CTest 2.6.0 Docs: Difference between revisions

From KitwarePublic
Jump to navigationJump to search
(New page: <pre> ctest version 2.6-patch 0 ------------------------------------------------------------------------------ Name ctest - Testing driver provided by CMake. --------------------------...)
 
(Replaced content with "https://cmake.org/cmake/help/ctest2.6docs.html")
Line 1: Line 1:
<pre>
https://cmake.org/cmake/help/ctest2.6docs.html
ctest version 2.6-patch 0
------------------------------------------------------------------------------
Name
 
  ctest - Testing driver provided by CMake.
 
------------------------------------------------------------------------------
Usage
 
  ctest [options]
 
------------------------------------------------------------------------------
Description
 
The "ctest" executable is the CMake test driver program.  CMake-generated
build trees created for projects that use the ENABLE_TESTING and ADD_TEST
commands have testing support.  This program will run the tests and report
results.
 
------------------------------------------------------------------------------
Options
 
  -C <cfg>, --build-config <cfg>
      Choose configuration to test.
 
      Some CMake-generated build trees can have multiple build
      configurations in the same tree.  This option can be used to specify
      which one should be tested.  Example configurations are "Debug" and
      "Release".
 
  -V,--verbose
      Enable verbose output from tests.
 
      Test output is normally suppressed and only summary information is
      displayed.  This option will show all test output.
 
  -VV,--extra-verbose
      Enable more verbose output from tests.
 
      Test output is normally suppressed and only summary information is
      displayed.  This option will show even more test output.
 
  --debug
      Displaying more verbose internals of CTest.
 
      This feature will result in large number of output that is mostly
      useful for debugging dashboard problems.
 
  -Q,--quiet
      Make ctest quiet.
 
      This option will suppress all the output.  The output log file will
      still be generated if the --output-log is specified.  Options such as
      --verbose, --extra-verbose, and --debug are ignored if --quiet is
      specified.
 
  -O <file>, --output-log <file>
      Output to log file
 
      This option tells ctest to write all its output to a log file.
 
  -N,--show-only
      Disable actual execution of tests.
 
      This option tells ctest to list the tests that would be run but not
      actually run them.  Useful in conjunction with the -R and -E options.
 
  -R <regex>, --tests-regex <regex>
      Run tests matching regular expression.
 
      This option tells ctest to run only the tests whose names match the
      given regular expression.
 
  -E <regex>, --exclude-regex <regex>
      Exclude tests matching regular expression.
 
      This option tells ctest to NOT run the tests whose names match the
      given regular expression.
 
  -D <dashboard>, --dashboard <dashboard>
      Execute dashboard test
 
      This option tells ctest to perform act as a Dart client and perform a
      dashboard test.  All tests are <Mode><Test>, where Mode can be
      Experimental, Nightly, and Continuous, and Test can be Start, Update,
      Configure, Build, Test, Coverage, and Submit.
 
  -M <model>, --test-model <model>
      Sets the model for a dashboard
 
      This option tells ctest to act as a Dart client where the TestModel
      can be Experimental, Nightly, and Continuous.  Combining -M and -T is
      similar to -D
 
  -T <action>, --test-action <action>
      Sets the dashboard action to perform
 
      This option tells ctest to act as a Dart client and perform some
      action such as start, build, test etc.  Combining -M and -T is similar
      to -D
 
  --track <track>
      Specify the track to submit dashboard to
 
      Submit dashboard to specified track instead of default one.  By
      default, the dashboard is submitted to Nightly, Experimental, or
      Continuous track, but by specifying this option, the track can be
      arbitrary.
 
  -S <script>, --script <script>
      Execute a dashboard for a configuration
 
      This option tells ctest to load in a configuration script which sets a
      number of parameters such as the binary and source directories.  Then
      ctest will do what is required to create and run a dashboard.  This
      option basically sets up a dashboard and then runs ctest -D with the
      appropriate options.
 
  -SP <script>, --script-new-process <script>
      Execute a dashboard for a configuration
 
      This option does the same operations as -S but it will do them in a
      seperate process.  This is primarily useful in cases where the script
      may modify the environment and you do not want the modified enviroment
      to impact other -S scripts.
 
  -A <file>, --add-notes <file>
      Add a notes file with submission
 
      This option tells ctest to include a notes file when submitting
      dashboard.
 
  -I [Start,End,Stride,test#,test#|Test file], --tests-information
      Run a specific number of tests by number.
 
      This option causes ctest to run tests starting at number Start, ending
      at number End, and incrementing by Stride.  Any additional numbers
      after Stride are considered individual test numbers.  Start, End,or
      stride can be empty.  Optionally a file can be given that contains the
      same syntax as the command line.
 
  -U, --union
      Take the Union of -I and -R
 
      When both -R and -I are specified by default the intersection of tests
      are run.  By specifying -U the union of tests is run instead.
 
  --interactive-debug-mode [0|1]
      Set the interactive mode to 0 or 1.
 
      This option causes ctest to run tests in either an interactive mode or
      a non-interactive mode.  On Windows this means that in non-interactive
      mode, all system debug pop up windows are blocked.  In dashboard mode
      (Experimental, Nightly, Continuous), the default is non-interactive.
      When just running tests not for a dashboard the default is to allow
      popups and interactive debugging.
 
  --build-and-test
      Configure, build and run a test.
 
      This option tells ctest to configure (i.e.  run cmake on), build, and
      or execute a test.  The configure and test steps are optional.  The
      arguments to this command line are the source and binary directories.
      By default this will run CMake on the Source/Bin directories specified
      unless --build-nocmake is specified.  Both --build-makeprogram and
      --build-generator MUST be provided to use --built-and-test.  If
      --test-command is specified then that will be run after the build is
      complete.  Other options that affect this mode are --build-target
      --build-nocmake, --build-run-dir, --build-two-config, --build-exe-dir,
      --build-project,--build-noclean, --build-options
 
  --build-target
      Specify a specific target to build.
 
      This option goes with the --build-and-test option, if left out the all
      target is built.
 
  --build-nocmake
      Run the build without running cmake first.
 
      Skip the cmake step.
 
  --build-run-dir
      Specify directory to run programs from.
 
      Directory where programs will be after it has been compiled.
 
  --build-two-config
      Run CMake twice
 
  --build-exe-dir
      Specify the directory for the executable.
 
  --build-generator
      Specify the generator to use.
 
  --build-project
      Specify the name of the project to build.
 
  --build-makeprogram
      Specify the make program to use.
 
  --build-noclean
      Skip the make clean step.
 
  --build-config-sample
      A sample executable to use to determine the configuraiton
 
      A sample executable to use to determine the configuraiton that should
      be used.  e.g.  Debug/Release/etc
 
  --build-options
      Add extra options to the build step.
 
      This option must be the last option with the exception of
      --test-command
 
  --test-command
      The test to run with the --build-and-test option.
 
