Installing Python

Because SCons is written in Python, you must obviously have Python installed on your system to use SCons Before you try to install Python, you should check to see if Python is already available on your system by typing python at your system's command-line prompt. You should see something like the following on a UNIX or Linux system that has Python installed:

       $ python
       Python 2.2.2 (#1, Feb 24 2003, 19:13:11)
       [GCC 3.2.2 20030222 (Red Hat Linux 3.2.2-4)] on linux2
       Type "help", "copyright", "credits" or "license" for more information.
       >>> ^D
    

And on a Windows system with Python installed:

       C:\>python
       Python 2.2.2 (#34, Apr 9 2002, 19:34:33) [MSC 32 bit (Intel)] on win32
       Type "help", "copyright", "credits" or "license" for more information.
       >>> ^Z
    

The >>> is the input prompt for the Python interpreter. The ^D and ^Z represent the CTRL-D and CTRL-Z characters that you will need to type to get out of the interpreter before proceeding to installing SCons.

If Python is not installed on your system, you will see an error message stating something like "command not found" (on UNIX or Linux) or "'python' is not recognized as an internal or external command, operable progam or batch file" (on Windows). In that case, you need to install Python before you can install SCons.

The standard location for information about downloading and installing Python is http://www.python.org/download/. See that page for information about how to download and install Python on your system.

Installing SCons From Pre-Built Packages

SCons comes pre-packaged for installation on a number of systems, including Linux and Windows systems. You do not need to read this entire section, you should only need to read the section appropriate to the type of system you're running on.

        # rpm -Uvh scons-0.96-1.noarch.rpm
      

        # apt-get install scons
      

Building and Installing SCons on Any System

If a pre-built SCons package is not available for your system, then you can still easily build and install SCons using the native Python distutils package.

The first step is to download either the scons-0.96.tar.gz or scons-0.96.zip, which are available from the SCons download page at http://www.scons.org/download.html.

Unpack the archive you downloaded, using a utility like tar on Linux or UNIX, or WinZip on Windows. This will create a directory called scons-0.96, usually in your local directory. Then change your working directory to that directory and install SCons by executing the following commands:

      # cd scons-0.96
      # python setup.py install
    

This will build SCons, install the scons script in the default system scripts directory (/usr/local/bin or C:\Python2.2\Scripts), and will install the SCons build engine in an appropriate stand-alone library directory (/usr/local/lib/scons or C:\Python2.2\scons). Because these are system directories, you may need root (on Linux or UNIX) or Administrator (on Windows) privileges to install SCons like this.

        # python setup.py install --prefix=$HOME
      

        # python setup.py install --version-lib
      

Building Simple C / C++ Programs

Here's the famous "Hello, World!" program in C:

      int
      main()
      {
          printf("Hello, world!\n");
      }
   

And here's how to build it using SCons. Enter the following into a file named SConstruct:

      Program('hello.c')
   

This minimal configuration file gives SCons two pieces of information: what you want to build (an executable program), and the input file from which you want it built (the hello.c file). Program is a builder method, a Python call that tells SCons that you want to build an executable program.

That's it. Now run the scons command to build the program. On a POSIX-compliant system like Linux or UNIX, you'll see something 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.
   

On a Windows system with the Microsoft Visual C++ compiler, you'll see something 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.
   

First, notice that you only need to specify the name of the source file, and that SCons correctly deduces the names of the object and executable files to be built from the base of the source file name.

Second, notice that the same input SConstruct file, without any changes, generates the correct output file names on both systems: hello.o and hello on POSIX systems, hello.obj and hello.exe on Windows systems. This is a simple example of how SCons makes it extremely easy to write portable software builds.

(Note that we won't provide duplicate side-by-side POSIX and Windows output for all of the examples in this guide; just keep in mind that, unless otherwise specified, any of the examples should work equally well on both types of systems.)

Building Object Files

The Program builder method is only one of many builder methods that SCons provides to build different types of files. Another is the Object builder method, which tells SCons to build an object file from the specified source file:

      Object('hello.c')
   

Now when you run the scons command to build the program, it will build just the hello.o object file on a POSIX system:

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

And just the hello.obj object file on a Windows system (with the Microsoft Visual C++ compiler):

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

Simple Java Builds

SCons also makes building with Java extremely easy. Unlike the Program and Object builder methods, however, the Java builder method requires that you specify the name of a destination directory in which you want the class files placed, followed by the source directory in which the .java files live:

     Java('classes', 'src')
   

If the src directory contains a single hello.java file, then the output from running the scons command would look something like this (on a POSIX system):

      % scons
      scons: Reading SConscript files ...
      scons: done reading SConscript files.
      scons: Building targets ...
      javac -d classes -sourcepath src src/hello.java
      scons: done building targets.
   

