCons to Not Duplicate Source Files in the Build Directory

In most cases and with most tool sets, SCons can place its target files in a build subdirectory without duplicating the source files and everything will work just fine. You can disable the default SCons behavior by specifying duplicate=0 when you call the SConscript function:

      SConscript('src/SConscript', build_dir='build', duplicate=0)
    

When this flag is specified, SCons uses the build directory like most people expect--that is, the output files are placed in the build directory while the source files stay in the source directory:

      % ls src
      SConscript
      hello.c
      % scons -Q
      cc -c src/hello.c -o build/hello.o
      cc -o build/hello build/hello.o
      % ls build
      hello
      hello.o
    

The BuildDir Function

Use the BuildDir function to establish that target files should be built in a separate directory from the source files:

      BuildDir('build', 'src')
      env = Environment()
      env.Program('build/hello.c')
    

Note that when you're not using an SConscript file in the src subdirectory, you must actually specify that the program must be built from the build/hello.c file that SCons will duplicate in the build subdirectory.

When using the BuildDir function directly, SCons still duplicates the source files in the build directory by default:

      % ls src
      hello.c
      % scons -Q
      cc -c -o build/hello.o build/hello.c
      cc -o build/hello build/hello.o
      % ls build
      hello  hello.c  hello.o
    

You can specify the same duplicate=0 argument that you can specify for an SConscript call:

      BuildDir('build', 'src', duplicate=0)
      env = Environment()
      env.Program('build/hello.c')
    

In which case SCons will disable duplication of the source files:

      % ls src
      hello.c
      % scons -Q
      cc -c -o build/hello.o src/hello.c
      cc -o build/hello build/hello.o
      % ls build
      hello  hello.o
    

Using BuildDir With an SConscript File

Even when using the BuildDir function, it's much more natural to use it with a subsidiary SConscript file. For example, if the src/SConscript looks like this:

      env = Environment()
      env.Program('hello.c')
    

Then our SConstruct file could look like:

      BuildDir('build', 'src')
      SConscript('build/SConscript')
      

Yielding the following output:

      % ls src
      SConscript  hello.c
      % scons -Q
      cc -c -o build/hello.o build/hello.c
      cc -o build/hello build/hello.o
      % ls build
      SConscript  hello  hello.c  hello.o
    

Notice that this is completely equivalent to the use of SConscript that we learned about in the previous section.

Writing Builders That Execute External Commands

The simplest Builder to create is one that executes an external command. For example, if we want to build an output file by running the contents of the input file through a command named foobuild, creating that Builder might look like:

       bld = Builder(action = 'foobuild < $SOURCE > $TARGET')
    

All the above line does is create a free-standing Builder object. The next section will show us how to actually use it.

Attaching a Builder to a Construction Environment

A Builder object isn't useful until it's attached to a construction environment so that we can call it to arrange for files to be built. This is done through the BUILDERS construction variable in an environment. The BUILDERS variable is a Python dictionary that maps the names by which you want to call various Builder objects to the objects themselves. For example, if we want to call the Builder we just defined by the name Foo, our SConstruct file might look like:

       bld = Builder(action = 'foobuild < $SOURCE > $TARGET')
       env = Environment(BUILDERS = {'Foo' : bld})
    

With the Builder so attached to our construction environment we can now actually call it like so:

       env.Foo('file.foo', 'file.input')
    

Then when we run SCons it looks like:

      % scons -Q
      foobuild < file.input > file.foo
    

Note, however, that the default BUILDERS variable in a construction environment comes with a default set of Builder objects already defined: Program, Library, etc. And when we explicitly set the BUILDERS variable when we create the construction environment, the default Builders are no longer part of the environment:

       bld = Builder(action = 'foobuild < $SOURCE > $TARGET')
       env = Environment(BUILDERS = {'Foo' : bld})
       env.Foo('file.foo', 'file.input')
       env.Program('hello.c')
    

