/usr/share/doc/fcm/user_guide/build.html

<!DOCTYPE html>
<html>
<head>
  <title>FCM: User Guide: Annex: The FCM 1 Build System</title>
  <meta name="author" content="FCM team" />
  <meta name="viewport" content="width=device-width, initial-scale=1.0" />
  <link rel="icon" href="../etc/fcm-icon.png" type="image/png" />
  <link rel="shortcut icon" href="../etc/fcm-icon.png" type="image/png" />
  <link href="../etc/bootstrap/css/bootstrap.min.css" rel="stylesheet" media="screen" />
  <link href="../etc/fcm.css" rel="stylesheet" media="screen" />
</head>
<body>
  <div class="navbar navbar-inverse">
    <div class="container-fluid">
      <div class="navbar-header">
        <a class="navbar-brand" href=".."><span class="fcm-version">FCM</span></a>
      </div>
      <div class="collapse navbar-collapse">
        <ul class="nav navbar-nav">
          <li><a href="../installation/">Installation</a></li>

          <li><a class="active" href="#">User Guide</a></li>
        </ul>
      </div>
    </div>
  </div>

  <div class="page-header">
    <div class="fcm-page-content pull-right well well-sm"></div>
    <h1>FCM: User Guide: Annex: The FCM 1 Build System</h1>
  </div>

  <div class="container">
  <div class="row">
  <div class="col-md-12">

  <h2 id="introduction">Introduction</h2>

  <p><em>The FCM 1 build system is deprecated. The documentation for the current
  build system can be found at <a href="make.html">FCM Make</a>.</em></p>

  <p>The build system analyses the directory tree containing a set of source
  code, processes the configuration, and invokes <code>make</code> to
  compile/build the source code into the project executables. In this chapter,
  we shall use many examples to explain how to use the build system. At the end
  of this chapter, you should be able to use the build system, either by
  defining the build configuration file directly or by using the extract system
  to generate a suitable build configuration file.</p>

  <h2 id="command">The Build Command</h2>

  <p>To invoke the build system, simply issue the command:</p>
  <pre>
fcm build
</pre>

  <p>By default, the build system searches for a build configuration file
  <samp>bld.cfg</samp> in <samp>$PWD</samp> and then <samp>$PWD/cfg</samp>. If
  a build configuration file is not found in these directories, the command
  fails with an error. If a build configuration file is found, the system will
  use the configuration specified in the file to perform the build. If you use
  the extract system to extract your source tree, a build configuration should
  be written for you automatically at the <samp>cfg/</samp> sub-directory of
  the destination root directory.</p>

  <p>If the root directory of the build does not exist, the system performs a
  new full build at this directory. If a previous build already exists at this
  directory, the system performs an incremental build. If a full (fresh) build
  is required for whatever reason, you can invoke the build system using the
  <code>-f</code> option, (i.e. the command becomes <code>fcm build -f</code>).
  If you simply want to remove all the items generated by a previous build in
  the destination, you can invoke the build system using the
  <code>--clean</code> option.</p>

  <p>The build system uses GNU <code>make</code> to perform the majority of the
  build. GNU <code>make</code> has a <code>-j jobs</code> option to specify the
  number of <var>jobs</var> to run simultaneously. Invoking the build system
  with the same option triggers this option when the build system invokes the
  <code>make</code> command. The argument to the option <var>jobs</var> must be
  an integer. The default is <samp>1</samp>. For example, the command <code>fcm
  build -j 4</code> will allow <code>make</code> to perform 4 jobs
  simultaneously.</p>

  <p>For further information on the build command, please see <a href=
  "command_ref.html#fcm-build">FCM Command Reference &gt; fcm build</a>.</p>

  <h2 id="basic">Basic Features</h2>

  <p>The build configuration file is the user interface of the build system. It
  is a line based text file. You can create your own build configuration file
  or you can use the extract system to create one for you. For a complete set
  of build configuration file declarations, please refer to the <a href=
  "annex_bld_cfg.html">Annex: Declarations in FCM build configuration
  file</a>.</p>

  <h3 id="basic_build">Basic build configuration</h3>

  <p>Suppose we have a directory at <samp>$HOME/example</samp>. Its
  sub-directory at <samp>$HOME/example/src</samp> contains a source tree to be
  built. You may want to have a build configuration file
  <samp>$HOME/example/cfg/bld.cfg</samp>, which may contain:</p>
  <pre id="example_1">
# Example 1
# ----------------------------------------------------------------------
cfg::type     bld                           # line 1
cfg::version  1.0                           # line 2

dest          $HOME/example                 # line 4

target        foo.exe bar.exe               # line 6

tool::fc      ifort                         # line 8
tool::fflags  -O3                           # line 9
tool::cc      gcc                           # line 10
tool::cflags  -O3                           # line 11

tool::ldflags -O3 -L$(HOME)/lib -legg -lham # line 13
</pre>

  <p>Here is an explanation of what each line does:</p>

  <ul>
    <li><dfn>line 1</dfn>: the label <code>CFG::TYPE</code> declares the type
    of the configuration file. The value <samp>bld</samp> tells the system that
    it is a build configuration file.</li>

    <li><dfn>line 2</dfn>: the label <code>CFG::VERSION</code> declares the
    version of the build configuration file. The current default is
    <samp>1.0</samp>. Although it is not currently used, if we have to change
    the format of the configuration file at a later stage, we shall be able to
    use this number to determine whether we are reading a file with an older
    format or one with a newer format.</li>

    <li><dfn>line 4</dfn>: the label <code>DEST</code> declares the root
    directory of the current build.</li>

    <li><dfn>line 6</dfn>: the label <code>TARGET</code> declares a list of
    <em>default</em> targets. The default targets of the current build will be
    <samp>foo.exe</samp> and <samp>bar.exe</samp>.</li>

    <li><dfn>line 8</dfn>: the label <code>TOOL::FC</code> declares the Fortran
    compiler command.</li>

    <li><dfn>line 9</dfn>: the label <code>TOOL::FFLAGS</code> declares the
    options to be used when invoking the Fortran compiler command.</li>

    <li><dfn>line 10</dfn>: the label <code>TOOL::CC</code> declares the C
    compiler command.</li>

    <li><dfn>line 11</dfn>: the label <code>TOOL::CFLAGS</code> declares the
    options to be used when invoking the C compiler command.</li>

    <li><dfn>line 13</dfn>: the label <code>TOOL::LDFLAGS</code> declares the
    options to be used when invoking the linker command.</li>
  </ul>

  <p>When we invoke the build system, it reads the above configuration file. It
  will go through various internal processes, such as dependency generations,
  to obtain the required information to prepare the <samp>Makefile</samp> of
  the build. (All of which will be described in later sections.) The
  <samp>Makefile</samp> of the build will be placed at
  <samp>$HOME/example/bld</samp>. The system will then invoke <code>make</code>
  to build the targets specified in line 6, i.e. <samp>foo.exe</samp> and
  <samp>bar.exe</samp> using the build tools specified between line 8 to line
  13. On a successful build, the target executables will be sent to
  <samp>$HOME/example/bin/</samp>. The build system also creates a shell script
  called <samp>fcm_env.sh</samp> in <samp>$HOME/example/</samp>. If you source
  the shell script, it will export your <var>PATH</var> environment variable to
  search the <samp>$HOME/example/bin/</samp> directory for executables.</p>

  <p>N.B. You may have noticed that the <code>-c</code> (compile to object file
  only) option is missing from the compiler flags declarations. This is because
  the option is inserted automatically by the build system, unless it is
  already declared.</p>

  <p>N.B. You can declare the linker using <code>TOOL::LD</code>. If it is not
  specified, the default is to use the compiler command for the source file
  containing the main program.</p>

  <dl>
    <dt>Note - declaration of source files for build</dt>

    <dd>
      <p>Source files do not have to reside in the <samp>src/</samp>
      sub-directory of the build root directory. They can be anywhere, but you
      will have to declare them using the label <code>SRC::&lt;pcks&gt;</code>,
      where <code>&lt;pcks&gt;</code> is the sub-package name in which the
      source belongs. If a directory is specified then the build system
      automatically searches for all source files in this directory. E.g.</p>
      <pre>
# Declare a source in the sub-package "foo/bar"
src::foo/bar  $HOME/foo/bar
</pre>

      <p>By default, the build system searches the <samp>src/</samp>
      sub-directory of the build root directory for source files. If all source
      files are already declared explicitly, you can switch off the automatic
      directory search by setting the <code>SEARCH_SRC</code> flag to false.
      E.g.</p>
      <pre>
search_src  false
</pre>