  --test-timeout
      The time limit in seconds, internal use only.
 
  --tomorrow-tag
      Nightly or experimental starts with next day tag.
 
      This is useful if the build will not finish in one day.
 
  --ctest-config
      The configuration file used to initialize CTest state when submitting
      dashboards.
 
      This option tells CTest to use different initialization file instead
      of CTestConfiguration.tcl.  This way multiple initialization files can
      be used for example to submit to multiple dashboards.
 
  --overwrite
      Overwrite CTest configuration option.
 
      By default ctest uses configuration options from configuration file.
      This option will overwrite the configuration option.
 
  --extra-submit <file>[;<file>]
      Submit extra files to the dashboard.
 
      This option will submit extra files to the dashboard.
 
  --force-new-ctest-process
      Run child CTest instances as new processes
 
      By default CTest will run child CTest instances within the same
      process.  If this behavior is not desired, this argument will enforce
      new processes for child CTest processes.
 
  --submit-index
      Submit individual dashboard tests with specific index
 
      This option allows performing the same CTest action (such as test)
      multiple times and submit all stages to the same dashboard (Dart2
      required).  Each execution requires different index.
 
------------------------------------------------------------------------------
Generators
 
The following generators are available on this platform:
 
------------------------------------------------------------------------------
Commands
 
  break
      Break from an enclosing foreach or while loop.
 
        break()
 
      Breaks from an enclosing foreach loop or while loop
 
  build_name
      Deprecated.  Use ${CMAKE_SYSTEM} and ${CMAKE_CXX_COMPILER} instead.
 
        build_name(variable)
 
      Sets the specified variable to a string representing the platform and
      compiler settings.  These values are now available through the
      CMAKE_SYSTEM and CMAKE_CXX_COMPILER variables.
 
  cmake_minimum_required
      Set the minimum required version of cmake for a project.
 
        cmake_minimum_required(VERSION major[.minor[.patch]]
                                [FATAL_ERROR])
 
      If the current version of CMake is lower than that required it will
      stop processing the project and report an error.  When a version
      higher than 2.4 is specified the command implicitly invokes
 
        cmake_policy(VERSION major[.minor[.patch]])
 
      which sets the cmake policy version level to the version specified.
      When version 2.4 or lower is given the command implicitly invokes
 
        cmake_policy(VERSION 2.4)
 
      which enables compatibility features for CMake 2.4 and lower.
 
      The FATAL_ERROR option is accepted but ignored.  It is left from CMake
      versions 2.4 and lower in which failure to meet the minimum version
      was a warning by default.
 
  cmake_policy
      Manage CMake Policy settings.
 
      As CMake evolves it is sometimes necessary to change existing behavior
      in order to fix bugs or improve implementations of existing features.
      The CMake Policy mechanism is designed to help keep existing projects
      building as new versions of CMake introduce changes in behavior.  Each
      new policy (behavioral change) is given an identifier of the form
      "CMP<NNNN>" where "<NNNN>" is an integer index.  Documentation
      associated with each policy describes the OLD and NEW behavior and the
      reason the policy was introduced.  Projects may set each policy to
      select the desired behavior.  When CMake needs to know which behavior
      to use it checks for a setting specified by the project.  If no
      setting is available the OLD behavior is assumed and a warning is
      produced requesting that the policy be set.
 
      The cmake_policy command is used to set policies to OLD or NEW
      behavior.  While setting policies individually is supported, we
      encourage projects to set policies based on CMake versions.
 
        cmake_policy(VERSION major.minor[.patch])
 
      Specify that the current CMake list file is written for the given
      version of CMake.  All policies introduced in the specified version or
      earlier will be set to use NEW behavior.  All policies introduced
      after the specified version will be reset to use OLD behavior with a
      warning.  This effectively requests behavior preferred as of a given
      CMake version and tells newer CMake versions to warn about their new
      policies.  The policy version specified must be at least 2.4 or the
      command will report an error.  In order to get compatibility features
      supporting versions earlier than 2.4 see documentation of policy
      CMP0001.
 
        cmake_policy(SET CMP<NNNN> NEW)
        cmake_policy(SET CMP<NNNN> OLD)
 
      Tell CMake to use the OLD or NEW behavior for a given policy.
      Projects depending on the old behavior of a given policy may silence a
      policy warning by setting the policy state to OLD.  Alternatively one
      may fix the project to work with the new behavior and set the policy
      state to NEW.
 
        cmake_policy(PUSH)
        cmake_policy(POP)
 
      Push and pop the current policy setting state on a stack.  Each PUSH
      must have a matching POP.  This is useful when mixing multiple
      projects, subprojects, and files included from external projects that
      may each have been written for a different version of CMake.  Each
      subdirectory entered by the project automatically pushes a new level
      on the stack to isolate the subdirectories from their parents.
 
  configure_file
      Copy a file to another location and modify its contents.
 
        configure_file(InputFile OutputFile
                        [COPYONLY] [ESCAPE_QUOTES] [@ONLY])
 
      The Input and Output files have to have full paths.  This command
      replaces any variables in the input file referenced as ${VAR} or @VAR@
      with their values as determined by CMake.  If a variable is not
      defined, it will be replaced with nothing.  If COPYONLY is specified,
      then no variable expansion will take place.  If ESCAPE_QUOTES is
      specified then any substituted quotes will be C-style escaped.  The
      file will be configured with the current values of CMake variables.
      If @ONLY is specified, only variables of the form @VAR@ will be
      replaces and ${VAR} will be ignored.  This is useful for configuring
      scripts that use ${VAR}.  Any occurrences of #cmakedefine VAR will be
      replaced with either #define VAR or /* #undef VAR */ depending on the
      setting of VAR in CMake
 
  CTEST_BUILD
      Builds the repository.
 
        CTEST_BUILD([BUILD build_dir] [RETURN_VALUE res])
 
      Builds the given build directory and stores results in Build.xml.
 
  CTEST_CONFIGURE
      Configures the repository.
 
        CTEST_CONFIGURE(BUILD build_dir RETURN_VALUE res)
 
      Configures the given build directory and stores results in
      Configure.xml.  The second argument is a variable that will hold
      return value.
 
  CTEST_COVERAGE
      Tests the repository.
 
        CTEST_COVERAGE([BUILD build_dir] [RETURN_VALUE res])
 
      Perform the coverage of the given build directory and stores results
      in Coverage.xml.  The second argument is a variable that will hold
      value.
 
  CTEST_EMPTY_BINARY_DIRECTORY
      empties the binary directory
 
        CTEST_EMPTY_BINARY_DIRECTORY( directory )
 
      Removes a binary directory.  This command will perform some checks
      prior to deleting the directory in an attempt to avoid malicious or
      accidental directory deletion.
 
  CTEST_MEMCHECK
      Tests the repository.
 
        CTEST_MEMCHECK([BUILD build_dir] [RETURN_VALUE res])
 
      Performs a memory checking of tests in the given build directory and
      stores results in MemCheck.xml.  The second argument is a variable
      that will hold value.
 