      % scons -Q
      AttributeError: SConsEnvironment instance has no attribute 'Program':
    

To be able use both our own defined Builder objects and the default Builder objects in the same construction environment, you can either add to the BUILDERS variable using the Append function:

       env = Environment()
       bld = Builder(action = 'foobuild < $SOURCE > $TARGET')
       env.Append(BUILDERS = {'Foo' : bld})
       env.Foo('file.foo', 'file.input')
       env.Program('hello.c')
    

Or you can explicitly set the appropriately-named key in the BUILDERS dictionary:

       env = Environment()
       bld = Builder(action = 'foobuild < $SOURCE > $TARGET')
       env['BUILDERS']['Foo'] = bld
       env.Foo('file.foo', 'file.input')
       env.Program('hello.c')
    

Either way, the same construction environment can then use both the newly-defined Foo Builder and the default Program Builder:

      % scons -Q
      foobuild < file.input > file.foo
      cc -c -o hello.o hello.c
      cc -o hello hello.o
    

Letting SCons Handle The File Suffixes

By supplying additional information when you create a Builder, you can let SCons add appropriate file suffixes to the target and/or the source file. For example, rather than having to specify explicitly that you want the Foo Builder to build the file.foo target file from the file.input source file, you can give the .foo and .input suffixes to the Builder, making for more compact and readable calls to the Foo Builder:

       bld = Builder(action = 'foobuild < $SOURCE > $TARGET',
                     suffix = '.foo',
                     src_suffix = '.input')
       env = Environment(BUILDERS = {'Foo' : bld})
       env.Foo('file1')
       env.Foo('file2')
    

      % scons -Q
      foobuild < file1.input > file1.foo
      foobuild < file2.input > file2.foo
    

You can also supply a prefix keyword argument if it's appropriate to have SCons append a prefix to the beginning of target file names.

Builders That Execute Python Functions

In SCons, you don't have to call an external command to build a file. You can, instead, define a Python function that a Builder object can invoke to build your target file (or files). Such a builder function definition looks like:

       def build_function(target, source, env):
           # Code to build "target" from "source"
           return None
    

The arguments of a builder function are:

The builder function must return a 0 or None value if the target(s) are built successfully. The builder function may raise an exception or return any non-zero value to indicate that the build is unsuccessful,

Once you've defined the Python function that will build your target file, defining a Builder object for it is as simple as specifying the name of the function, instead of an external command, as the Builder's action argument:

       def build_function(target, source, env):
           # Code to build "target" from "source"
           return None
       bld = Builder(action = build_function,
                     suffix = '.foo',
                     src_suffix = '.input')
       env = Environment(BUILDERS = {'Foo' : bld})
       env.Foo('file')
    

And notice that the output changes slightly, reflecting the fact that a Python function, not an external command, is now called to build the target file:

      % scons -Q
      build_function("file.foo", "file.input")
    

Builders That Create Actions Using a Generator

SCons Builder objects can create an action "on the fly" by using a function called a generator. This provides a great deal of flexibility to construct just the right list of commands to build your target. A generator looks like:

       def generate_actions(source, target, env, for_signature):
           return 'foobuild < %s > %s' % (target[0], source[0])
    

The arguments of a generator are:

The generator must return a command string or other action that will be used to build the specified target(s) from the specified source(s).

Once you've defined a generator, you create a Builder to use it by specifying the generator keyword argument instead of action.

       def generate_actions(source, target, env, for_signature):
           return 'foobuild < %s > %s' % (source[0], target[0])
       bld = Builder(generator = generate_actions,
                     suffix = '.foo',
                     src_suffix = '.input')
       env = Environment(BUILDERS = {'Foo' : bld})
       env.Foo('file')
    

      % scons -Q
      foobuild < file.input > file.foo
    

Note that it's illegal to specify both an action and a generator for a Builder.