      <p>As mentioned in the previous chapter, the name of a sub-package
      &lt;pcks&gt; provides a unique namespace for a file. The name of a
      sub-package is a list of words delimited by a slash <code>/</code>. (The
      system uses the double colons <code>::</code> and the double underscores
      <code>__</code> internally. Please avoid using <code>::</code> and
      <code>__</code> for naming your files and directories.)</p>

      <p>Currently, the build system only supports non-space characters in the
      package name, as the space character is used as a delimiter between the
      declaration label and its value. If there are spaces in the path name to
      a file or directory, you should explicity re-define the package name of
      that path to a package name with no space using the above method.
      However, we recommend that only non-space characters are used for naming
      directories and files to make life simple.</p>

      <p>In the build system, the sub-package name also provides an
      <em>inheritance</em> relationship for sub-packages. For instance, we may
      have a sub-package called <samp>foo/bar/egg</samp>, which belongs to the
      sub-package <samp>foo/bar</samp>, which belongs to the package
      <samp>foo</samp>.</p>

      <ul>
        <li>If we declare a global build tool, it applies to all packages.</li>

        <li>If we declare a build tool for <samp>foo</samp>, it applies also to
        the sub-package <samp>foo/bar</samp> and <samp>foo/bar/egg</samp>.</li>

        <li>If we declare a build tool for <samp>foo/bar</samp>, it applies
        also to <samp>foo/bar/egg</samp>, but not to other sub-packages in
        <samp>foo</samp>.</li>
      </ul>
    </dd>
  </dl>

  <h3 id="basic_extract">Build configuration via the extract system</h3>

  <p>As mentioned earlier, you can obtain a build configuration file through
  the extract system. The following example is what you may have in your
  extract configuration in order to obtain a similar configuration as <a href=
  "#example_1">example 1</a>:</p>
  <pre id="example_2">
# Example 2
# ----------------------------------------------------------------------
cfg::type          ext                           # line 1
cfg::version       1.0                           # line 2

dest               $HOME/example                 # line 4

bld::target        foo.exe bar.exe               # line 6

bld::tool::fc      ifort                         # line 8
bld::tool::fflags  -O3                           # line 9
bld::tool::cc      gcc                           # line 10
bld::tool::cflags  -O3                           # line 11

bld::tool::ldflags -O3 -L$(HOME)/lib -legg -lham # line 13

# ... and other declarations for source locations ...
</pre>

  <p>It is easy to note the similarities and differences between <a href=
  "#example_1">example 1</a> and <a href="#example_2">example 2</a>. <a href=
  "#example_2">Example 2</a> is an extract configuration file. It extracts to a
  destination root directory that will become the root directory of the build.
  Line 6 to line 13 are the same declarations, except that they are now
  prefixed with <code>BLD::</code>. In an extract configuration file, any lines
  prefixed with <code>BLD::</code> means that they are build configuration
  setting. These lines are ignored by the extract system but are parsed down to
  the output build configuration file, with the <code>BLD::</code> prefix
  removed. (Note: the <code>BLD::</code> prefix is optional for declarations in
  a build configuration file.)</p>

  <p>N.B. If you use the extract system to mirror an extract to an alternate
  location, the extract system will assume that the root directory of the
  alternate destination is the root directory of the build, and that the build
  will be carried out in that destination.</p>

  <h3 id="basic_exename">Naming of executables</h3>

  <p>If a source file called <samp>foo.f90</samp> contains a main program, the
  default behaviour of the system is to name its executable
  <samp>foo.exe</samp>. The root name of the executable is the same as the
  original file name, but its file extension is replaced with
  <samp>.exe</samp>. The output extension can be altered by re-registering the
  extension for output EXE files. How this can be done will be discussed later
  in the sub-section <a href="#advanced_file-type">File Type</a>.</p>

  <p>If you need to alter the full name of the executable, you can use the
  <code>EXE_NAME::</code> declaration. For example, the declaration:</p>
  <pre>
bld::exe_name::foo  bar
</pre>

  <p>will rename the executable of <samp>foo.f90</samp> from
  <samp>foo.exe</samp> to <samp>bar</samp>.</p>

  <p>Note: the declaration label is <code>bld::exe_name::foo</code> (not
  <code>bld::exe_name::foo.exe</code>) and the executable will be named
  <samp>bar</samp> (not <samp>bar.exe</samp>).</p>

  <h3 id="basic_flags">Setting the compiler flags</h3>

  <p>As discussed in the first example, the compiler commands and their flags
  can be set via the <code>TOOL::</code> declarations. A simple
  <code>TOOL::FFLAGS</code> declaration, for example, alters the compiler
  options for compiling all Fortran source files in the build. If you need to
  alter the compiler options only for the source files in a particular
  sub-package, it is possible to do so by adding the sub-package name to the
  declaration label. For example, the declaration label
  <code>TOOL::FFLAGS::foo/bar</code> will ensure that the declaration only
  applies to the code in the sub-package <samp>foo/bar</samp>. You can even
  make declarations down to the individual source file level. For example, the
  declaration label <code>TOOL::FFLAGS::foo/bar/egg.f90</code> will ensure that
  the declaration applies only for the file <samp>foo/bar/egg.f90</samp>.</p>

  <p>N.B. Although the prefix <code>TOOL::</code> and the tool names are
  case-insensitive, sub-package names are case sensitive in the declarations.
  Internally, tool names are turned into uppercase, and the sub-package
  delimiters are changed from the slash <code>/</code> (or double colons
  <code>::</code>) to the double underscores <code>__</code>. When the system
  generates the <samp>Makefile</samp> for the build, each <code>TOOL</code>
  declaration will be exported as an environment variable. For example, the
  declaration <code>tool::fflags/foo/bar</code> will be exported as
  <samp>FFLAGS__foo__bar</samp>.</p>

  <p>N.B. <code>TOOL</code> declarations for sub-packages are only accepted by
  the system when it is sensible to do so. For example, it allows you to
  declare different compiler flags, linker commands and linker flags for
  different sub-packages, but it does not accept different compilers for
  different sub-packages. If you attempt to make a <code>TOOL</code>
  declaration for a sub-package that does not exist, the build system will exit
  with an error.</p>

  <p>The following is an example setting in an extract configuration file based
  on <a href="#example_2">example 2</a>:</p>
  <pre id="example_3">
# Example 3
# ----------------------------------------------------------------------
cfg::type              ext
cfg::version           1.0

dest                   $HOME/example

bld::target            foo.exe bar.exe

bld::tool::fc          ifort
bld::tool::fflags      -O3    # line 9
bld::tool::cc          gcc
bld::tool::cflags      -O3

bld::tool::ldflags     -L$(HOME)/lib -legg -lham

bld::tool::fflags::ops -O1 -C # line 15
bld::tool::fflags::gen -O2    # line 16

# ... and other declarations for repositories and source directories ...
</pre>

  <p>In the example above, line 15 alters the Fortran compiler flags for
  <samp>ops</samp>, so that all source files in <samp>ops</samp> will be
  compiled with optimisation level 1 and will have runtime error checking
  switched on. Line 16, alters the Fortran compiler flags for <samp>gen</samp>,
  so that all source files in <samp>gen</samp> will be compiled with
  optimisation level 2. All other Fortran source files will use the global
  setting declared at line 9, so they they will all be compiled with
  optimisation level 3.</p>

  <dl>
    <dt>Note - changing compiler flags in incremental builds</dt>

    <dd>
      <p>Suppose you have performed a successful build using the configuration
      in <a href="#example_3">example 3</a>, and you have decided to change
      some of the compiler flags, you can do so by altering the appropriate
      flags in the build configuration file. When you trigger an incremental
      build, the system will detect changes in compiler flags automatically,
      and update only the required targets. The following hierarchy is
      followed:</p>

      <ul>
        <li>If the compiler flags for a particular source file change, only
        that source file and any targets depending on that source file are
        re-built.</li>

        <li>If the compiler flags for a container package change, only source
        files within that container package and any targets depending on those
        source files are re-built.</li>

        <li>If the global compiler flags change, all source files are
        re-built.</li>

        <li>If the compiler command changes, all source files are
        re-built.</li>
      </ul>
    </dd>
  </dl>

  <p>N.B. For a full list of build tools declarations, please see <a href=
  "annex_bld_cfg.html#tools-list">Annex: Declarations in FCM build
  configuration file &gt; list of tools</a>.</p>