  CTEST_READ_CUSTOM_FILES
      read CTestCustom files.
 
        CTEST_READ_CUSTOM_FILES( directory ... )
 
      Read all the CTestCustom.ctest or CTestCustom.cmake files from the
      given directory.
 
  CTEST_RUN_SCRIPT
      runs a ctest -S script
 
        CTEST_RUN_SCRIPT([NEW_PROCESS] script_file_name script_file_name1
                    script_file_name2 ...)
 
      Runs a script or scripts much like if it was run from ctest -S.  If no
      argument is provided then the current script is run using the current
      settings of the variables.  If NEW_PROCESS is specified then each
      script will be run in a seperate process.
 
  CTEST_SLEEP
      sleeps for some amount of time
 
        CTEST_SLEEP( seconds )
        CTEST_SLEEP( time1 duration time2 )
 
      With one argument it will sleep for a given number of seconds.  With
      three arguments it will wait for time2 - time1 - duration seconds.
 
  CTEST_START
      Starts the testing for a given model
 
        CTEST_START(Model [TRACK <track>] [source [binary]])
 
      Starts the testing for a given model.  The command should be called
      after the binary directory is initialized.  If the 'source' and
      'binary' directory are not specified, it reads the
      CTEST_SOURCE_DIRECTORY and CTEST_BINARY_DIRECTORY.  If the track is
      specified, the submissions will go to the specified track.
 
  CTEST_SUBMIT
      Submits the repository.
 
        CTEST_SUBMIT([RETURN_VALUE res])
 
      Submits the test results for the project.
 
  CTEST_TEST
      Tests the repository.
 
        CTEST_TEST([BUILD build_dir] [RETURN_VALUE res])
 
      Tests the given build directory and stores results in Test.xml.  The
      second argument is a variable that will hold value.
 
  CTEST_UPDATE
      Updates the repository.
 
        CTEST_UPDATE([SOURCE source] [RETURN_VALUE res])
 
      Updates the given source directory and stores results in Update.xml.
      The second argument is a variable that will hold the number of files
      modified.  If there is a problem, the variable will be -1.
 
  else
      Starts the else portion of an if block.
 
        else(expression)
 
      See the if command.
 
  elseif
      Starts the elseif portion of an if block.
 
        elseif(expression)
 
      See the if command.
 
  endforeach
      Ends a list of commands in a FOREACH block.
 
        endforeach(expression)
 
      See the FOREACH command.
 
  endfunction
      Ends a list of commands in a function block.
 
        endfunction(expression)
 
      See the function command.
 
  endif
      Ends a list of commands in an if block.
 
        endif(expression)
 
      See the if command.
 
  endmacro
      Ends a list of commands in a macro block.
 
        endmacro(expression)
 
      See the macro command.
 
  endwhile
      Ends a list of commands in a while block.
 
        endwhile(expression)
 
      See the while command.
 
  exec_program
      Deprecated.  Use the execute_process() command instead.
 
      Run an executable program during the processing of the CMakeList.txt
      file.
 
        exec_program(Executable [directory in which to run]
                      [ARGS <arguments to executable>]
                      [OUTPUT_VARIABLE <var>]
                      [RETURN_VALUE <var>])
 
      The executable is run in the optionally specified directory.  The
      executable can include arguments if it is double quoted, but it is
      better to use the optional ARGS argument to specify arguments to the
      program.  This is because cmake will then be able to escape spaces in
      the executable path.  An optional argument OUTPUT_VARIABLE specifies a
      variable in which to store the output.  To capture the return value of
      the execution, provide a RETURN_VALUE.  If OUTPUT_VARIABLE is
      specified, then no output will go to the stdout/stderr of the console
      running cmake.
 
 
  execute_process
      Execute one or more child processes.
 
        execute_process(COMMAND <cmd1> [args1...]]
                        [COMMAND <cmd2> [args2...] [...]]
                        [WORKING_DIRECTORY <directory>]
                        [TIMEOUT <seconds>]
                        [RESULT_VARIABLE <variable>]
                        [OUTPUT_VARIABLE <variable>]
                        [ERROR_VARIABLE <variable>]
                        [INPUT_FILE <file>]
                        [OUTPUT_FILE <file>]
                        [ERROR_FILE <file>]
                        [OUTPUT_QUIET]
                        [ERROR_QUIET]
                        [OUTPUT_STRIP_TRAILING_WHITESPACE]
                        [ERROR_STRIP_TRAILING_WHITESPACE])
 
      Runs the given sequence of one or more commands with the standard
      output of each process piped to the standard input of the next.  A
      single standard error pipe is used for all processes.  If
      WORKING_DIRECTORY is given the named directory will be set as the
      current working directory of the child processes.  If TIMEOUT is given
      the child processes will be terminated if they do not finish in the
      specified number of seconds (fractions are allowed).  If
      RESULT_VARIABLE is given the variable will be set to contain the
      result of running the processes.  This will be an integer return code
      from the last child or a string describing an error condition.  If
      OUTPUT_VARIABLE or ERROR_VARIABLE are given the variable named will be
      set with the contents of the standard output and standard error pipes
      respectively.  If the same variable is named for both pipes their
      output will be merged in the order produced.  If INPUT_FILE,
      OUTPUT_FILE, or ERROR_FILE is given the file named will be attached to
      the standard input of the first process, standard output of the last
      process, or standard error of all processes respectively.  If
      OUTPUT_QUIET or ERROR_QUIET is given then the standard output or
      standard error results will be quietly ignored.  If more than one
      OUTPUT_* or ERROR_* option is given for the same pipe the precedence
      is not specified.  If no OUTPUT_* or ERROR_* options are given the
      output will be shared with the corresponding pipes of the CMake
      process itself.
 
      The execute_process command is a newer more powerful version of
      exec_program, but the old command has been kept for compatibility.
 
  file
      File manipulation command.
 
        file(WRITE filename "message to write"... )
        file(APPEND filename "message to write"... )
        file(READ filename variable [LIMIT numBytes] [OFFSET offset] [HEX])
        file(STRINGS filename variable [LIMIT_COUNT num]
              [LIMIT_INPUT numBytes] [LIMIT_OUTPUT numBytes]
              [LENGTH_MINIMUM numBytes] [LENGTH_MAXIMUM numBytes]
              [NEWLINE_CONSUME] [REGEX regex]
              [NO_HEX_CONVERSION])
        file(GLOB variable [RELATIVE path] [globbing expressions]...)
        file(GLOB_RECURSE variable [RELATIVE path]
              [globbing expressions]...)
        file(REMOVE [file1 ...])
        file(REMOVE_RECURSE [file1 ...])
        file(MAKE_DIRECTORY [directory1 directory2 ...])
        file(RELATIVE_PATH variable directory file)
        file(TO_CMAKE_PATH path result)
        file(TO_NATIVE_PATH path result)
        file(DOWNLOAD url file [TIMEOUT timeout] [STATUS status] [LOG log])
 
      WRITE will write a message into a file called 'filename'.  It
      overwrites the file if it already exists, and creates the file if it
      does not exist.
 