Builders That Modify the Target or Source Lists Using an Emitter

SCons supports the ability for a Builder to modify the lists of target(s) from the specified source(s).

       def modify_targets(target, source, env):
           target.append('new_target')
           source.append('new_source')
           return target, source
       bld = Builder(action = 'foobuild $TARGETS - $SOURCES',
                     suffix = '.foo',
                     src_suffix = '.input',
                     emitter = modify_targets)
       env = Environment(BUILDERS = {'Foo' : bld})
       env.Foo('file')
    

      % scons -Q
      foobuild file.foo new_target - file.input new_source
    

       bld = Builder(action = 'XXX',
                     suffix = '.foo',
                     src_suffix = '.input',
                     emitter = 'MY_EMITTER')
       def modify1(target, source, env):
           return target, source
       def modify2(target, source, env):
           return target, source
       env1 = Environment(BUILDERS = {'Foo' : bld},
                          MY_EMITTER = modify1)
       env2 = Environment(BUILDERS = {'Foo' : bld},
                          MY_EMITTER = modify2)
       env1.Foo('file1')
       env2.Foo('file2')
    

A Simple Scanner Example

Suppose, for example, that we want to create a simple scanner for .foo files. A .foo file contains some text that will be processed, and can include other files on lines that begin with include followed by a file name:

      include filename.foo
    

Scanning a file will be handled by a Python function that you must supply. Here is a function that will use the Python re module to scan for the include lines in our example:

      import re
      
      include_re = re.compile(r'^include\\s+(\\S+)$', re.M)
      
      def kfile_scan(node, env, path, arg):
          contents = node.get_contents()
          return include_re.findall(contents)
    

The scanner function must accept the four specified arguments and return a list of implicit dependencies. Presumably, these would be dependencies found from examining the contents of the file, although the function can perform any manipulation at all to generate the list of dependencies.

A Scanner object is created using the Scanner function, which typically takes an skeys argument to associate the type of file suffix with this scanner. The Scanner object must then be associated with the SCANNERS construction variable of a construction environment, typically by using the Append method:

       kscan = Scanner(function = kfile_scan,
                       skeys = ['.k'])
       env.Append(SCANNERS = kscan)
    

When we put it all together, it looks like:

        import re

        include_re = re.compile(r'^include\\s+(\\S+)$', re.M)

        def kfile_scan(node, env, path):
            contents = node.get_contents()
            includes = include_re.findall(contents)
            return includes

        kscan = Scanner(function = kfile_scan,
                        skeys = ['.k'])

        env = Environment(ENV = {'PATH' : '/usr/local/bin'})
        env.Append(SCANNERS = kscan)

        env.Command('foo', 'foo.k', 'kprocess < $SOURCES > $TARGET')
    

The Repository Method

It's often useful to allow multiple programmers working on a project to build software from source files and/or derived files that are stored in a centrally-accessible repository, a directory copy of the source code tree. (Note that this is not the sort of repository maintained by a source code management system like BitKeeper, CVS, or Subversion. For information about using SCons with these systems, see the section, "Fetching Files From Source Code Management Systems," below.) You use the Repository method to tell SCons to search one or more central code repositories (in order) for any source files and derived files that are not present in the local build tree:

       env = Environment()
       env.Program('hello.c')
       Repository('/usr/repository1', '/usr/repository2')
    

Multiple calls to the Repository method will simply add repositories to the global list that SCons maintains, with the exception that SCons will automatically eliminate the current directory and any non-existent directories from the list.

Finding source files in repositories

The above example specifies that SCons will first search for files under the /usr/repository1 tree and next under the /usr/repository2 tree. SCons expects that any files it searches for will be found in the same position relative to the top-level directory. In the above example, if the hello.c file is not found in the local build tree, SCons will search first for a /usr/repository1/hello.c file and then for a /usr/repository1/hello.c file to use in its place.