We'll cover Java builds in more detail, including building Java archive (.jar) and other types of file, in the chapter called Java Builds.

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.

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

       print "Calling Program('hello.c')"
       Program('hello.c')
       print "Calling Program('goodbye.c')"
       Program('goodbye.c')
       print "Finished calling Program()"
     

       % scons
       scons: Reading SConscript files ...
       Calling Program('hello.c')
       Calling Program('goodbye.c')
       Finished calling Program()
       scons: done reading SConscript files.
       scons: Building targets ...
       cc -c -o goodbye.o goodbye.c
       cc -o goodbye goodbye.o
       cc -c -o hello.o hello.c
       cc -o hello hello.o
       scons: done building targets.
     

Making the SCons 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: all of 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.

One 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.

Specifying the Name of the Target (Output) File

You've seen that when you call the Program builder method, it builds the resulting program with the same base name as the source file. That is, the following call to build an executable program from the hello.c source file will build an executable program named hello on POSIX systems, and an executable program named hello.exe on Windows systems:

       Program('hello.c')
    

If you want to build a program with a different name than the base of the source file name, you simply put the target file name to the left of the source file name:

       Program('new_hello', 'hello.c')
    

(SCons requires the target file name first, followed by the source file name, so that the order mimics that of an assignment statement in most programming languages, including Python: "program = source files".)

Now SCons will build an executable program named new_hello when run on a POSIX system:

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

And SCons will build an executable program named new_hello.exe when run on a Windows system:

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

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(['main.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 main.o main.c
       cc -o main main.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 (as we've seen in the previous section) 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
    

Specifying Single Files Vs. Lists of Files

We've now shown you two ways to specify the source for a program, one with a list of files:

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

And one with a single file:

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

You could actually put a single file name in a list, too, which you might prefer just for the sake of consistency:

       Program('hello', ['hello.c'])
    

SCons functions will accept a single file name in either form. In fact, internally, SCons treats all input as lists of files, but allows you to omit the square brackets to cut down a little on the typing when there's only a single file name.

Although SCons functions are forgiving about whether or not you use a string vs. a list for a single file name, Python itself is more strict about treating lists and strings differently. So where SCons allows either a string or list: # The following two calls both work correctly: Program('program1', 'program1.c') Program('program2', ['program2.c']) Trying to do "Python things" that mix strings and lists will cause errors or lead to incorrect results: common_sources = ['file1.c', 'file2.c'] # THE FOLLOWING IS INCORRECT AND GENERATES A PYTHON ERROR # BECAUSE IT TRIES TO ADD A STRING TO A LIST: Program('program1', common_sources + 'program1.c') # The following works correctly, because it's adding two # lists together to make another list. Program('program2', common_sources + ['program2.c'])

Making Lists of Files Easier 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)
    

(Note in this example that we used the Python "triple-quote" syntax, which allows a string to contain multiple lines. The three quotes can be either single or double quotes.)

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 the chapter called Building and Linking with Libraries, 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'])
      

        % scons -Q
        cc -c -o f1.os f1.c
        cc -c -o f2.os f2.c
        cc -c -o f3.os f3.c
        cc -shared -o libfoo.so f1.os f2.os f3.os
      

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

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', 'bar'], 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 -lbar
    

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 bar.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.

Note also that, if you only have a single library to link with, you can specify the library name in single string, instead of a Python list, so that:

      Program('prog.c', LIBS='foo', LIBPATH='.')
    

is equivalent to:

      Program('prog.c', LIBS=['foo'], LIBPATH='.')
    

This is similar to the way that SCons handles either a string or a list to specify a single source file.

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'
    

(Note that Python requires that the backslash separators in a Windows path name be escaped within strings.)

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.

Builder Methods Return Lists of Target Nodes

All builder methods return a list of Node objects that identify the target file or files that will be built. These returned Nodes can be passed as source files to other builder methods,

For example, suppose that we want to build the two object files that make up a program with different options. This would mean calling the Object builder once for each object file, specifying the desired options:

    Object('hello.c', CCFLAGS='-DHELLO')
    Object('goodbye.c', CCFLAGS='-DGOODBYE')
    

One way to combine these object files into the resulting program would be to call the Program builder with the names of the object files listed as sources:

    Object('hello.c', CCFLAGS='-DHELLO')
    Object('goodbye.c', CCFLAGS='-DGOODBYE')
    Program(['hello.o', 'goodbye.o'])
    

The problem with listing the names as strings is that our SConstruct file is no longer portable across operating systems. It won't, for example, work on Windows because the object files there would be named hello.obj and goodbye.obj, not hello.o and goodbye.o.