  <h3 id="basic_interface">Automatic Fortran 9X interface block</h3>

  <p>For each Fortran 9X source file containing standalone subroutines and/or
  functions, the system generates an interface file and sends it to the
  <samp>inc/</samp> sub-directory of the build root. An interface file contains
  the interface blocks for the subroutines and functions in the original source
  file. In an incremental build, if you have modified a Fortran 9X source file,
  its interface file will only be re-generated if the content of the interface
  has changed.</p>

  <p>Consider a source file <samp>foo.f90</samp> containing a subroutine called
  <samp>foo</samp>. In a normal operation, the system writes the interface file
  to <samp>foo.interface</samp> in the <samp>inc/</samp> sub-directory of the
  build root. By default, the root name of the interface file is the same as
  that of the source file, and is case sensitive. You can change this behaviour
  using a <code>TOOL::INTERFACE</code> declaration. E.g.:</p>
  <pre>
bld::tool::interface  program # The default is "file"
</pre>

  <p>In such case, the root name of the interface file will be named in lower
  case after the first program unit in the file.</p>

  <p>The default extension for an interface file is <samp>.interface</samp>.
  This can be modified through the input and output file type register, which
  will be discussed in a later section on <a href="#advanced_file-type">File
  Type</a>.</p>

  <p>In most cases, we modify procedures without altering their calling
  interfaces. Consider another source file <samp>bar.f90</samp> containing a
  subroutine <samp>bar</samp>. If <samp>bar</samp> calls <samp>foo</samp>, it
  is good practice for <samp>bar</samp> to have an explicit interface for
  <samp>foo</samp>. This can be achieved if the subroutine <samp>bar</samp> has
  the following within its declaration section:</p>
  <pre>
INCLUDE 'foo.interface'
</pre>

  <p>The source file <samp>bar.f90</samp> is now dependent on the interface
  file <samp>foo.interface</samp>. This can make incremental build very
  efficient, as changes in the <samp>foo.f90</samp> file will not normally
  trigger the re-compilation of <samp>bar.f90</samp>, provided that the
  interface of the <code>subroutine foo</code> remains unchanged. (However, the
  system is clever enough to know that it needs to re-link any executables that
  are dependent on the object file for the <code>subroutine bar</code>.)</p>

  <p>By default, the system uses its own internal logic to extract the calling
  interfaces of top level subroutines and functions in a Fortran source file to
  generate an interface block. However, the system can also work with the
  interface generator <code>f90aib</code>, which is a freeware obtained from
  <a href=
  "http://www.ifremer.fr/ditigo/molagnon/fortran90/contenu.html">Fortran 90
  texts and programs, assembled by Michel Olagnon</a> at the French Research
  Institute for Exploitation of the Sea. To do so, you need to make a
  declaration in the build configuration file using the label
  <code>TOOL::GENINTERFACE</code>. As for any other <code>TOOL</code>
  declarations, you can attach a sub-package name to the label. The change will
  then apply only to source files within that sub-package. If
  <code>TOOL::GENINTERFACE</code> is declared to have the value
  <code>NONE</code>, interface generation will be switched off. The following
  are some examples:</p>
  <pre id="example_4">
# Example 4
# ----------------------------------------------------------------------
# This is an EXTRACT configuration file ...

# ... some other declarations ...

bld::tool::geninterface       f90aib # line 5
bld::tool::geninterface::bar  none   # line 6

# ... some other declarations ...
</pre>

  <p>In line 5, the global interface generator is now set to
  <code>f90aib</code>. In line 6, by setting the interface generator for the
  package <samp>bar</samp> to the <code>none</code> keyword, no interface file
  will be generated for source files under the package <samp>bar</samp>.</p>

  <p>Switching off the interface block generator can be useful in many
  circumstances. For example, if the interface block is already provided
  manually within the source tree, or if the interface block is never used by
  other program units, it is worth switching off the interface generator for
  the source file to speed up the build process.</p>

  <h3 id="basic_dependency">Automatic dependency</h3>

  <p>The build system has a built-in dependency scanner, which works out the
  dependency relationship between source files, so that they can be built in
  the correct order. The system scans all source files of known types for all
  supported dependency patterns. Dependencies of source files in a sub-package
  are written in a cache, which can be retrieved for incremental builds. (In an
  incremental build, only changed source files need to be re-scanned for
  dependency information. Dependency information for other files are retrieved
  from the cache.) The dependency information is passed to the
  <code>make</code> rule generator, which writes the <samp>Makefile</samp>.</p>

  <p>The <code>make</code> rule generator generates different <code>make</code>
  rules for different dependency types. The following dependency patterns are
  automatically detected by the current system:</p>

  <ul>
    <li>The <code>USE &lt;module&gt;</code> statement in a Fortran source file
    is the first pattern. The statement has two implications: 1) The current
    file compiles only if the module has been successfully compiled, and needs
    to be re-compiled if the module has changed. 2) The executable depending on
    the current file can only resolve all its externals by linking with the
    object file of the compiled module. The executable needs to be re-linked if
    the module and its dependencies has changed.</li>

    <li>The <code>INCLUDE '&lt;name&gt;.interface'</code> statement in a
    Fortran source file is the second pattern. (The default extension for an
    interface file is <samp>.interface</samp>. This can be modified through the
    input and output file type register, which will be discussed in a later
    section on <a href="#advanced_file-type">File Type</a>.) It has two
    implications: 1) The current file compiles only if the included interface
    file is in the INCLUDE search path, and needs to be re-compiled if the
    interface file changes. 2) The executable depending on the current file can
    only resolve all its externals by linking with the object file of the
    source file that generates the interface file. The executable needs to be
    re-linked if the source file (and its dependencies) associated with the
    interface file has changed. It is worth noting that for this dependency to
    work, the root &lt;name&gt; of the interface file should match with that of
    the source file associated with the interface file. (Please note that you
    can use pre-processor [#include "&lt;name&gt;.interface] instead of Fortran
    INCLUDE, but it will not work if you switch on the <a href=
    "#advanced_pp">pre-processing</a> stage, which will be discussed in a later
    section.)</li>

    <li>The <code>INCLUDE '&lt;file&gt;'</code> statement (excluding the
    INCLUDE interface file statement) in a Fortran source file is the third
    pattern. It has two implications: 1) The current file compiles only if the
    included file is in the INCLUDE search path, and needs to be re-compiled if
    the include file changes. 2) The executable needs to be linked with any
    objects the include file is dependent on. It needs to be re-linked if these
    objects have changed.</li>

    <li>The <code>#include '&lt;file&gt;'</code> statement in a Fortran/C
    source or header file is the fourth pattern. It has similar implications as
    the Fortran INCLUDE statement. However, they have to be handled differently
    because <code>#include</code> statements are processed by the
    pre-processor, which may be performed in a separate stage of the FCM build
    process. This will be further discussed in a later sub-section on <a href=
    "#advanced_pp">Pre-processing</a>.</li>
  </ul>

  <p>If you want your code to be built automatically by the FCM build system,
  you should also design your code to conform to the following rules:</p>

  <ol>
    <li>Single compilable program unit, (i.e. program, subroutine, function or
    module), per file.</li>

    <li>Unique name for each compilable program unit.</li>

    <li>Always supply an interface for subroutines and functions, i.e.:

      <ul>
        <li>Put them in modules.</li>

        <li>Put them in the CONTAINS section within the main program unit.</li>

        <li>Use interface files.</li>
      </ul>
    </li>

    <li>If interface files are used, it is good practice to name each source
    file after the program unit it contains. It will make life a lot simpler
    when using the <a href="#basic_interface">Automatic Fortran 9X interface
    block</a> feature, which has already been discussed in the previous
    section.