So given the SConstruct file above, if the hello.c file exists in the local build directory, SCons will rebuild the hello program as normal:

      % scons -Q
      cc -c -o hello.o hello.c
      cc -o hello hello.o
    

If, however, there is no local hello.c file, but one exists in /usr/repository1, SCons will recompile the hello program from the source file it finds in the repository:

      % scons -Q
      cc -c -o hello.o hello.c
      cc -o hello hello.o
      gcc -c /usr/repository1/hello.c -o hello.o
      gcc -o hello hello.o
    

And similarly, if there is no local hello.c file and no /usr/repository1/hello.c, but one exists in /usr/repository2:

      % scons -Q
      cc -c -o hello.o hello.c
      cc -o hello hello.o
    

Finding the SConstruct file in repositories

SCons will also search in repositories for the SConstruct file and any specified SConscript files. This poses a problem, though: how can SCons search a repository tree for an SConstruct file if the SConstruct file itself contains the information about the pathname of the repository? To solve this problem, SCons allows you to specify repository directories on the command line using the -Y option:

      % scons -Q -Y /usr/repository1 -Y /usr/repository2
    

When looking for source or derived files, SCons will first search the repositories specified on the command line, and then search the repositories specified in the SConstruct or SConscript files.

Finding derived files in repositories

If a repository contains not only source files, but also derived files (such as object files, libraries, or executables), SCons will perform its normal MD5 signature calculation to decide if a derived file in a repository is up-to-date, or the derived file must be rebuilt in the local build directory. For the SCons signature calculation to work correctly, a repository tree must contain the .sconsign files that SCons uses to keep track of signature information.

Usually, this would be done by a build integrator who would run SCons in the repository to create all of its derived files and .sconsign files, or who would SCons in a separate build directory and copying the resulting tree to the desired repository:

      % cd /usr/repository1
      % scons -Q
      cc -c -o file1.o file1.c
      cc -c -o file2.o file2.c
      cc -c -o hello.o hello.c
      cc -o hello hello.o file1.o file2.o
    

(Note that this is safe even if the SConstruct file lists /usr/repository1 as a repository, because SCons will remove the current build directory from its repository list for that invocation.)

Now, with the repository populated, we only need to create the one local source file we're interested in working with at the moment, and use the -Y option to tell SCons to fetch any other files it needs from the repository:

      % cd $HOME/build
      % edit hello.c
      % scons -Q -Y /usr/repository1
      cc -c -o hello.o hello.c
      cc -o hello hello.o /usr/repository1/file1.o /usr/repository1/file2.o
    

Notice that SCons realizes that it does not need to rebuild local copies file1.o and file2.o files, but instead uses the already-compiled files from the repository.

Guaranteeing local copies of files

If the repository tree contains the complete results of a build, and we try to build from the repository without any files in our local tree, something moderately surprising happens:

      % mkdir $HOME/build2
      % cd $HOME/build2
      % scons -Q -Y /usr/all/repository hello
      scons: `hello' is up-to-date.
    

Why does SCons say that the hello program is up-to-date when there is no hello program in the local build directory? Because the repository (not the local directory) contains the up-to-date hello program, and SCons correctly determines that nothing needs to be done to rebuild that up-to-date copy of the file.

There are, however, many times when you want to ensure that a local copy of a file always exists. A packaging or testing script, for example, may assume that certain generated files exist locally. To tell SCons to make a copy of any up-to-date repository file in the local build directory, use the Local function:

       env = Environment()
       hello = env.Program('hello.c')
       Local(hello)
    

If we then run the same command, SCons will make a local copy of the program from the repository copy, and tell you that it is doing so:

      % scons -Y /usr/all/repository hello
      Local copy of hello from /usr/all/repository/hello
      scons: `hello' is up-to-date.
    

(Notice that, because the act of making the local copy is not considered a "build" of the hello file, SCons still reports that it is up-to-date.)