A better solution is to assign the lists of targets returned by the calls to the Object builder to variables, which we can then concatenate in our call to the Program builder:

      hello_list = Object('hello.c', CCFLAGS='-DHELLO')
      goodbye_list = Object('goodbye.c', CCFLAGS='-DGOODBYE')
      Program(hello_list + goodbye_list)
    

This makes our SConstruct file portable again, the build output on Linux looking like:

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

And on Windows:

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

We'll see examples of using the list of nodes returned by builder methods throughout the rest of this guide.

Explicitly Creating File and Directory Nodes

It's worth mentioning here that SCons maintains a clear distinction between Nodes that represent files and Nodes that represent directories. SCons supports File and Dir functions that, repectively, return a file or directory Node:

      hello_c = File('hello.c')
      Program(hello_c)

      classes = Dir('classes')
      Java(classes, 'src')
    

Normally, you don't need to call File or Dir directly, because calling a builder method automatically treats strings as the names of files or directories, and translates them into the Node objects for you. The File and Dir functions can come in handy in situations where you need to explicitly instruct SCons about the type of Node being passed to a builder or other function, or unambiguously refer to a specific file in a directory tree.

There are also times when you may need to refer to an entry in a file system without knowing in advance whether it's a file or a directory. For those situations, SCons also supports an Entry function, which returns a Node that can represent either a file or a directory.

    xyzzy = Entry('xyzzy')
    

The returned xyzzy Node will be turned into a file or directory Node the first time it is used by a builder method or other function that requires one vs. the other.

Printing Node File Names

One of the most common things you can do with a Node is use it to print the file name that the node represents. For example, the following SConstruct file:

      hello_c = File('hello.c')
      Program(hello_c)

      classes = Dir('classes')
      Java(classes, 'src')

      object_list = Object('hello.c')
      program_list = Program(object_list)
      print "The object file is:", object_list[0]
      print "The program file is:", program_list[0]
    

Would print the following file names on a POSIX system:

      % scons -Q
      The object file is: hello.o
      The program file is: hello
      cc -c -o hello.o hello.c
      cc -o hello hello.o
    

And the following file names on a Windows system:

      C:\>scons -Q
      The object file is: hello.obj
      The program file is: hello.exe
      cl /nologo /c hello.c /Fohello.obj
      link /nologo /OUT:hello.exe hello.obj
    

Using a Node's File Name as a String

Printing a Node's name as described in the previous section works because the string representation of a Node is the name of the file. If you want to do something other than print the name of the file, you can fetch it by using the builtin Python str function. For example, if you want to use the Python os.path.exists to figure out whether a file exists while the SConstruct file is being read and executed, you can fetch the string as follows:

      import os.path
      program_list = Program('hello.c')
      program_name = str(program_list[0])
      if not os.path.exists(program_name)
          print program_name, "does not exist!"
    

Which executes as follows on a POSIX system:

      % scons -Q
      The object file is: hello.o
      The program file is: hello
      cc -c -o hello.o hello.c
      cc -o hello hello.o
    

Deciding When a Source File Has Changed: the SourceSignatures Function

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.

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

         % scons -Q hello
         cc -c -o hello.o hello.c
         cc -o hello hello.o
         % edit hello.c
             [CHANGE THE CONTENTS OF hello.c]
         % scons -Q hello
         cc -c -o hello.o hello.c
         cc -o hello hello.o
      

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

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

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

Deciding When a Target File Has Changed: the TargetSignatures Function

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 configure separately how the signatures of "intermediate" target files are used when deciding if a dependent target file must be rebuilt.

         % scons -Q hello
         cc -c -o hello.o hello.c
         cc -o hello hello.o
         % edit hello.c
             [CHANGE THE CONTENTS OF hello.c]
         % scons -Q hello
         cc -c -o hello.o hello.c
         cc -o hello hello.o
      

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

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

         % scons -Q hello
         cc -c -o hello.o hello.c
         cc -o hello hello.o
         % edit hello.c
           [CHANGE A COMMENT IN hello.c]
         % scons -Q hello
         cc -c -o hello.o hello.c
         scons: `hello' is up to date.
      

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)
    

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

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

Ignoring Dependencies: 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')
    

Explicit Dependencies: 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 an 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.
   

If you want to loop through and print the values of all of the construction variables in a construction environment, the Python code to do that in sorted order might look something like:

      env = Environment()
      dict = env.Dictionary()
      keys = dict.keys()
      keys.sort()
      for key in keys:
          print "construction variable = '%s', value = '%s'" % (key, dict[key])
   

Expanding Values From a Construction Environment

Another way to get information from a construction environment. is to use the subst method on a string containing $-expansions of construction variable names. As a simple example, the example from the previous section that used env['CC'] to fetch the value of CC could also be written as:

     env = Environment()
     print "CC is:", env.subst('$CC')
   

The real advantage of using subst to expand strings is that construction variables in the result get re-expanded until there are no expansions left in the string. So a simple fetch of a value like $CCCOM:

     env = Environment(CCFLAGS = '-DFOO')
     print "CCCOM is:", env['CCCOM']
   

Will print the unexpanded value of CCCOM, showing us the construction variables that still need to be expanded:

     % scons -Q
     CCCOM is: $CC $CCFLAGS $CPPFLAGS $_CPPDEFFLAGS $_CPPINCFLAGS -c -o $TARGET $SOURCES
     scons: `.' is up to date.
   