      APPEND will write a message into a file same as WRITE, except it will
      append it to the end of the file
 
      READ will read the content of a file and store it into the variable.
      It will start at the given offset and read up to numBytes.  If the
      argument HEX is given, the binary data will be converted to
      hexadecimal representation and this will be stored in the variable.
 
      STRINGS will parse a list of ASCII strings from a file and store it in
      a variable.  Binary data in the file are ignored.  Carriage return
      (CR) characters are ignored.  It works also for Intel Hex and Motorola
      S-record files, which are automatically converted to binary format
      when reading them.  Disable this using NO_HEX_CONVERSION.
 
      LIMIT_COUNT sets the maximum number of strings to return.  LIMIT_INPUT
      sets the maximum number of bytes to read from the input file.
      LIMIT_OUTPUT sets the maximum number of bytes to store in the output
      variable.  LENGTH_MINIMUM sets the minimum length of a string to
      return.  Shorter strings are ignored.  LENGTH_MAXIMUM sets the maximum
      length of a string to return.  Longer strings are split into strings
      no longer than the maximum length.  NEWLINE_CONSUME allows newlines to
      be included in strings instead of terminating them.
 
      REGEX specifies a regular expression that a string must match to be
      returned.  Typical usage
 
        file(STRINGS myfile.txt myfile)
 
      stores a list in the variable "myfile" in which each item is a line
      from the input file.
 
      GLOB will generate a list of all files that match the globbing
      expressions and store it into the variable.  Globbing expressions are
      similar to regular expressions, but much simpler.  If RELATIVE flag is
      specified for an expression, the results will be returned as a
      relative path to the given path.
 
      Examples of globbing expressions include:
 
          *.cxx      - match all files with extension cxx
          *.vt?      - match all files with extension vta,...,vtz
          f[3-5].txt - match files f3.txt, f4.txt, f5.txt
 
      GLOB_RECURSE will generate similar list as the regular GLOB, except it
      will traverse all the subdirectories of the matched directory and
      match the files.
 
      Examples of recursive globbing include:
 
          /dir/*.py  - match all python files in /dir and subdirectories
 
      MAKE_DIRECTORY will create the given directories, also if their parent
      directories don't exist yet
 
      REMOVE will remove the given files, also in subdirectories
 
      REMOVE_RECURSE will remove the given files and directories, also
      non-empty directories
 
      RELATIVE_PATH will determine relative path from directory to the given
      file.
 
      TO_CMAKE_PATH will convert path into a cmake style path with unix /.
      The input can be a single path or a system path like "$ENV{PATH}".
      Note the double quotes around the ENV call TO_CMAKE_PATH only takes
      one argument.
 
      TO_NATIVE_PATH works just like TO_CMAKE_PATH, but will convert from a
      cmake style path into the native path style \ for windows and / for
      UNIX.
 
      DOWNLOAD will download the givin URL to the given file.  If LOG var is
      specified a log of the download will be put in var.  If STATUS var is
      specified the status of the operation will be put in var.  The status
      is returned in a list of length 2.  The first element is the numeric
      return value for the operation, and the second element is a string
      value for the error.  A 0 numeric error means no error in the
      operation.  If TIMEOUT time is specified, the operation will timeout
      after time seconds, time can be specified as a float.
 
 
  find_file
      Find the full path to a file.
 
          find_path(<VAR> name1 [path1 path2 ...])
 
      This is the short-hand signature for the command that is sufficient in
      many cases.  It is the same as find_path(<VAR> name1 [PATHS path1
      path2 ...])
 
          find_path(
                    <VAR>
                    name | NAMES name1 [name2 ...]
                    [PATHS path1 [path2 ... ENV var]]
                    [PATH_SUFFIXES suffix1 [suffix2 ...]]
                    [DOC "cache documentation string"]
                    [NO_DEFAULT_PATH]
                    [NO_CMAKE_ENVIRONMENT_PATH]
                    [NO_CMAKE_PATH]
                    [NO_SYSTEM_ENVIRONMENT_PATH]
                    [NO_CMAKE_SYSTEM_PATH]
                    [CMAKE_FIND_ROOT_PATH_BOTH |
                    ONLY_CMAKE_FIND_ROOT_PATH |
                    NO_CMAKE_FIND_ROOT_PATH]
                  )
 
      This command is used to find a full path to named file.  A cache entry
      named by <VAR> is created to store the result of this command.  If the
      full path to a file is found the result is stored in the variable and
      the search will not be repeated unless the variable is cleared.  If
      nothing is found, the result will be <VAR>-NOTFOUND, and the search
      will be attempted again the next time find_path is invoked with the
      same variable.  The name of the full path to a file that is searched
      for is specified by the names listed after the NAMES argument.
      Additional search locations can be specified after the PATHS argument.
      If ENV var is found in the PATHS section the environment variable var
      will be read and converted from a system environment variable to a
      cmake style list of paths.  For example ENV PATH would be a way to
      list the system path variable.  The argument after DOC will be used
      for the documentation string in the cache.  PATH_SUFFIXES can be used
      to give sub directories that will be appended to the search paths.
 
      If NO_DEFAULT_PATH is specified, then no additional paths are added to
      the search.  If NO_DEFAULT_PATH is not specified, the search process
      is as follows:
 
      1.  Search cmake specific environment variables.  This can be skipped
      if NO_CMAKE_ENVIRONMENT_PATH is passed.
 
          <prefix>/include for each <prefix> in CMAKE_PREFIX_PATH
          CMAKE_INCLUDE_PATH
          CMAKE_FRAMEWORK_PATH
 
      2.  Search cmake variables with the same names as the cmake specific
      environment variables.  These are intended to be used on the command
      line with a -DVAR=value.  This can be skipped if NO_CMAKE_PATH is
      passed.
 
          <prefix>/include for each <prefix> in CMAKE_PREFIX_PATH
          CMAKE_INCLUDE_PATH
          CMAKE_FRAMEWORK_PATH
 
      3.  Search the standard system environment variables.  This can be
      skipped if NO_SYSTEM_ENVIRONMENT_PATH is an argument.
 
          PATH
          INCLUDE
 
      4.  Search cmake variables defined in the Platform files for the
      current system.  This can be skipped if NO_CMAKE_SYSTEM_PATH is
      passed.
 