Specifying the Shared Cache Directory

To enable sharing of derived files, use the CacheDir function in any SConscript file:

       CacheDir('/usr/local/build_cache')
    

Note that the directory you specify must already exist and be readable and writable by all developers who will be sharing derived files. It should also be in some central location that all builds will be able to access. In environments where developers are using separate systems (like individual workstations) for builds, this directory would typically be on a shared or NFS-mounted file system.

Here's what happens: When a build has a CacheDir specified, every time a file is built, it is stored in the shared cache directory along with its MD5 build signature. On subsequent builds, before an action is invoked to build a file, SCons will check the shared cache directory to see if a file with the exact same build signature already exists. If so, the derived file will not be built locally, but will be copied into the local build directory from the shared cache directory, like so:

      % scons -Q
      cc -c -o hello.o hello.c
      cc -o hello hello.o
      % scons -Q -c
      Removed hello.o
      Removed hello
      % scons -Q
      Retrieved `hello.o' from cache
      Retrieved `hello' from cache
    

Keeping Build Output Consistent

One potential drawback to using a shared cache is that your build output can be inconsistent from invocation to invocation, because any given file may be rebuilt one time and retrieved from the shared cache the next time. This can make analyzing build output more difficult, especially for automated scripts that expect consistent output each time.

If, however, you use the --cache-show option, SCons will print the command line that it would have executed to build the file, even when it is retrieving the file from the shared cache. This makes the build output consistent every time the build is run:

      % scons -Q
      cc -c -o hello.o hello.c
      cc -o hello hello.o
      % scons -Q -c
      Removed hello.o
      Removed hello
      % scons -Q --cache-show
      cc -c -o hello.o hello.c
      cc -o hello hello.o
    

The trade-off, of course, is that you no longer know whether or not SCons has retrieved a derived file from cache or has rebuilt it locally.

Not Retrieving Files From a Shared Cache

Retrieving an already-built file from the shared cache is usually a significant time-savings over rebuilding the file, but how much of a savings (or even whether it saves time at all) can depend a great deal on your system or network configuration. For example, retrieving cached files from a busy server over a busy network might end up being slower than rebuilding the files locally.

In these cases, you can specify the --cache-disable command-line option to tell SCons to not retrieve already-built files from the shared cache directory:

      % scons -Q
      cc -c -o hello.o hello.c
      cc -o hello hello.o
      % scons -Q -c
      Removed hello.o
      Removed hello
      % scons -Q
      Retrieved `hello.o' from cache
      Retrieved `hello' from cache
      % scons -Q -c
      Removed hello.o
      Removed hello
      % scons -Q --cache-disable
      cc -c -o hello.o hello.c
      cc -o hello hello.o
    

Populating a Shared Cache With Already-Built Files

Sometimes, you may have one or more derived files already built in your local build tree that you wish to make available to other people doing builds. For example, you may find it more effective to perform integration builds with the cache disabled (per the previous section) and only populate the shared cache directory with the built files after the integration build has completed successfully. This way, the cache will only get filled up with derived files that are part of a complete, successful build not with files that might be later overwritten while you debug integration problems.

In this case, you can use the the --cache-force option to tell SCons to put all derived files in the cache, even if the files had already been built by a previous invocation:

      % scons -Q --cache-disable
      cc -c -o hello.o hello.c
      cc -o hello hello.o
      % scons -Q -c
      Removed hello.o
      Removed hello
      % scons -Q --cache-disable
      cc -c -o hello.o hello.c
      cc -o hello hello.o
      % scons -Q --cache-force
      scons: `.' is up to date.
      % scons -Q -c
      Removed hello.o
      Removed hello
      % scons -Q
      Retrieved `hello.o' from cache
      Retrieved `hello' from cache
    

Notice how the above sample run demonstrates that the --cache-disable option avoids putting the built hello.o and hello files in the cache, but after using the --cache-force option, the files have been put in the cache for the next invocation to retrieve.