Calling the subst method on $CCOM, however:

     env = Environment(CCFLAGS = '-DFOO')
     print "CCCOM is:", env.subst('$CCCOM')
   

Will recursively expand all of the $-prefixed construction variables, showing us the final output:

     % scons -Q
     CCCOM is: gcc -DFOO -c -o
     scons: `.' is up to date.
   

(Note that because we're not expanding this in the context of building something there are no target or source files for $TARGET and $SOURCES to expand.

Modifying a Construction Environment

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

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

        % scons -Q
        cc -DDEFINE2 -c -o foo.o foo.c
        cc -o foo foo.o
     

        env = Environment()
        env.Replace(NEW_VARIABLE = 'xyzzy')
        print "NEW_VARIABLE =", env['NEW_VARIABLE']
     

        % scons -Q
        NEW_VARIABLE = xyzzy
        scons: `.' is up to date.
     

        env = Environment(CCFLAGS = '-DDEFINE1')
        print "CCFLAGS =", env['CCFLAGS']
        env.Program('foo.c')

        env.Replace(CCFLAGS = '-DDEFINE2')
        print "CCFLAGS =", env['CCFLAGS']
        env.Program('bar.c')
     

        % scons
        scons: Reading SConscript files ...
        CCFLAGS = -DDEFINE1
        CCFLAGS = -DDEFINE2
        scons: done reading SConscript files.
        scons: Building targets ...
        cc -DDEFINE2 -c -o bar.o bar.c
        cc -o bar bar.o
        cc -DDEFINE2 -c -o foo.o foo.c
        cc -o foo foo.o
        scons: done building targets.
     

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

        % scons -Q
        cc -DMY_VALUE -DLAST -c -o foo.o foo.c
        cc -o foo foo.o
     

        env = Environment()
        env.Append(NEW_VARIABLE = 'added')
        print "NEW_VARIABLE =", env['NEW_VARIABLE']
     

        % scons -Q
        NEW_VARIABLE = added
        scons: `.' is up to date.
     

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

        % scons -Q
        cc -DFIRST -DMY_VALUE -c -o foo.o foo.c
        cc -o foo foo.o
     

        env = Environment()
        env.Prepend(NEW_VARIABLE = 'added')
        print "NEW_VARIABLE =", env['NEW_VARIABLE']
     

        % scons -Q
        NEW_VARIABLE = added
        scons: `.' is up to date.
     

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)
      

         % scons
         scons: Reading SConscript files ...
         DEFAULT_TARGETS is ['prog1']
         scons: done reading SConscript files.
         scons: Building targets ...
         cc -c -o prog1.o prog1.c
         cc -o prog1 prog1.o
         scons: done building 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)
      

         % scons
         scons: Reading SConscript files ...
         DEFAULT_TARGETS is now ['prog1']
         DEFAULT_TARGETS is now ['prog1', 'prog2']
         scons: done reading SConscript files.
         scons: Building targets ...
         cc -c -o prog1.o prog1.c
         cc -o prog1 prog1.o
         cc -c -o prog2.o prog2.c
         cc -o prog2 prog2.o
         scons: done building 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', 'Set to build for release', 0))
           env = Environment(options = opts,
                             CPPDEFINES={'RELEASE_BUILD' : '${RELEASE}'})
           env.Program('foo.c')
      

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

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

        % scons -Q RELEASE=no foo.o
        cc -DRELEASE_BUILD=0 -c -o foo.o foo.c
      

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

        % scons -Q RELEASE=bad_value foo.o
        
        scons: *** Error converting option: RELEASE
        Invalid value for boolean option: bad_value
        File "SConstruct", line 4, in ?
      

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

        % scons -Q COLOR=red foo.o
        cc -DCOLOR="red" -c -o foo.o foo.c
        % scons -Q COLOR=blue foo.o
        cc -DCOLOR="blue" -c -o foo.o foo.c
        % scons -Q COLOR=green foo.o
        cc -DCOLOR="green" -c -o foo.o foo.c
      

        % scons -Q COLOR=magenta foo.o
        
        scons: *** Invalid value for option COLOR: magenta
        File "SConstruct", line 5, in ?
      