          <prefix>/include for each <prefix> in CMAKE_SYSTEM_PREFIX_PATH
          CMAKE_SYSTEM_INCLUDE_PATH
          CMAKE_SYSTEM_FRAMEWORK_PATH
 
      5.  Search the paths specified after PATHS or in the short-hand
      version of the command.
 
      On Darwin or systems supporting OS X Frameworks, the cmake variable
      CMAKE_FIND_FRAMEWORK can be set to empty or one of the following:
 
          "FIRST"  - Try to find frameworks before standard
                    libraries or headers. This is the default on Darwin.
          "LAST"  - Try to find frameworks after standard
                    libraries or headers.
          "ONLY"  - Only try to find frameworks.
          "NEVER". - Never try to find frameworks.
 
      On Darwin or systems supporting OS X Application Bundles, the cmake
      variable CMAKE_FIND_APPBUNDLE can be set to empty or one of the
      following:
 
          "FIRST"  - Try to find application bundles before standard
                    programs. This is the default on Darwin.
          "LAST"  - Try to find application bundles after standard
                    programs.
          "ONLY"  - Only try to find application bundles.
          "NEVER". - Never try to find application bundles.
 
      The CMake variable CMAKE_FIND_ROOT_PATH specifies one or more
      directories to be prepended to all other search directories.  This
      effectively "re-roots" the entire search under given locations.  By
      default it is empty.  It is especially useful when cross-compiling to
      point to the root directory of the target environment and CMake will
      search there too.  By default at first the directories listed in
      CMAKE_FIND_ROOT_PATH and then the non-rooted directories will be
      searched.  The default behavior can be adjusted by setting
      CMAKE_FIND_ROOT_PATH_MODE_INCLUDE.  This behavior can be manually
      overridden on a per-call basis.  By using CMAKE_FIND_ROOT_PATH_BOTH
      the search order will be as described above.  If
      NO_CMAKE_FIND_ROOT_PATH is used then CMAKE_FIND_ROOT_PATH will not be
      used.  If ONLY_CMAKE_FIND_ROOT_PATH is used then only the re-rooted
      directories will be searched.
 
      The reason the paths listed in the call to the command are searched
      last is that most users of CMake would expect things to be found first
      in the locations specified by their environment.  Projects may
      override this behavior by simply calling the command twice:
 
          find_path(<VAR> NAMES name PATHS paths... NO_DEFAULT_PATH)
          find_path(<VAR> NAMES name)
 
      Once one of these calls succeeds the result variable will be set and
      stored in the cache so that neither call will search again.
 
  find_library
      Find a library.
 
          find_library(<VAR> name1 [path1 path2 ...])
 
      This is the short-hand signature for the command that is sufficient in
      many cases.  It is the same as find_library(<VAR> name1 [PATHS path1
      path2 ...])
 
          find_library(
                    <VAR>
                    name | NAMES name1 [name2 ...]
                    [PATHS path1 [path2 ... ENV var]]
                    [PATH_SUFFIXES suffix1 [suffix2 ...]]
                    [DOC "cache documentation string"]
                    [NO_DEFAULT_PATH]
                    [NO_CMAKE_ENVIRONMENT_PATH]
                    [NO_CMAKE_PATH]
                    [NO_SYSTEM_ENVIRONMENT_PATH]
                    [NO_CMAKE_SYSTEM_PATH]
                    [CMAKE_FIND_ROOT_PATH_BOTH |
                    ONLY_CMAKE_FIND_ROOT_PATH |
                    NO_CMAKE_FIND_ROOT_PATH]
                  )
 
      This command is used to find a library.  A cache entry named by <VAR>
      is created to store the result of this command.  If the library is
      found the result is stored in the variable and the search will not be
      repeated unless the variable is cleared.  If nothing is found, the
      result will be <VAR>-NOTFOUND, and the search will be attempted again
      the next time find_library is invoked with the same variable.  The
      name of the library that is searched for is specified by the names
      listed after the NAMES argument.  Additional search locations can be
      specified after the PATHS argument.  If ENV var is found in the PATHS
      section the environment variable var will be read and converted from a
      system environment variable to a cmake style list of paths.  For
      example ENV PATH would be a way to list the system path variable.  The
      argument after DOC will be used for the documentation string in the
      cache.  PATH_SUFFIXES can be used to give sub directories that will be
      appended to the search paths.
 
      If NO_DEFAULT_PATH is specified, then no additional paths are added to
      the search.  If NO_DEFAULT_PATH is not specified, the search process
      is as follows:
 
      1.  Search cmake specific environment variables.  This can be skipped
      if NO_CMAKE_ENVIRONMENT_PATH is passed.
 
          <prefix>/lib for each <prefix> in CMAKE_PREFIX_PATH
          CMAKE_LIBRARY_PATH
          CMAKE_FRAMEWORK_PATH
 
      2.  Search cmake variables with the same names as the cmake specific
      environment variables.  These are intended to be used on the command
      line with a -DVAR=value.  This can be skipped if NO_CMAKE_PATH is
      passed.
 
          <prefix>/lib for each <prefix> in CMAKE_PREFIX_PATH
          CMAKE_LIBRARY_PATH
          CMAKE_FRAMEWORK_PATH
 
      3.  Search the standard system environment variables.  This can be
      skipped if NO_SYSTEM_ENVIRONMENT_PATH is an argument.
 
          PATH
          LIB
 
      4.  Search cmake variables defined in the Platform files for the
      current system.  This can be skipped if NO_CMAKE_SYSTEM_PATH is
      passed.
 
          <prefix>/lib for each <prefix> in CMAKE_SYSTEM_PREFIX_PATH
          CMAKE_SYSTEM_LIBRARY_PATH
          CMAKE_SYSTEM_FRAMEWORK_PATH
 
      5.  Search the paths specified after PATHS or in the short-hand
      version of the command.
 
      On Darwin or systems supporting OS X Frameworks, the cmake variable
      CMAKE_FIND_FRAMEWORK can be set to empty or one of the following:
 
          "FIRST"  - Try to find frameworks before standard
                    libraries or headers. This is the default on Darwin.
          "LAST"  - Try to find frameworks after standard
                    libraries or headers.
          "ONLY"  - Only try to find frameworks.
          "NEVER". - Never try to find frameworks.
 
      On Darwin or systems supporting OS X Application Bundles, the cmake
      variable CMAKE_FIND_APPBUNDLE can be set to empty or one of the
      following:
 
          "FIRST"  - Try to find application bundles before standard
                    programs. This is the default on Darwin.
          "LAST"  - Try to find application bundles after standard
                    programs.
          "ONLY"  - Only try to find application bundles.
          "NEVER". - Never try to find application bundles.
 
      The CMake variable CMAKE_FIND_ROOT_PATH specifies one or more
      directories to be prepended to all other search directories.  This
      effectively "re-roots" the entire search under given locations.  By
      default it is empty.  It is especially useful when cross-compiling to
      point to the root directory of the target environment and CMake will
      search there too.  By default at first the directories listed in
      CMAKE_FIND_ROOT_PATH and then the non-rooted directories will be
      searched.  The default behavior can be adjusted by setting
      CMAKE_FIND_ROOT_PATH_MODE_LIBRARY.  This behavior can be manually
      overridden on a per-call basis.  By using CMAKE_FIND_ROOT_PATH_BOTH
      the search order will be as described above.  If
      NO_CMAKE_FIND_ROOT_PATH is used then CMAKE_FIND_ROOT_PATH will not be
      used.  If ONLY_CMAKE_FIND_ROOT_PATH is used then only the re-rooted
      directories will be searched.
 