      <ul>
        <li>The problem is that, by default, the root name of the interface
        file is the same as that of the source file rather than the program
        unit. If they differ then the build system will create a dependency on
        the wrong object file (since the object files are named according to
        the program unit).</li>

        <li>This problem can be avoided by changing the behaviour of the
        interface file generator to use the name of the program unit instead
        (using a <code>TOOL::INTERFACE</code> declaration).</li>
      </ul>
    </li>
  </ol>

  <dl>
    <dt>Note - setting build targets</dt>

    <dd>
      <p>The <samp>Makefile</samp> generated by the build system contains a
      list of targets that can be built. The build system allows you to build
      (or perform the actions of) any targets that are present in the generated
      <samp>Makefile</samp>. There are two ways to specify the targets to be
      built.</p>

      <p>Firstly, you can use the <code>TARGET</code> declarations in your
      build configuration file to specify the default targets to be built.
      These targets will be set as dependencies of the <samp>all</samp> target
      in the generated <samp>Makefile</samp>, which is the default target to be
      built when <code>make</code> is invoked by FCM. It is worth noting that
      <code>TARGET</code> declarations are cumulative. A later declaration does
      not override an earlier one - it simply adds more targets to the
      list.</p>

      <p>Alternatively, you can use the <code>-t</code> option when you invoke
      the <code>fcm build</code> command. The option takes an argument, which
      should be a colon <code>:</code> separated list of targets to be built.
      When the <code>-t</code> option is set, FCM invokes <code>make</code> to
      build these targets instead. (E.g. if we invoke the build system with the
      command <code>fcm build -t foo.exe:bar.exe</code>, it will invoke
      <code>make</code> to build <samp>foo.exe</samp> and
      <samp>bar.exe</samp>.)</p>

      <p>If you do not specify any explicit targets, the system will search
      your source tree for main programs:</p>

      <ul>
        <li>If there are main programs in your source tree, they will be set as
        the default targets automatically.</li>

        <li>Otherwise, the default is to build the top level library archive
        containing objects compiled from the source files in the current source
        tree. (For more information on building library archives, please see
        the section on <a href="#advanced_library">Creating library
        archives</a>.)</li>
      </ul>
    </dd>
  </dl>

  <h2 id="advanced">Advanced Features</h2>

  <h3 id="advanced_dependency">Further dependency features</h3>

  <p>Apart from the usual dependency patterns described in the previous
  sub-section, the automatic dependency scanner also recognises two special
  directives when they are inserted into a source file:</p>

  <ul>
    <li>The directive <code>DEPENDS ON: &lt;object&gt;</code> in a comment line
    of a Fortran/C source file: It states that the current file is dependent on
    the declared external object. The executable depending on the current file
    needs to link with this external object in order to resolve all its
    external references. It needs to be re-linked if the declared external
    object (and its dependencies) has changed.</li>

    <li>The directive <code>CALLS: &lt;executable&gt;</code> in a comment line
    of a script: It states that the current script is dependent on the declared
    executable file, which can be another script or a binary executable. The
    current script can only function correctly if the declared executable is
    found in the search path. This directive is useful to ensure that all
    dependent executables are built or copied to the correct path.</li>
  </ul>

  <p>Another way to specify external dependency is to use the
  <code>EXE_DEP</code> declaration to declare extra dependencies. The
  declaration normally applies to all main programs, but if the form
  <code>EXE_DEP::&lt;target&gt;</code> is used, it will only apply to
  &lt;target&gt;, (which must be the name of a main program target). If the
  declaration is made without a value, the main programs will be set to depend
  on all object files. Otherwise, the value can be supplied as a space
  delimited list of items. Each item can be either the name of a sub-package or
  an object target. For the former, the main programs will be set to depend on
  all object files within the sub-package. For the latter, the main programs
  will be set to depend on the object target. The following are some
  examples:</p>
  <pre id="example_5">
# Example 5
# ----------------------------------------------------------------------
cfg::type          ext
cfg::version       1.0

bld::exe_dep::foo.exe  foo/bar egg.o # line 4
bld::exe_dep                         # line 5
# ... some other declarations ...
</pre>

  <p>Here is an explanation of what each line does:</p>

  <ul>
    <li><dfn>line 4</dfn>: this line declares the dependency on the sub-package
    <samp>foo/bar</samp> and the object target <samp>egg.o</samp> for building
    the main program target <samp>foo.exe</samp>. The target
    <samp>foo.exe</samp> will now depends on all object files in the
    <samp>foo/bar</samp> sub-package as well as the object target
    <samp>egg.o</samp>.</li>

    <li><dfn>line 5</dfn>: this line declares that all other main program
    targets will depend on all (non-program) object files in the build.</li>
  </ul>

  <dl>
    <dt>Note - naming of object files</dt>

    <dd>
      <p>By default, object files are named with the suffix <samp>.o</samp>.
      For a Fortran source file, the build system uses the lower case name of
      the first program unit within the file to name its object file. For
      example, if the first program unit in the Fortran source file
      <samp>foo.f90</samp> is <code>PROGRAM Bar</code>, the object file will be
      <samp>bar.o</samp>. For a C source file, the build system uses the lower
      case root name of the source file to name its object file. For example, a
      C source file called <samp>egg.c</samp> will have its object file named
      <samp>egg.o</samp>.</p>

      <p>The reason for using lower case to name the object files is because
      Fortran is a case insensitive language. Its symbols can either be in
      lower or upper case. E.g. the <code>SUBROUTINE Foo</code> is the same as
      the <code>SUBROUTINE foo</code>. It can be rather confusing if the
      subroutines are stored in different files. When they are compiled and
      archived into a library, there will be a clash of namespace, as the
      Fortran compiler thinks they are the same. However, this type of error
      does not normally get reported. If <samp>Foo</samp> and <samp>foo</samp>
      are very different code, the user may end up using the wrong subroutine,
      which may lead to a very long debugging session. By naming all object
      files in lower case, this type of situation can be avoided. If there is a
      clash in names due to the use of upper/lower cases, it will be reported
      as warnings by the build system, (as <em>duplicated targets</em> for
      building <samp>foo.o</samp>).</p>
    </dd>
  </dl>

  <p>It is realised that there are situations when an automatically detected
  dependency should not be written into the <samp>Makefile</samp>. For example,
  the dependency may be a standard module provided by the Fortran compiler, and
  does not need to be built in the usual way. In such case, we need to have a
  way to exclude this module during an automatic dependency scan.</p>

  <p>The <code>EXCL_DEP</code> declaration can be used to do just that. The
  following extract configuration contains some examples of the basic usage of
  the <code>EXCL_DEP</code> declaration:</p>
  <pre id="example_6">
# Example 6
# ----------------------------------------------------------------------
cfg::type          ext
cfg::version       1.0

bld::excl_dep  USE::YourFortranMod             # line 4
bld::excl_dep  INTERFACE::HerFortran.interface # line 5
bld::excl_dep  INC::HisFortranInc.inc          # line 6
bld::excl_dep  H::TheirHeader.h                # line 7
bld::excl_dep  OBJ::ItsObject.o                # line 8

# ... some other declarations ...
</pre>

  <p>Here is an explanation of what each line does:</p>

  <ul>
    <li><dfn>line 4</dfn>: this line declares that the Fortran module
    <samp>YourFortranMod</samp> should be excluded. The value of each
    <code>EXCL_DEP</code> declaration has two parts. The first part is a label
    that is used to define the type of dependency to be excluded. For a full
    list of these labels, please see the <a href=
    "annex_bld_cfg.html#dependency-types">dependency types table</a> in the
    <a href="annex_bld_cfg.html">Annex: Declarations in FCM build configuration
    file</a>. The label <code>USE</code> denotes a Fortran module. The second
    part of the label is the dependency itself. For instance, if a Fortran
    source file contains the line: <code>USE YourFortranMod</code>, the
    dependency scanner will ignore it.</li>

    <li><dfn>line 5</dfn>: this line declares that the include statement for
    the Fortran 9X interface file <samp>HerFortran.interface</samp> should be
    excluded. The label <code>INTERFACE</code> denotes a Fortran INCLUDE
    statement for a Fortran 9X interface block file. For example, if a Fortran
    source file contains the line: <code>INCLUDE 'HerFortran.interface'</code>,
    the dependency scanner will ignore it.</li>

    <li><dfn>line 6</dfn>: this line declares that the include statement for
    <samp>HisFortranInc.inc</samp> should be excluded. The label
    <code>INC</code> denotes a Fortran INCLUDE statement other than an INCLUDE
    statement for an interface block file. For example, if a Fortran source
    file contains the line: <code>INCLUDE 'HisFortranInc.inc'</code>, the
    dependency scanner will ignore it.</li>

    <li><dfn>line 7</dfn>: this line declares that the header include statement
    <samp>TheirHeader.h</samp> should be excluded. The label <code>H</code>
    denotes a pre-processing #include statement. For example, if a source file
    contains the line: <code>#include 'TheirHeader.h'</code>, the dependency
    scanner will ignore it.</li>