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

        % scons -Q COLOR=navy foo.o
        cc -DCOLOR="blue" -c -o foo.o foo.c
      

        % scons -Q COLOR=Red foo.o
        
        scons: *** Invalid value for option COLOR: Red
        File "SConstruct", line 5, in ?
        % scons -Q COLOR=BLUE foo.o
        
        scons: *** Invalid value for option COLOR: BLUE
        File "SConstruct", line 5, in ?
        % scons -Q COLOR=nAvY foo.o
        
        scons: *** Invalid value for option COLOR: nAvY
        File "SConstruct", line 5, in ?
      

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

        % scons -Q COLOR=Red foo.o
        cc -DCOLOR="Red" -c -o foo.o foo.c
        % scons -Q COLOR=BLUE foo.o
        cc -DCOLOR="BLUE" -c -o foo.o foo.c
        % scons -Q COLOR=nAvY foo.o
        cc -DCOLOR="blue" -c -o foo.o foo.c
        % scons -Q COLOR=green foo.o
        cc -DCOLOR="green" -c -o foo.o foo.c
      

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

        % scons -Q COLOR=Red foo.o
        cc -DCOLOR="red" -c -o foo.o foo.c
        % scons -Q COLOR=nAvY foo.o
        cc -DCOLOR="blue" -c -o foo.o foo.c
        % scons -Q COLOR=GREEN foo.o
        cc -DCOLOR="green" -c -o foo.o foo.c
      

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

        % scons -Q COLORS=red,blue foo.o
        cc -DCOLORS="red blue" -c -o foo.o foo.c
        % scons -Q COLORS=blue,green,red foo.o
        cc -DCOLORS="blue green red" -c -o foo.o foo.c
      

        % scons -Q COLORS=all foo.o
        cc -DCOLORS="red green blue" -c -o foo.o foo.c
        % scons -Q COLORS=none foo.o
        cc -DCOLORS="" -c -o foo.o foo.c
      

        % scons -Q COLORS=magenta foo.o
        
        scons: *** Error converting option: COLORS
        Invalid value(s) for option: magenta
        File "SConstruct", line 5, in ?
      

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

        % scons -Q foo.o
        cc -DCONFIG_FILE="/etc/my_config" -c -o foo.o foo.c
        % scons -Q CONFIG=/usr/local/etc/other_config foo.o
        scons: `foo.o' is up to date.
      

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

        % scons -Q foo.o
        cc -DPACKAGE="/opt/location" -c -o foo.o foo.c
        % scons -Q PACKAGE=/usr/local/location foo.o
        cc -DPACKAGE="/usr/local/location" -c -o foo.o foo.c
        % scons -Q PACKAGE=yes foo.o
        cc -DPACKAGE="1" -c -o foo.o foo.c
        % scons -Q PACKAGE=no foo.o
        cc -DPACKAGE="0" -c -o foo.o 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 the chapter called Separating Source and Build Directories, 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 the chapter called Separating Source and Build Directories, 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')
        

        % scons -Q
        cc -c -o bar/bar.o bar/bar.c
        cc -c -o foo/foo.o foo/foo.c
        ar r libprog.a foo/foo.o bar/bar.o
        ranlib libprog.a
      

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':
        File "SConstruct", line 4:
          env.Program('hello.c')
    

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.)

Configure Contexts

The basic framework for multi-platform build configuration in SCons is to attach a configure context to a construction environment by calling the Configure function, perform a number of checks for libraries, functions, header files, etc., and to then call the configure context's Finish method to finish off the configuration:

    env = Environment()
    conf = Configure(env)
    # Checks for libraries, header files, etc. go here!
    env = conf.Finish()
    

The next sections describe the basic checks that SCons supports, as well as how to add your own custom checks.

Checking for the Existence of Header Files

Testing the existence of a header file requires knowing what language the header file is. A configure context has a CheckCHeader method that checks for the existence of a C header file:

    env = Environment()
    conf = Configure(env)
    if not conf.CheckCHeader('math.h'):
        print 'Math.h must be installed!'
        Exit(1)
    if conf.CheckCHeader('foo.h'):
        conf.env.Append('-DHAS_FOO_H')
    env = conf.Finish()
    

Note that you can choose to terminate the build if a given header file doesn't exist, or you can modify the contstruction environment based on the existence of a header file.