      The reason the paths listed in the call to the command are searched
      last is that most users of CMake would expect things to be found first
      in the locations specified by their environment.  Projects may
      override this behavior by simply calling the command twice:
 
          find_library(<VAR> NAMES name PATHS paths... NO_DEFAULT_PATH)
          find_library(<VAR> NAMES name)
 
      Once one of these calls succeeds the result variable will be set and
      stored in the cache so that neither call will search again.
 
      If the library found is a framework, then VAR will be set to the full
      path to the framework <fullPath>/A.framework.  When a full path to a
      framework is used as a library, CMake will use a -framework A, and a
      -F<fullPath> to link the framework to the target.
 
  find_package
      Load settings for an external project.
 
        find_package(<package> [major[.minor[.patch]]] [EXACT] [QUIET]
                      [[REQUIRED|COMPONENTS] [components...]])
 
      Finds and loads settings from an external project.  <package>_FOUND
      will be set to indicate whether the package was found.  When the
      package is found package-specific information is provided through
      variables documented by the package itself.  The QUIET option disables
      messages if the package cannot be found.  The REQUIRED option stops
      processing with an error message if the package cannot be found.  A
      package-specific list of components may be listed after the REQUIRED
      option or after the COMPONENTS option if no REQUIRED option is given.
      The "[major[.minor[.patch]]]" version argument specifies a desired
      version with which the package found should be compatible.  The EXACT
      option requests that the version be matched exactly.  Version support
      is currently provided only on a package-by-package basis (details
      below).
 
      User code should generally look for packages using the above simple
      signature.  The remainder of this command documentation specifies the
      full command signature and details of the search process.  Project
      maintainers wishing to provide a package to be found by this command
      are encouraged to read on.
 
      The command has two modes by which it searches for packages: "Module"
      mode and "Config" mode.  Module mode is available when the command is
      invoked with the above reduced signature.  CMake searches for a file
      called "Find<package>.cmake" in the CMAKE_MODULE_PATH followed by the
      CMake installation.  If the file is found, it is read and processed by
      CMake.  It is responsible for finding the package, checking the
      version, and producing any needed messages.  Many find-modules provide
      limited or no support for versioning; check the module documentation.
      If no module is found the command proceeds to Config mode.
 
      The complete Config mode command signature is:
 
        find_package(<package> [major[.minor[.patch]]] [EXACT] [QUIET]
                      [[REQUIRED|COMPONENTS] [components...]] [NO_MODULE]
                      [NAMES name1 [name2 ...]]
                      [CONFIGS config1 [config2 ...]]
                      [PATHS path1 [path2 ... ]]
                      [PATH_SUFFIXES suffix1 [suffix2 ...]]
                      [NO_DEFAULT_PATH]
                      [NO_CMAKE_ENVIRONMENT_PATH]
                      [NO_CMAKE_PATH]
                      [NO_SYSTEM_ENVIRONMENT_PATH]
                      [NO_CMAKE_BUILDS_PATH]
                      [NO_CMAKE_SYSTEM_PATH]
                      [CMAKE_FIND_ROOT_PATH_BOTH |
                      ONLY_CMAKE_FIND_ROOT_PATH |
                      NO_CMAKE_FIND_ROOT_PATH])
 
      The NO_MODULE option may be used to skip Module mode explicitly.  It
      is also implied by use of options not specified in the reduced
      signature.
 
      Config mode attempts to locate a configuration file provided by the
      package to be found.  A cache entry called <package>_DIR is created to
      hold the directory containing the file.  By default the command
      searches for a package with the name <package>.  If the NAMES option
      is given the names following it are used instead of <package>.  The
      command searches for a file called "<name>Config.cmake" or
      "<lower-case-name>-config.cmake" for each name specified.  A
      replacement set of possible configuration file names may be given
      using the CONFIGS option.  The search procedure is specified below.
      Once found, the configuration file is read and processed by CMake.
      Since the file is provided by the package it already knows the
      location of package contents.  The full path to the configuration file
      is stored in the cmake variable <package>_CONFIG.
 
      If the package configuration file cannot be found CMake will generate
      an error describing the problem unless the QUIET argument is
      specified.  If REQUIRED is specified and the package is not found a
      fatal error is generated and the configure step stops executing.  If
      <package>_DIR has been set to a directory not containing a
      configuration file a fatal error is always generated because user
      intervention is required.
 
      When the "[major[.minor[.patch]]]" version argument is specified
      Config mode will only find a version of the package that claims
      compatibility with the requested version.  If the EXACT option is
      given only a version of the package claiming an exact match of the
      requested version may be found.  CMake does not establish any
      convention for the meaning of version numbers.  Package version
      numbers are checked by "version" files provided by the packages
      themselves.  For a candidate package confguration file
      "<config-file>.cmake" the corresponding version file is located next
      to it and named either "<config-file>-version.cmake" or
      "<config-file>Version.cmake".  If no such version file is available
      then the configuration file is assumed to not be compatible with any
      requested version.  When a version file is found it is loaded to check
      the requested version number.  The version file is loaded in a nested
      scope in which the following variables have been defined:
 
        PACKAGE_FIND_NAME          = the <package> name
        PACKAGE_FIND_VERSION      = full requested version string
        PACKAGE_FIND_VERSION_MAJOR = requested major version, if any
        PACKAGE_FIND_VERSION_MINOR = requested minor version, if any
        PACKAGE_FIND_VERSION_PATCH = requested patch version, if any
 
      The version file checks whether it satisfies the requested version and
      sets these variables:
 
        PACKAGE_VERSION            = package version (major[.minor[.patch]])
        PACKAGE_VERSION_EXACT      = true if version is exact match
        PACKAGE_VERSION_COMPATIBLE = true if version is compatible
 
      These variables are checked by the find_package command to determine
      whether the configuration file provides an acceptable version.  They
      are not available after the find_package call returns.  If the version
      is acceptable the following variables are set:
 
        <package>_VERSION      = package version (major[.minor[.patch]])
        <package>_VERSION_MAJOR = major from major[.minor[.patch]], if any
        <package>_VERSION_MINOR = minor from major[.minor[.patch]], if any
        <package>_VERSION_PATCH = patch from major[.minor[.patch]], if any
 
      and the corresponding package configuration file is loaded.  When
      multiple package configuration files are available whose version files
      claim compatibility with the version requested it is unspecified which
      one is chosen.  No attempt is made to choose a highest or closest
      version number.
 