    <li><dfn>line 8</dfn>: this line declares that the external dependency for
    <samp>ItsObject.o</samp> should be excluded. The label <code>OBJ</code>
    denotes a compiled binary object. These dependencies are normally inserted
    into the source files as special comments. For example, if a source file
    contains the line: <code>! depends on: ItsObject.o</code>, the dependency
    scanner will ignore it.</li>
  </ul>

  <p>An <code>EXCL_DEP</code> declaration normally applies to all files in the
  build. However, you can suffix it with the name of a sub-package, i.e.
  <code>EXCL_DEP::&lt;pcks&gt;</code>. In such case, the declaration will only
  apply while scanning for dependencies in the source files in the sub-package
  named &lt;pcks&gt;.</p>

  <p>You can also exclude all dependency scan of a particular type. To do so,
  simply declare the type in the value. For example, if you do not want the
  build system to scan for the <code>CALLS: &lt;executable&gt;</code> directive
  in the comment lines of your scripts, you can make the following
  declaration:</p>
  <pre>
bld::excl_dep  EXE
</pre>

  <p>The opposite of the <code>EXCL_DEP</code> declaration is the
  <code>DEP::&lt;pcks&gt;</code> declaration, which you can use to add a
  dependency to a source file (in the package name <code>&lt;pcks&gt;</code>).
  The syntax of the declaration is similar to that of <code>EXCL_DEP</code>,
  but you must specify the package name of a source file for DEP declarations.
  Please also note that a <code>DEP</code> declaration only works if the
  particular dependency is supported for the particular source file - as it
  makes no sense, for example, to specify a USE dependency for a shell
  script.</p>

  <p>If you need to switch off dependency checking completely, you can use the
  <code>NO_DEP</code> declaration. For example, to switch off dependency
  checking for all but the <samp>foo/bar</samp> sub-package, you can do:</p>
  <pre>
bld::no_dep           true
bld::no_dep::foo/bar  false
</pre>

  <h3 id="advanced_blockdata">Linking a Fortran executable with a BLOCKDATA
  program unit</h3>

  <p>If it is required to link Fortran executables with BLOCKDATA program
  units, you must declare the executable targets and the objects containing the
  BLOCKDATA program units using the <code>BLOCKDATA::&lt;target&gt;</code>
  declarations. For example, if <samp>foo.exe</samp> is an executable target
  depending on the objects of the BLOCKDATA program units
  <samp>blkdata.o</samp> and <samp>fbk.o</samp>, you will make the following
  declarations:</p>
  <pre>
bld::blockdata::foo.exe  blkdata fbk
</pre>

  <p>If all your executables are dependent on <samp>blkdata.o</samp> and
  <samp>fbk.o</samp>, you will make the following declarations:</p>
  <pre>
bld::blockdata  blkdata fbk
</pre>

  <h3 id="advanced_library">Creating library archives</h3>

  <p>If you are interested in building library archives, the build system
  allows you to do it in a relatively simple way. For each sub-package in the
  source tree, there is a target to build a library containing all the objects
  compiled from the source files (that are not main programs) within the
  sub-package. If the sub-package contains children sub-packages, the object
  files of the children will also be included recursively. By default, the
  library archive is named after the sub-package, in the format
  <code>lib&lt;pcks&gt;.a</code>. (For example, the library archive for the
  package <samp>foo/bar/egg</samp> will be named
  <samp>libfoo__bar__egg.a</samp> by default.) If you do not like the default
  name for the sub-package library, you can use the
  <code>LIB::&lt;pcks&gt;</code> declaration to rename it, as long as the new
  name does not clash with other targets. For example, to rename
  <samp>libfoo__bar__egg.a</samp> to <samp>libham.a</samp>, you will make the
  following declaration in your extract configuration file:</p>
  <pre>
bld::lib::foo/bar/egg  ham
</pre>

  <p>In addition to sub-package libraries, you can also build a global library
  archive for the whole source tree. By default, the library is named
  <samp>libfcm_default.a</samp>, but you can rename it using the
  <code>LIB</code> declaration as above. For example, to rename the library to
  <samp>libmy-lib.a</samp>, you will make the following declaration in your
  extract configuration file:</p>
  <pre>
bld::lib  my-lib
</pre>

  <p>When a library archive is created successfully, the build system will
  automatically generate the relevant exclude dependency configurations in the
  <samp>etc/</samp> sub-directory of the build root. You will be able to
  include these configurations in subsequent builds that utilise the library.
  The root names of the configuration files match those of the library archives
  that you can create in the current build, but the extension <samp>*.a</samp>
  is replaced with <samp>*.cfg</samp>. For example, the exclude dependency
  configuration for <samp>libmy-lib.a</samp> is <samp>libmy-lib.cfg</samp>.</p>

  <h3 id="advanced_pp">Pre-processing</h3>

  <p>As most modern compilers can handle pre-processing, the build system
  leaves pre-processing to the compiler by default. However, it is recognised
  that there are code written with pre-processor directives that can alter the
  argument list of procedures and/or their dependencies. If a source file
  requires pre-processing in such a way, we have to pre-process before running
  the interface block generator and the dependency scanner. The <code>PP</code>
  declaration can be used to switch on this pre-processing stage. The
  pre-processing stage can be switched on globally or for individual
  sub-packages only. The following is an example, using an extract
  configuration file:</p>
  <pre id="example_7">
# Example 7
# ----------------------------------------------------------------------
cfg::type          ext
cfg::version       1.0

bld::pp::gen       true                  # line 4
bld::pp::var/foo   true                  # line 5

bld::tool::cppkeys GOOD WEATHER FORECAST # line 7
bld::tool::fppkeys FOO BAR EGG HAM       # line 8

# ... some other declarations ...
</pre>

  <p>Here is an explanation of what each line does:</p>

  <ul>
    <li><dfn>line 4-5</dfn>: these switches on the pre-processing stage for all
    sub-packages under <samp>gen</samp> and <samp>var/foo</samp>.</li>

    <li><dfn>line 7</dfn>: this declares a list of pre-defined macros
    <samp>GOOD</samp>, <samp>WEATHER</samp> and <samp>FORECAST</samp> for
    pre-processing all C files.</li>

    <li><dfn>line 8</dfn>: this declares a list of pre-defined macros
    <samp>FOO</samp>, <samp>BAR</samp>, <samp>EGG</samp> and <samp>HAM</samp>
    for pre-processing all Fortran files that require processing.</li>
  </ul>

  <p>Source files requiring pre-processing may contain <code>#include</code>
  statements to include header files. For including a local file, its name
  should be embedded within a pair of quotes, i.e. <samp>'file.h'</samp> or
  <samp>"file.h"</samp>. If the header file is embedded within a pair of
  &lt;file.h&gt; angle brackets, the system will assume that the file can be
  found in a standard location.</p>

  <p>The build system allows header files to be placed anywhere within the
  declared source tree. The system uses the dependency scanner, as described in
  the previous sub-section to scan for any header file dependencies. All source
  files requiring pre-processing and all header files are scanned. Header files
  that are required are copied to the <samp>inc/</samp> subdirectory of the
  build root, which is automatically added to the pre-processor search path via
  the <code>-I&lt;dir&gt;</code> option. The build system uses an internal
  logic similar to <code>make</code> to perform pre-processing. Header files
  are only copied to the <samp>inc/</samp> sub-directory if they are used in
  <code>#include</code> statements.</p>

  <p>Unlike <code>make</code>, which only uses the timestamp to determine
  whether an item is out of date, the internal logic of the build system does
  this by inspecting the content of the file as well. In an incremental build,
  the pre-processed file is only updated if its content has changed. This
  avoids unnecessary updates (and hence unnecessary re-compilation) in an
  incremental build if the changed section of the code does not affect the
  output file.</p>

  <p>Pre-processed code generated during the pre-processing stage are sent to
  the <samp>ppsrc/</samp> sub-directory of the build root. It will have a
  relative path that reflects the name of the declared sub-package. The
  pre-processed source file will have the same root name as the original source
  file. For C files, the same extension <samp>.c</samp> will be used. For
  Fortran files, the case of the extension will normally be dropped, e.g. from
  <samp>.F90</samp> to <samp>.f90</samp>.</p>

  <p>Following pre-processing, the system will use the pre-processed source
  file as if it is the original source file. The interface generator will
  generate the interface file using the pre-processed file, the dependency
  scanner will scan the pre-processed file for dependencies, and the compiler
  will compile the pre-processed source.</p>