If you need to check for the existence a C++ header file, use the CheckCXXHeader method:

    env = Environment()
    conf = Configure(env)
    if not conf.CheckCXXHeader('vector.h'):
        print 'vector.h must be installed!'
        Exit(1)
    env = conf.Finish()
    

Checking for the Availability of a Function

Check for the availability of a specific function using the CheckFunc method:

    env = Environment()
    conf = Configure(env)
    if not conf.CheckFunc('strcpy'):
        print 'Did not find strcpy(), using local version'
        conf.env.Append('-Dstrcpy=my_local_strcpy')
    env = conf.Finish()
    

Checking for the Availability of a Library

Check for the availability of a library using the CheckLib method. You only specify the basename of the library, you don't need to add a lib prefix or a .a or .lib suffix:

    env = Environment()
    conf = Configure(env)
    if not conf.CheckLib('m'):
        print 'Did not find libm.a or m.lib, exiting!'
        Exit(1)
    env = conf.Finish()
    

Because the ability to use a library successfully often depends on having access to a header file that describes the library's interface, you can check for a library and a header file at the same time by using the CheckLibWithHeader method:

    env = Environment()
    conf = Configure(env)
    if not conf.CheckLibWithHeader('m', 'math.h'):
        print 'Did not find libm.a or m.lib, exiting!'
        Exit(1)
    env = conf.Finish()
    

This is essentially shorthand for separate calls to the CheckHeader and CheckLib functions.

Checking for the Availability of a typedef

Check for the availability of a typedef by using the CheckType method:

    env = Environment()
    conf = Configure(env)
    if not conf.CheckType('off_t'):
        print 'Did not find off_t typedef, assuming int'
        conf.env.Append(CCFLAGS = '-Doff_t=int')
    env = conf.Finish()
    

You can also add a string that will be placed at the beginning of the test file that will be used to check for the typedef. This provide a way to specify files that must be included to find the typedef:

    env = Environment()
    conf = Configure(env)
    if not conf.CheckType('off_t', '#include <sys/types.h>\n'):
        print 'Did not find off_t typedef, assuming int'
        conf.env.Append(CCFLAGS = '-Doff_t=int')
    env = conf.Finish()
    

Adding Your Own Custom Checks

A custom check is a Python function that checks for a certain condition to exist on the running system, usually using methods that SCons supplies to take care of the details of checking whether a compilation succeeds, a link succeeds, a program is runnable, etc. A simple custom check for the existence of a specific library might look as follows:

    mylib_test_source_file = """
    #include <mylib.h>
    int main(int argc, char **argv)
    {
        MyLibrary mylib(argc, argv);
        return 0;
    }
    """

    def CheckMyLibrary(context):
        context.Message('Checking for MyLibrary...')
        result = context.TryLink(mylib_test_source_file, '.c')
        context.Result(result)
        return result
    

The Message and Result methods should typically begin and end a custom check to let the user know what's going on: the Message call prints the specified message (with no trailing newline) and the Result call prints ok if the check succeeds and failed if it doesn't. The TryLink method actually tests for whether the specified program text will successfully link.

(Note that a custom check can modify its check based on any arguments you choose to pass it, or by using or modifying the configure context environment in the context.env attribute.)

This custom check function is then attached to the configure context by passing a dictionary to the Configure call that maps a name of the check to the underlying function:

    env = Environment()
    conf = Configure(env, custom_tests = {'CheckMyLibrary' : CheckMyLibrary})
    

You'll typically want to make the check and the function name the same, as we've done here, to avoid potential confusion.

We can then put these pieces together and actually call the CheckMyLibrary check as follows:

    mylib_test_source_file = """
    #include <mylib.h>
    int main(int argc, char **argv)
    {
        MyLibrary mylib(argc, argv);
        return 0;
    }
    """

    def CheckMyLibrary(context):
        context.Message('Checking for MyLibrary... ')
        result = context.TryLink(mylib_test_source_file, '.c')
        context.Result(result)
        return result

    env = Environment()
    conf = Configure(env, custom_tests = {'CheckMyLibrary' : CheckMyLibrary})
    if not conf.CheckMyLibrary():
        print 'MyLibrary is not installed!'
        Exit(1)
    env = conf.Finish()

    # We would then add actual calls like Program() to build
    # something using the "env" construction environment.
    

If MyLibrary is not installed on the system, the output will look like:

    % scons
    scons: Reading SConscript file ...
    Checking for MyLibrary... failed
    MyLibrary is not installed!
    

If MyLibrary is installed, the output will look like:

    % scons
    scons: Reading SConscript file ...
    Checking for MyLibrary... failed
    scons: done reading SConscript
    scons: Building targets ...
        .
        .
        .
    

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.

Building Java Class Files: the Java Builder

The basic activity when programming in Java, of course, is to take one or more .java files containing Java source code and to call the Java compiler to turn them into one or more .class files. In SCons, you do this by giving the Java Builder a target directory in which to put the .class files, and a source directory that contains the .java files:

      Java('classes', 'src')
    

If the src directory contains three .java source files, then running SCons might look like this:

      % scons -Q
      javac -d classes -sourcepath src src/Example1.java src/Example2.java src/Example3.java
    

SCons will actually search the src directory tree for all of the .java files. The Java compiler will then create the necessary class files in the classes subdirectory, based on the class names found in the .java files.