      Config mode provides an elaborate interface and search procedure.
      Much of the interface is provided for completeness and for use
      internally by find-modules loaded by Module mode.  Most user code
      should simply call
 
        find_package(<package> [major[.minor]] [EXACT] [REQUIRED|QUIET])
 
      in order to find a package.  Package maintainers providing CMake
      package configuration files are encouraged to name and install them
      such that the procedure outlined below will find them without
      requiring use of additional options.
 
      CMake constructs a set of possible installation prefixes for the
      package.  Under each prefix several directories are searched for a
      configuration file.  The tables below show the directories searched.
      Each entry is meant for installation trees following Windows (W), UNIX
      (U), or Apple (A) conventions.
 
        <prefix>/                                              (W)
        <prefix>/(cmake|CMake)/                                (W)
        <prefix>/(share|lib)/<name>*/                          (U)
        <prefix>/(share|lib)/<name>*/(cmake|CMake)/            (U)
 
      On systems supporting OS X Frameworks and Application Bundles the
      following directories are searched for frameworks or bundles
      containing a configuration file:
 
        <prefix>/<name>.framework/Resources/                    (A)
        <prefix>/<name>.framework/Resources/CMake/              (A)
        <prefix>/<name>.framework/Versions/*/Resources/        (A)
        <prefix>/<name>.framework/Versions/*/Resources/CMake/  (A)
        <prefix>/<name>.app/Contents/Resources/                (A)
        <prefix>/<name>.app/Contents/Resources/CMake/          (A)
 
      In all cases the <name> is treated as case-insensitive and corresponds
      to any of the names specified (<package> or names given by NAMES).  If
      PATH_SUFFIXES is specified the suffixes are appended to each (W) or
      (U) directory entry one-by-one.
 
      This set of directories is intended to work in cooperation with
      projects that provide configuration files in their installation trees.
      Directories above marked with (W) are intended for installations on
      Windows where the prefix may point at the top of an application's
      installation directory.  Those marked with (U) are intended for
      installations on UNIX platforms where the prefix is shared by multiple
      packages.  This is merely a convention, so all (W) and (U) directories
      are still searched on all platforms.  Directories marked with (A) are
      intended for installations on Apple platforms.  The cmake variables
      CMAKE_FIND_FRAMEWORK and CMAKE_FIND_APPBUNDLE determine the order of
      preference as specified below.
 
      The set of installation prefixes is constructed using the following
      steps.  If NO_DEFAULT_PATH is specified steps 1-5 are skipped.
 
      1.  Search cmake specific environment variables.  This can be skipped
      if NO_CMAKE_ENVIRONMENT_PATH is passed.
 
          CMAKE_PREFIX_PATH
          CMAKE_FRAMEWORK_PATH
          CMAKE_APPBUNDLE_PATH
 
      2.  Search cmake variables with the same names as the cmake specific
      environment variables.  These are intended to be used on the command
      line with a -DVAR=value.  This can be skipped if NO_CMAKE_PATH is
      passed.
 
          CMAKE_PREFIX_PATH
          CMAKE_FRAMEWORK_PATH
          CMAKE_APPBUNDLE_PATH
 
      3.  Search the standard system environment variables.  This can be
      skipped if NO_SYSTEM_ENVIRONMENT_PATH is passed.  Path entries ending
      in "/bin" or "/sbin" are automatically converted to their parent
      directories.
 
          PATH
 
      4.  Search project build trees recently configured in a CMake GUI.
      This can be skipped if NO_CMAKE_BUILDS_PATH is passed.  It is intended
      for the case when a user is building multiple dependent projects one
      after another.
 
      5.  Search cmake variables defined in the Platform files for the
      current system.  This can be skipped if NO_CMAKE_SYSTEM_PATH is
      passed.
 
          CMAKE_SYSTEM_PREFIX_PATH
          CMAKE_SYSTEM_FRAMEWORK_PATH
          CMAKE_SYSTEM_APPBUNDLE_PATH
 
      6.  Search paths specified by the PATHS option.
 
      On Darwin or systems supporting OS X Frameworks, the cmake variable
      CMAKE_FIND_FRAMEWORK can be set to empty or one of the following:
 
          "FIRST"  - Try to find frameworks before standard
                    libraries or headers. This is the default on Darwin.
          "LAST"  - Try to find frameworks after standard
                    libraries or headers.
          "ONLY"  - Only try to find frameworks.
          "NEVER". - Never try to find frameworks.
 
      On Darwin or systems supporting OS X Application Bundles, the cmake
      variable CMAKE_FIND_APPBUNDLE can be set to empty or one of the
      following:
 
          "FIRST"  - Try to find application bundles before standard
                    programs. This is the default on Darwin.
          "LAST"  - Try to find application bundles after standard
                    programs.
          "ONLY"  - Only try to find application bundles.
          "NEVER". - Never try to find application bundles.
 
      The CMake variable CMAKE_FIND_ROOT_PATH specifies one or more
      directories to be prepended to all other search directories.  This
      effectively "re-roots" the entire search under given locations.  By
      default it is empty.  It is especially useful when cross-compiling to
      point to the root directory of the target environment and CMake will
      search there too.  By default at first the directories listed in
      CMAKE_FIND_ROOT_PATH and then the non-rooted directories will be
      searched.  The default behavior can be adjusted by setting
      CMAKE_FIND_ROOT_PATH_MODE_PACKAGE.  This behavior can be manually
      overridden on a per-call basis.  By using CMAKE_FIND_ROOT_PATH_BOTH
      the search order will be as described above.  If
      NO_CMAKE_FIND_ROOT_PATH is used then CMAKE_FIND_ROOT_PATH will not be
      used.  If ONLY_CMAKE_FIND_ROOT_PATH is used then only the re-rooted
      directories will be searched.
 
      The reason the paths listed in the call to the command are searched
      last is that most users of CMake would expect things to be found first
      in the locations specified by their environment.  Projects may
      override this behavior by simply calling the command twice:
 
          find_package(<package> PATHS paths... NO_DEFAULT_PATH)
          find_package(<package>)
 
      Once one of these calls succeeds the result variable will be set and
      stored in the cache so that neither call will search again.
 
  find_path
      Find the directory containing a file.
 
          find_path(<VAR> name1 [path1 path2 ...])
 