  <p>The <code>TOOL::CPPKEYS</code> and <code>TOOL::FPPKEYS</code> declarations
  are used to pre-define macros in the C and Fortran pre-processor
  respectively. This is implemented by the build system using the pre-processor
  <code>-D</code> option on each word in the list. The use of these
  declarations are not confined to the pre-process stage. If any source files
  requiring pre-processing are left to the compiler, the declarations will be
  used to set up the commands for compiling these source files.</p>

  <p>The <code>TOOL::CPPKEYS</code> and <code>TOOL::FPPKEYS</code> declarations
  normally applies globally, but like any other <code>TOOL</code> declarations,
  they can be suffixed with sub-package names. In such cases, the declarations
  will apply only to the specified sub-packages.</p>

  <dl>
    <dt>Note - changing pre-processor flags</dt>

    <dd>
      <p>As for compiler flags, the build system detects changes in
      pre-processor flags (<code>TOOL::CPPFLAGS</code> and
      <code>TOOL::FPPFLAGS</code>) and macro definitions
      (<code>TOOL::CPPKEYS</code> and <code>TOOL::FPPKEYS</code>). If the
      pre-processor flags or the macro definitions have changed in an
      incremental build, the system will re-do all the necessary
      pre-processing. The following hierarchy is followed:</p>

      <ul>
        <li>If the pre-processor flags or macro definitions for a particular
        source file change, only that source file will be pre-processed
        again.</li>

        <li>If the pre-processor flags or macro definitions for a particular
        container package change, only source files within that container will
        be pre-processed again.</li>

        <li>If the global pre-processor flags or macro definitions change, all
        source files will be pre-processed again.</li>

        <li>If the pre-processor command changes, all source files are
        pre-processed again.</li>
      </ul>
    </dd>
  </dl>

  <h3 id="advanced_file-type">File type</h3>

  <p>The build system only knows what to do with an input source file if it
  knows what type of file it is. The type of a source file is normally
  determined automatically using one of the following three methods (in
  order):</p>

  <ol>
    <li>If the file is named with an extension, its extension will be matched
    against a set of registered file extensions. If a match is found, the file
    type will be set according to the register.</li>

    <li>If a file does not have an extension or does not match with a
    registered extension, its name is compared with a set of pre-defined
    patterns. If a match is found, the file type will be set according to the
    file type associated with the pattern.</li>

    <li>If the above two methods failed and if the file is a text file, the
    system will attempt to read the first line of the file. If the first line
    begins with a <code>#!</code> pattern, the line will be compared with a set
    of pre-defined patterns. If a match is found, the file type will be set
    according to the file type associated with the pattern.</li>
  </ol>

  <p>In addition to the above, if a file is a Fortran or C source file, the
  system will attempt to open the source file to determine whether it contains
  a main program, module (Fortran only) or just standalone procedures. All
  these information will be used later by the build system to process the
  source file.</p>

  <p>The build system registers a file type with a set of type flags delimited
  by the double colons <code>::</code>. For example, a Fortran 9X source file
  is registered as <code>FORTRAN::FORTRAN9X::SOURCE</code>. (Please note that
  the order of the type flags in the list is insignificant. For example,
  <code>FORTRAN::SOURCE</code> is the same as <code>SOURCE::FORTRAN</code>.)
  For a list of all the type flags used by the build system, please see the
  <a href="annex_bld_cfg.html#infile-ext-types">input file extension type flags
  table</a> in the <a href="annex_bld_cfg.html">Annex: Declarations in FCM
  build configuration file</a>.</p>

  <p>The following is a list of default input file extensions and their
  associated types:</p>

  <dl>
    <dt><samp>.f .for .ftn .f77</samp></dt>

    <dd><code>FORTRAN::SOURCE</code> Fortran 77 source file (assumed to be
    fixed format)</dd>

    <dt><samp>.f90 .f95</samp></dt>

    <dd><code>FORTRAN::FORTRAN9X::SOURCE</code> Fortran 9X source file (assumed
    to be free format)</dd>

    <dt><samp>.F .FOR .FTN .F77</samp></dt>

    <dd><code>FPP::SOURCE</code> Fortran 77 source file (assumed to be fixed
    format) that requires pre-processing</dd>

    <dt><samp>.F90 .F95</samp></dt>

    <dd><code>FPP::FPP9X::SOURCE</code> Fortran 9X source file (assumed to be
    free format) that requires pre-processing</dd>

    <dt><samp>.c</samp></dt>

    <dd><code>C::SOURCE</code> C source file</dd>

    <dt><samp>.h .h90</samp></dt>

    <dd><code>CPP::INCLUDE</code> Pre-processor <code>#include</code> header
    file</dd>

    <dt><samp>.o .obj</samp></dt>

    <dd><code>BINARY::OBJ</code> Compiled binary object</dd>

    <dt><samp>.exe</samp></dt>

    <dd><code>BINARY::EXE</code> Binary executable</dd>

    <dt><samp>.a</samp></dt>

    <dd><code>BINARY::LIB</code> Binary object library archive</dd>

    <dt><samp>.sh .ksh .bash .csh</samp></dt>

    <dd><code>SHELL::SCRIPT</code> Unix shell script</dd>

    <dt><samp>.pl .pm</samp></dt>

    <dd><code>PERL::SCRIPT</code> Perl script</dd>

    <dt><samp>.py</samp></dt>

    <dd><code>PYTHON::SCRIPT</code> Python script</dd>

    <dt><samp>.tcl</samp></dt>

    <dd><code>TCL::SCRIPT</code> Tcl/Tk script</dd>

    <dt><samp>.pro</samp></dt>

    <dd><code>PVWAVE::SCRIPT</code> IDL/PVWave program</dd>

    <dt><samp>.cfg</samp></dt>

    <dd><code>CFGFILE</code> FCM configuration file</dd>

    <dt><samp>.inc</samp></dt>

    <dd><code>FORTRAN::FORTRAN9X::INCLUDE</code> Fortran INCLUDE file</dd>

    <dt><samp>.interface</samp></dt>

    <dd><code>FORTRAN::FORTRAN9X::INCLUDE::INTERFACE</code> Fortran 9X INCLUDE
    interface block file</dd>
  </dl>

  <p>N.B. The extension must be unique. For example, the system does not
  support the use of <samp>.inc</samp> files for both <code>#include</code> and
  Fortran <code>INCLUDE</code>.</p>

  <p>The following is a list of supported file name patterns and their
  associated types:</p>

  <dl>
    <dt><samp>*Scr_* *Comp_* *IF_* *Suite_* *Interface_*</samp></dt>

    <dd><code>SHELL::SCRIPT</code> Unix shell script, GEN-based project naming
    conventions</dd>

    <dt><samp>*List_*</samp></dt>

    <dd><code>SHELL::SCRIPT::GENLIST</code> Unix shell script, GEN list
    file</dd>

    <dt><samp>*Sql_*</samp></dt>

    <dd><code>SCRIPT::SQL</code> SQL script, GEN-based project naming
    conventions</dd>
  </dl>

  <p>The following is a list of supported <code>#!</code> line patterns and
  their associated types:</p>

  <dl>
    <dt><samp>*sh* *ksh* *bash* *csh*</samp></dt>

    <dd><code>SHELL::SCRIPT</code> Unix shell script</dd>

    <dt><samp>*perl*</samp></dt>

    <dd><code>PERL::SCRIPT</code> Perl script</dd>

    <dt><samp>*python*</samp></dt>

    <dd><code>PYTHON::SCRIPT</code> Python script</dd>

    <dt><samp>*tclsh* *wish*</samp></dt>

    <dd><code>TCL::SCRIPT</code> Tcl/Tk script</dd>
  </dl>

  <p>The build system allows you to add or modify the register for input file
  extensions and their associated type using the
  <code>INFILE_EXT::&lt;ext&gt;</code> declaration, where &lt;ext&gt; is a file
  name extension without the leading dot. If file extension alone is
  insufficient for defining the type of your source file, you can use the
  <code>SRC_TYPE::&lt;pcks&gt;</code> declaration, (where &lt;pcks&gt; is the
  package name of the source file). For example, in an extract configuration
  file, you may have:</p>
  <pre id="example_8">
# Example 8
# ----------------------------------------------------------------------
cfg::type                ext
cfg::version             1.0

bld::infile_ext::foo     CPP::INCLUDE                # line 4
bld::infile_ext::bar     FORTRAN::FORTRAN9X::INCLUDE # line 5
bld::src_type::egg/ham.f FORTRAN::FORTRAN9X::INCLUDE # line 6