How SCons Handles Java Dependencies

In addition to searching the source directory for .java files, SCons actually runs the .java files through a stripped-down Java parser that figures out what classes are defined. In other words, SCons knows, without you having to tell it, what .class files will be produced by the javac call. So our one-liner example from the preceding section:

      Java('classes', 'src')
    

Will not only tell you reliably that the .class files in the classes subdirectory are up-to-date:

      % scons -Q
      javac -d classes -sourcepath src src/Example1.java src/Example2.java src/Example3.java
      % scons -Q classes
      scons: `classes' is up to date.
    

But it will also remove all of the generated .class files, even for inner classes, without you having to specify them manually. For example, if our Example1.java and Example3.java files both define additional classes, and the class defined in Example2.java has an inner class, running scons -c will clean up all of those .class files as well:

      % scons -Q
      javac -d classes -sourcepath src src/Example1.java src/Example2.java src/Example3.java
      % scons -Q -c classes
      Removed classes/Example1.class
      Removed classes/AdditionalClass1.class
      Removed classes/Example2$Inner2.class
      Removed classes/Example2.class
      Removed classes/Example3.class
      Removed classes/AdditionalClass3.class
    

Building Java Archive (.jar) Files: the Jar Builder

After building the class files, it's common to collect them into a Java archive (.jar) file, which you do by calling the Jar Builder method. If you want to just collect all of the class files within a subdirectory, you can just specify that subdirectory as the Jar source:

      Java(target = 'classes', source = 'src')
      Jar(target = 'test.jar', source = 'classes')
    

SCons will then pass that directory to the jar command, which will collect all of the underlying .class files:

      % scons -Q
      javac -d classes -sourcepath src src/Example1.java src/Example2.java src/Example3.java
      jar cf test.jar classes
    

If you want to keep all of the .class files for multiple programs in one location, and only archive some of them in each .jar file, you can pass the Jar builder a list of files as its source. It's extremely simple to create multiple .jar files this way, using the lists of target class files created by calls to the Java builder as sources to the various Jar calls:

      prog1_class_files = Java(target = 'classes', source = 'prog1')
      prog2_class_files = Java(target = 'classes', source = 'prog2')
      Jar(target = 'prog1.jar', source = prog1_class_files)
      Jar(target = 'prog2.jar', source = prog2_class_files)
    

This will then create prog1.jar and prog2.jar next to the subdirectories that contain their .java files:

      % scons -Q
      javac -d classes -sourcepath prog1 prog1/Example1.java prog1/Example2.java
      javac -d classes -sourcepath prog2 prog2/Example3.java prog2/Example4.java
      jar cf prog1.jar classes/Example1.class classes/Example2.class
      jar cf prog2.jar classes/Example3.class classes/Example4.class
    

Building C Header and Stub Files: the JavaH Builder

You can generate C header and source files for implementing native methods, by using the JavaH Builder. There are several ways of using the JavaH Builder. One typical invocation might look like:

      classes = Java(target = 'classes', source = 'src/pkg/sub')
      JavaH(target = 'native', source = classes)
    

The source is a list of class files generated by the call to the Java Builder, and the target is the output directory in which we want the C header files placed. The target gets converted into the -d when SCons runs javah:

      % scons -Q
      javac -d classes -sourcepath src/pkg/sub src/pkg/sub/Example1.java src/pkg/sub/Example2.java src/pkg/sub/Example3.java
      javah -d native -classpath classes pkg.sub.Example1 pkg.sub.Example2 pkg.sub.Example3
    

In this case, the call to javah will generate the header files native/pkg_sub_Example1.h, native/pkg_sub_Example2.h and native/pkg_sub_Example3.h. Notice that SCons remembered that the class files were generated with a target directory of classes, and that it then specified that target directory as the -classpath option to the call to javah.

Although it's more convenient to use the list of class files returned by the Java Builder as the source of a call to the JavaH Builder, you can specify the list of class files by hand, if you prefer. If you do, you need to set the JAVACLASSDIR construction variable when calling JavaH:

      Java(target = 'classes', source = 'src/pkg/sub')
      class_file_list = ['classes/pkg/sub/Example1.class',
                         'classes/pkg/sub/Example2.class',
                         'classes/pkg/sub/Example3.class']
      JavaH(target = 'native', source = class_file_list, JAVACLASSDIR = 'classes')
    

The JAVACLASSDIR value then gets converted into the -classpath when SCons runs javah:

      % scons -Q
      javac -d classes -sourcepath src/pkg/sub src/pkg/sub/Example1.java src/pkg/sub/Example2.java src/pkg/sub/Example3.java
      javah -d native -classpath classes pkg.sub.Example1 pkg.sub.Example2 pkg.sub.Example3
    