# ... some other declarations ...
</pre>

  <p>Here is an explanation of what each line does:</p>

  <ul>
    <li><dfn>line 4</dfn>: this line registers the extension <samp>.foo</samp>
    to be of type <code>CPP::INCLUDE</code>. This means that any input files
    with <samp>.foo</samp> extension will be treated as if they are
    pre-processor header files.</li>

    <li><dfn>line 5</dfn>: this line registers the extension <samp>.bar</samp>
    to be of type <code>FORTRAN::FORTRAN9X::INCLUDE</code>. This means that any
    input file with <samp>.bar</samp> extension will be treated as if they are
    Fortran 9X INCLUDE files.</li>

    <li><dfn>line 6</dfn>: this line declares the type for the source file in
    the package <samp>egg::ham.f</samp> to be
    <code>FORTRAN::FORTRAN9X::INCLUDE</code>. Without this declaration, this
    file would normally be given the type <code>FORTRAN::SOURCE</code>.</li>
  </ul>

  <p>The <code>INFILE_EXT</code> declarations deal with extensions of input
  files. There is also a <code>OUTFILE_EXT::&lt;type&gt;</code> declaration
  that deals with extensions of output files. The declaration is opposite that
  of <code>INFILE_EXT</code>. The file &lt;type&gt; is now declared with the
  label, and the extension is declared as the value. It is worth noting that
  <code>OUTFILE_EXT</code> declarations use very different syntax for
  &lt;type&gt;, and the declared extension must include the leading dot. For a
  list of output types used by the build system, please see the <a href=
  "annex_bld_cfg.html#outfile-ext-types">output file extension types table</a>
  in the <a href="annex_bld_cfg.html">Annex: Declarations in FCM build
  configuration file</a>. An example is given below:</p>
  <pre id="example_9">
# Example 9
# ----------------------------------------------------------------------
cfg::type                   ext
cfg::version                1.0

bld::outfile_ext::mod       .MOD   # line 4
bld::outfile_ext::interface .intfb # line 5

# ... some other declarations ...
</pre>

  <p>Here is an explanation of what each line does:</p>

  <ul>
    <li><dfn>line 4</dfn>: this line modifies the extension of compiled Fortran
    9X module information files from the default <samp>.mod</samp> to
    <samp>.MOD</samp>.</li>

    <li><dfn>line 5</dfn>: this line modifies the extension of INCLUDE Fortran
    9X interface block files from the default <samp>.interface</samp> to
    <samp>.intfb</samp>.</li>
  </ul>

  <p>N.B. If you have made changes to the file type registers, whether it is
  for input files or output files, it is always worth re-building your code in
  full-build mode to avoid unexpected behaviour.</p>

  <h3 id="advanced_inherit">Inherit from a previous build</h3>

  <p>As you can inherit from previous extracts, you can inherit from previous
  builds. The very same <code>USE</code> statement can be used to declare a
  build, which the current build will depend on. The only difference is that
  the declared location must contain a valid build configuration file. In fact,
  if you use the extract system to obtain your build configuration file, any
  <code>USE</code> declarations in the extract configuration file will also be
  <code>USE</code> declarations in the output build configuration file.</p>

  <p>By declaring a previous build with a <code>USE</code> statement, the
  current build automatically inherits settings from it. The following points
  are worth noting:</p>

  <ul>
    <li>Build targets are not normally inherited. However, you can switch on
    inheritance of build targets using an <code>INHERIT::TARGET</code>
    declaration, such as:
      <pre>
inherit::target  true
</pre>
    </li>

    <li>The build root directory and its sub-directories of the inherited build
    are placed into the search paths. For example, if we have an inherited
    build at <samp>/path/to/inherited</samp>, and it is used by a build at
    <samp>/path/to/my_build</samp>, the search path of executable files will
    become <samp>/path/to/my_build/bin:/path/to/inherited/bin</samp>, so that
    the <samp>bin/</samp> sub-directory of the current build is searched before
    the <samp>bin/</samp> sub-directory of the inherited build. If two or more
    <code>USE</code> statements are declared, the <code>USE</code> statement
    declared last will have higher priority. For example, if the current build
    is <var>C</var>, and it USEs build <var>A</var> before build <var>B</var>,
    the search path will be <samp>C:B:A</samp>.</li>

    <li>Source files are inherited by default. If a source file is declared in
    the current build that has the same package name as a source file of the
    inherited build, it will override that in the inherited build. Any source
    files missing from the current build will be taken from the inherited
    build.

      <p>You can switch off inheritance of source files using an
      <code>INHERIT::SRC</code> declaration. This declaration can be suffixed
      with the name of a sub-package. In such case, the declaration applies
      only to the inheritance of the sub-package. Otherwise, it applies
      globally. For example:</p>
      <pre>
# Switch off inheritance of source files in the <samp>gen</samp> sub-package
inherit::src::gen  false
</pre>
    </li>

    <li><code>BLOCKDATA</code>, <code>DEP</code>, <code>EXCL_DEP</code>,
    <code>EXE_DEP</code>, <code>INFILE_EXT</code>, <code>LIB</code>,
    <code>OUTFILE_EXT</code>, <code>PP</code>, <code>TOOL</code> and
    <code>SRC_TYPE</code> declarations are automatically inherited. If the same
    setting is declared in the current incremental build, it overrides the
    inherited declaration.</li>
  </ul>

  <p>As an example, suppose we have already performed an extract and build
  based on the configuration in <a href="#example_2">example 2</a>, we can set
  up an extract configuration file as follows:</p>
  <pre id="example_10">
# Example 10
# ----------------------------------------------------------------------
cfg::type            ext
cfg::version         1.0

use                  $HOME/example               # line 4

dest                 $HOME/example10             # line 6

bld::inherit::target true                        # line 8 
bld::target          ham.exe egg.exe             # line 9

bld::tool::fflags    -O2 -w                      # line 11
bld::tool::cflags    -O2                         # line 12

# ... and other declarations for repositories and source directories ...
</pre>

  <p>Here is an explanation of what each line does:</p>

  <ul>
    <li><dfn>line 4</dfn>: this line declares a previous extract at
    <samp>$HOME/example</samp> which the current extract will inherit from. The
    same line will be written to the output build configuration file. The
    subsequent build will then inherit from the build at
    <samp>$HOME/example</samp>.</li>

    <li><dfn>line 6</dfn>: this declares the destination root directory of the
    current extract, which will become the root directory of the current build.
    Search paths of the build sub-directories will be set automatically. For
    example, the search path for executable files created by the current build
    will be <samp>$HOME/example10/bin:$HOME/example/bin</samp>.</li>

    <li><dfn>line 8</dfn>: this line switches on inheritance of build targets.
    The build targets in <a href="#example_1">example 1</a>, i.e.
    <samp>foo.exe</samp> and <samp>bar.exe</samp> will be built as part of the
    current build.</li>

    <li><dfn>line 9</dfn>: this declares two new build targets
    <samp>ham.exe</samp> and <samp>egg.exe</samp> to be added to the inherited
    ones. The default build targets of the current build will now be
    <samp>foo.exe</samp>, <samp>bar.exe</samp>, <samp>ham.exe</samp> and
    <samp>egg.exe</samp>.</li>

    <li><dfn>line 11-12</dfn>: these lines modify options used by the Fortran
    and the C compilers, overriding those inherited from <a href=
    "#example_1">example 1</a>.</li>
  </ul>

  <dl>
    <dt>Build inheritance limitation: handling of include files</dt>

    <dd>
      <p>The build system uses the compiler/pre-processor's <code>-I</code>
      option to specify the search path for include files. For example, it uses
      the option to specify the <samp>inc/</samp> sub-directories of the
      current build and its inherited build.</p>

      <p>However, some compilers/pre-processors (e.g. <code>cpp</code>) search
      for include files from the container directory of the source file before
      searching for the paths specified by the <code>-I</code> options. This
      behaviour may cause the build to behave incorrectly.</p>

      <p>Consider a source file <samp>egg/hen.c</samp> that includes
      <samp>fried.h</samp>. If the directory structure looks like:</p>
      <pre>
# Sources in inherited build:
egg/hen.c
egg/fried.h

# Sources in current build:
egg/fried.h
</pre>

      <p>The system will correctly identify that <samp>fried.h</samp> is out of
      date, and trigger a re-compilation of <samp>egg/hen.c</samp>. However, if
      the compiler searches for the include files from the container directory
      of the source file first, it will wrongly use the include file in the
      inherited build instead of the current one.</p>

      <p>Some compilers (e.g. <code>gfortran</code>) do not behave this way and
      others (e.g. <code>ifort</code>) have options to prevent include file
      search in the container directory of the source file. If you are using
      such a compiler you can avoid the problem for Fortran compilation
      although this does not fix the problem entirely if you have switched on
      the pre-processing stage. Otherwise you may have to work around the
      problem, (e.g. by making a comment change in the source file, or by not
      using an inherited build at all).</p>
    </dd>
  </dl>

  <h3 id="advanced_data">Building data files</h3>

  <p>While the usual targets to be built are the executables associated with
  source files containing main programs, libraries or scripts, the build system
  also allows you to build <em>data</em> files. All files with no registered
  type are considered to be <em>data</em> files. For each container
  sub-package, there is an automatic target for copying all <em>data</em> files
  to the <samp>etc/</samp> sub-directory of the build root. The name of the
  target has the form <code>&lt;pcks&gt;.etc</code>, where &lt;pcks&gt; is the
  name of the sub-package (with package names delimited by the double
  underscore <code>__</code>). For example, the target name for sub-package
  <samp>foo/bar</samp> is <samp>foo__bar.etc</samp>. This target is
  particularly useful for copying, say, all namelists in a sub-package to the
  <samp>etc/</samp> sub-directory of the build root.</p>

  <p>At the end of a successful build, if the <samp>etc/</samp> sub-directory
  is not empty, the <samp>fcm_env.sh</samp> script will export the environment
  variable <var>FCM_ETCDIR</var> to point to the <samp>etc/</samp>
  sub-directory. You should be able to use this environment variable to locate
  your data files.</p>

  <h2 id="verbose">Diagnostic verbose level</h2>

  <p>The amount of diagnostic messages generated by the build system is
  normally set to a level suitable for normal everyday operation. This is the
  default diagnostic verbose level 1. If you want a minimum amount of
  diagnostic messages, you should set the verbose level to 0. If you want more
  diagnostic messages, you can set the verbose level to 2 or 3. You can modify
  the verbose level in two ways. The first way is to set the environment
  variable <var>FCM_VERBOSE</var> to the desired verbose level. The second way
  is to invoke the build system with the <code>-v &lt;level&gt;</code> option.
  (If set, the command line option overrides the environment variable.)</p>

  <p>The following is a list of diagnostic output at each verbose level:</p>

  <dl>
    <dt>Level 0</dt>

    <dd>
      <ul>
        <li>Report the time taken at the end of each stage of the build
        process.</li>

        <li>Run the <code>make</code> command in silent mode.</li>
      </ul>
    </dd>

    <dt>Level 1</dt>

    <dd>
      <ul>
        <li>Everything at verbose level 0.</li>

        <li>Report the name of the build configuration file.</li>

        <li>Report the location of the build destination.</li>

        <li>Report date/time at the beginning of each stage of the build
        process.</li>

        <li>Report removed directories.</li>

        <li>Report number of pre-processed files.</li>

        <li>Report number of generated F9X interface files.</li>

        <li>Report number of source files scanned for dependencies.</li>

        <li>Report name of updated <samp>Makefile</samp>.</li>

        <li>Print compiler/linker commands.</li>

        <li>Report total time.</li>
      </ul>
    </dd>

    <dt>Level 2</dt>

    <dd>
      <ul>
        <li>Everything at verbose level 1.</li>

        <li>For incremental build in archive mode, report the commands used to
        extract the archives.</li>

        <li>Report creation and removal of directories.</li>

        <li>Report pre-processor commands.</li>

        <li>Print compiler/linker commands with timestamps.</li>
      </ul>
    </dd>

    <dt>Level 3</dt>

    <dd>
      <ul>
        <li>Everything at verbose level 2.</li>

        <li>Report update of dummy files.</li>

        <li>Report all shell commands.</li>

        <li>Report pre-processor commands with timestamps.</li>

        <li>Report any F9X interface files generated.</li>

        <li>Report number of lines and number of automatic dependencies for
        each source file which is scanned.</li>

        <li>Run <code>make</code> on normal mode (as opposed to silent
        mode).</li>

        <li>Report start date/time and time taken of <code>make</code>
        commands.</li>
      </ul>
    </dd>
  </dl>

  <h2 id="overview">Overview of the build process</h2>

  <p>The FCM build process can be summarised in five stages. Here is a summary
  of what is done in each stage:</p>

  <ol>
    <li><dfn>Parse configuration and setup destination</dfn>: in this
    pre-requisite stage, the build system parses the configuration file. The
    <samp>src/</samp> sub-directory is searched recursively for source files.
    For full builds, it ensures that the sub-directories and files created by
    the build system are removed. If you invoke <code>fcm build</code> with a
    <code>--clean</code> option, the system will not go any further.</li>

    <li><dfn>Setup build</dfn>: in this first stage, the system determines
    whether any settings have changed by using the cache. If so, the cache is
    updated with the current settings.</li>

    <li><dfn>Pre-process</dfn>: if any files in any source files require
    pre-processing, they will be pre-processed at this stage. The resulting
    pre-processed source files will be sent to the <samp>ppsrc/</samp>
    sub-directory of the build root.</li>

    <li><dfn>Generate dependency</dfn>: the system scans source files of
    registered types for dependency information. For an incremental build, the
    information is only updated if a source file is changed. The system then
    uses the information to write a <samp>Makefile</samp> for the main
    build.</li>

    <li><dfn>Generate interface</dfn>: if there are Fortran 9X source files
    with standalone subroutines and functions, the build system generates
    interface blocks for them. The result of which will be written to the
    interface files in the <samp>inc/</samp> sub-directory of the build
    root.</li>

    <li>
      <dfn>Make</dfn>: the system invokes <code>make</code> on the
      <samp>Makefile</samp> generated in the previous stage to perform the main
      build. Following a build, the <em>root</em> directory of the build may
      contain the following sub-directories (empty ones are removed
      automatically at the end of the build process):

      <dl>
        <dt><samp>.cache/.bld/</samp></dt>

        <dd>Cache files, used internally by FCM.</dd>

        <dt><samp>bin/</samp></dt>

        <dd>Executable binaries and scripts.</dd>

        <dt><samp>cfg/</samp></dt>

        <dd>Configuration files.</dd>

        <dt><samp>done/</samp></dt>

        <dd>Dummy <em>done</em> files used internally by the
        <samp>Makefile</samp> generated by FCM.</dd>

        <dt><samp>etc/</samp></dt>

        <dd>Miscellaneous data files.</dd>

        <dt><samp>flags/</samp></dt>

        <dd>Dummy <em>flags</em> files used internally by the
        <samp>Makefile</samp> generated by FCM.</dd>

        <dt><samp>inc/</samp></dt>

        <dd>Include files, such as <samp>*.h</samp>, <samp>*.inc</samp>,
        <samp>*.interface</samp>, and <samp>*.mod</samp>.</dd>

        <dt><samp>lib/</samp></dt>

        <dd>Object library archives.</dd>

        <dt><samp>obj/</samp></dt>

        <dd>Compiled object files.</dd>

        <dt><samp>ppsrc/</samp></dt>

        <dd>Source directories with pre-processed files.</dd>

        <dt><samp>src/</samp></dt>

        <dd>Source directories. This directory is not changed by the build
        system.</dd>

        <dt><samp>tmp/</samp></dt>

        <dd>Temporary objects and binaries. Files generated by the
        compiler/linker may be left here.</dd>
      </dl>
    </li>
  </ol>

  </div>
  </div>
  </div>

  <hr/>
  <div class="container-fluid text-center">
    <div class="row"><div class="col-md-12">
    <address><small>
      &copy; British Crown Copyright 2006-16
      <a href="http://www.metoffice.gov.uk">Met Office</a>.
      See <a href="../etc/fcm-terms-of-use.html">Terms of Use</a>.<br />
      This document is released under the British <a href=
      "http://www.nationalarchives.gov.uk/doc/open-government-licence/" rel=
      "license">Open Government Licence</a>.<br />
    </small></address>
    </div></div>
  </div>

  <script type="text/javascript" src="../etc/jquery.min.js"></script>
  <script type="text/javascript" src="../etc/bootstrap/js/bootstrap.min.js"></script>
  <script type="text/javascript" src="../etc/fcm.js"></script>
  <script type="text/javascript" src="../etc/fcm-version.js"></script>
</body>
</html>
fcm 2016.12.0-1 / usr / share / doc / fcm / user_guide / build.html