Lastly, if you don't want a separate header file generated for each source file, you can specify an explicit File Node as the target of the JavaH Builder:

      classes = Java(target = 'classes', source = 'src/pkg/sub')
      JavaH(target = File('native.h'), source = classes)
    

Because SCons assumes by default that the target of the JavaH builder is a directory, you need to use the File function to make sure that SCons doesn't create a directory named native.h. When a file is used, though, SCons correctly converts the file name into the javah -o option:

      % scons -Q
      javac -d classes -sourcepath src/pkg/sub src/pkg/sub/Example1.java src/pkg/sub/Example2.java src/pkg/sub/Example3.java
      javah -o native.h -classpath classes pkg.sub.Example1 pkg.sub.Example2 pkg.sub.Example3
    

Building RMI Stub and Skeleton Class Files: the RMIC Builder

You can generate Remote Method Invocation stubs by using the RMIC Builder. The source is a list of directories, typically returned by a call to the Java Builder, and the target is an output directory where the _Stub.class and _Skel.class files will be placed:

      classes = Java(target = 'classes', source = 'src/pkg/sub')
      RMIC(target = 'outdir', source = classes)
    

As it did with the JavaH Builder, SCons remembers the class directory and passes it as the -classpath option to rmic:

      % scons -Q
      javac -d classes -sourcepath src/pkg/sub src/pkg/sub/Example1.java src/pkg/sub/Example2.java
      rmic -d outdir -classpath classes pkg.sub.Example1 pkg.sub.Example2
    

This example would generate the files outdir/pkg/sub/Example1_Skel.class, outdir/pkg/sub/Example1_Stub.class, outdir/pkg/sub/Example2_Skel.class and outdir/pkg/sub/Example2_Stub.class.

Why is That Target Being Rebuilt? the --debug=explain Option

Let's take a simple example of a misconfigured build that causes a target to be rebuilt every time SCons is run:

      # Intentionally misspell the output file name in the
      # command used to create the file:
      Command('file.out', 'file.in', 'cp $SOURCE file.oout')
    

(Note to Windows users: The POSIX cp command copies the first file named on the command line to the second file. In our example, it copies the file.in file to the file.out file.)

Now if we run SCons multiple on this example, we see that it re-runs the cp command every time:

      % scons -Q
      cp file.in file.oout
      % scons -Q
      cp file.in file.oout
      % scons -Q
      cp file.in file.oout
    

In this example, the underlying cause is obvious: we've intentionally misspelled the output file name in the cp command, so the command doesn't actually build the file.out file that we've told SCons to expect. But if the problem weren't obvious, it would be helpful to specify the --debug=explain option on the command line to have SCons tell us very specifically why it's decided to rebuild the target:

      % scons -Q --debug=explain
      scons: building `file.out' because it doesn't exist
      cp file.in file.oout
    

If this had been a more complicated example involving a lot of build output, having SCons tell us that it's trying to rebuild the target file because it doesn't exist would be an important clue that something was wrong with the command that we invoked to build it.

The --debug=explain option also comes in handy to help figure out what input file changed. Given a simple configuration that builds a program from three source files, changing one of the source files and rebuilding with the --debug=explain option shows very specifically why SCons rebuilds the files that it does:

      % scons -Q
      cc -c -o file1.o file1.c
      cc -c -o file2.o file2.c
      cc -c -o file3.o file3.c
      cc -o prog file1.o file2.o file3.o
      % edit file2.c
          [CHANGE THE CONTENTS OF file2.c]
      % scons -Q --debug=explain
      scons: rebuilding `file2.o' because `file2.c' changed
      cc -c -o file2.o file2.c
      scons: rebuilding `prog' because `file2.o' changed
      cc -o prog file1.o file2.o file3.o
    

This becomes even more helpful in identifying when a file is rebuilt due to a change in an implicit dependency, such as an incuded .h file. If the file1.c and file3.c files in our example both included a hello.h file, then changing that included file and re-running SCons with the --debug=explain option will pinpoint that it's the change to the included file that starts the chain of rebuilds:

      % scons -Q
      cc -I. -c -o file1.o file1.c
      cc -I. -c -o file2.o file2.c
      cc -I. -c -o file3.o file3.c
      cc -o prog file1.o file2.o file3.o
      % edit hello.h
          [CHANGE THE CONTENTS OF hello.h]
      % scons -Q --debug=explain
      scons: rebuilding `file1.o' because `hello.h' changed
      cc -I. -c -o file1.o file1.c
      scons: rebuilding `file3.o' because `hello.h' changed
      cc -I. -c -o file3.o file3.c
      scons: rebuilding `prog' because:
                 `file1.o' changed
                 `file3.o' changed
      cc -o prog file1.o file2.o file3.o