      This is the short-hand signature for the command that is sufficient in
      many cases.  It is the same as find_path(<VAR> name1 [PATHS path1
      path2 ...])
 
          find_path(
                    <VAR>
                    name | NAMES name1 [name2 ...]
                    [PATHS path1 [path2 ... ENV var]]
                    [PATH_SUFFIXES suffix1 [suffix2 ...]]
                    [DOC "cache documentation string"]
                    [NO_DEFAULT_PATH]
                    [NO_CMAKE_ENVIRONMENT_PATH]
                    [NO_CMAKE_PATH]
                    [NO_SYSTEM_ENVIRONMENT_PATH]
                    [NO_CMAKE_SYSTEM_PATH]
                    [CMAKE_FIND_ROOT_PATH_BOTH |
                    ONLY_CMAKE_FIND_ROOT_PATH |
                    NO_CMAKE_FIND_ROOT_PATH]
                  )
 
      This command is used to find a directory containing the named file.  A
      cache entry named by <VAR> is created to store the result of this
      command.  If the file in a directory is found the result is stored in
      the variable and the search will not be repeated unless the variable
      is cleared.  If nothing is found, the result will be <VAR>-NOTFOUND,
      and the search will be attempted again the next time find_path is
      invoked with the same variable.  The name of the file in a directory
      that is searched for is specified by the names listed after the NAMES
      argument.  Additional search locations can be specified after the
      PATHS argument.  If ENV var is found in the PATHS section the
      environment variable var will be read and converted from a system
      environment variable to a cmake style list of paths.  For example ENV
      PATH would be a way to list the system path variable.  The argument
      after DOC will be used for the documentation string in the cache.
      PATH_SUFFIXES can be used to give sub directories that will be
      appended to the search paths.
 
      If NO_DEFAULT_PATH is specified, then no additional paths are added to
      the search.  If NO_DEFAULT_PATH is not specified, the search process
      is as follows:
 
      1.  Search cmake specific environment variables.  This can be skipped
      if NO_CMAKE_ENVIRONMENT_PATH is passed.
 
          <prefix>/include for each <prefix> in CMAKE_PREFIX_PATH
          CMAKE_INCLUDE_PATH
          CMAKE_FRAMEWORK_PATH
 
      2.  Search cmake variables with the same names as the cmake specific
      environment variables.  These are intended to be used on the command
      line with a -DVAR=value.  This can be skipped if NO_CMAKE_PATH is
      passed.
 
          <prefix>/include for each <prefix> in CMAKE_PREFIX_PATH
          CMAKE_INCLUDE_PATH
          CMAKE_FRAMEWORK_PATH
 
      3.  Search the standard system environment variables.  This can be
      skipped if NO_SYSTEM_ENVIRONMENT_PATH is an argument.
 
          PATH
          INCLUDE
 
      4.  Search cmake variables defined in the Platform files for the
      current system.  This can be skipped if NO_CMAKE_SYSTEM_PATH is
      passed.
 
          <prefix>/include for each <prefix> in CMAKE_SYSTEM_PREFIX_PATH
          CMAKE_SYSTEM_INCLUDE_PATH
          CMAKE_SYSTEM_FRAMEWORK_PATH
 
      5.  Search the paths specified after PATHS or in the short-hand
      version of the command.
 
      On Darwin or systems supporting OS X Frameworks, the cmake variable
      CMAKE_FIND_FRAMEWORK can be set to empty or one of the following:
 
          "FIRST"  - Try to find frameworks before standard
                    libraries or headers. This is the default on Darwin.
          "LAST"  - Try to find frameworks after standard
                    libraries or headers.
          "ONLY"  - Only try to find frameworks.
          "NEVER". - Never try to find frameworks.
 
      On Darwin or systems supporting OS X Application Bundles, the cmake
      variable CMAKE_FIND_APPBUNDLE can be set to empty or one of the
      following:
 
          "FIRST"  - Try to find application bundles before standard
                    programs. This is the default on Darwin.
          "LAST"  - Try to find application bundles after standard
                    programs.
          "ONLY"  - Only try to find application bundles.
          "NEVER". - Never try to find application bundles.
 
      The CMake variable CMAKE_FIND_ROOT_PATH specifies one or more
      directories to be prepended to all other search directories.  This
      effectively "re-roots" the entire search under given locations.  By
      default it is empty.  It is especially useful when cross-compiling to
      point to the root directory of the target environment and CMake will
      search there too.  By default at first the directories listed in
      CMAKE_FIND_ROOT_PATH and then the non-rooted directories will be
      searched.  The default behavior can be adjusted by setting
      CMAKE_FIND_ROOT_PATH_MODE_INCLUDE.  This behavior can be manually
      overridden on a per-call basis.  By using CMAKE_FIND_ROOT_PATH_BOTH
      the search order will be as described above.  If
      NO_CMAKE_FIND_ROOT_PATH is used then CMAKE_FIND_ROOT_PATH will not be
      used.  If ONLY_CMAKE_FIND_ROOT_PATH is used then only the re-rooted
      directories will be searched.
 
      The reason the paths listed in the call to the command are searched
      last is that most users of CMake would expect things to be found first
      in the locations specified by their environment.  Projects may
      override this behavior by simply calling the command twice:
 
          find_path(<VAR> NAMES name PATHS paths... NO_DEFAULT_PATH)
          find_path(<VAR> NAMES name)
 
      Once one of these calls succeeds the result variable will be set and
      stored in the cache so that neither call will search again.
 
      When searching for frameworks, if the file is specified as A/b.h, then
      the framework search will look for A.framework/Headers/b.h.  If that
      is found the path will be set to the path to the framework.  CMake
      will convert this to the correct -F option to include the file.
 
  find_program
      Find an executable program.
 
          find_program(<VAR> name1 [path1 path2 ...])
 
      This is the short-hand signature for the command that is sufficient in
      many cases.  It is the same as find_program(<VAR> name1 [PATHS path1
      path2 ...])
 
          find_program(
                    <VAR>
                    name | NAMES name1 [name2 ...]
                    [PATHS path1 [path2 ... ENV var]]
                    [PATH_SUFFIXES suffix1 [suffix2 ...]]
                    [DOC "cache documentation string"]
                    [NO_DEFAULT_PATH]
                    [NO_CMAKE_ENVIRONMENT_PATH]
                    [NO_CMAKE_PATH]
                    [NO_SYSTEM_ENVIRONMENT_PATH]
                    [NO_CMAKE_SYSTEM_PATH]
                    [CMAKE_FIND_ROOT_PATH_BOTH |
                    ONLY_CMAKE_FIND_ROOT_PATH |
                    NO_CMAKE_FIND_ROOT_PATH]
                  )
 
      This command is used to find a program.  A cache entry named by <VAR>
      is created to store the result of this command.  If the program is
      found the result is stored in the variable and the search will not be
      repeated unless the variable is cleared.  If nothing is found, the
      result will be <VAR>-NOTFOUND, and the search will be attempted again
      the next time find_program is invoked with the same variable.  The
      name of the program that is searched for is specified by the names
      listed after the NAMES argument.  Additional search locations can be
      specified after the PATHS argument.  If ENV var is found in the PATHS
      section the environment variable var will be read and converted from a
      system environment variable to a cmake style list of paths.  For
      example ENV PATH would be a way to list the system path variable.  The
      argument after DOC will be used for the documentation string in the
      cache.  PATH_SUFFIXES can be used to give sub directories that will be
      appended to the search paths.
 
      If NO_DEFAULT_PATH is specified, then no additional paths are added to
      the search.  If NO_DEFAULT_PATH is not specified, the search process
      is as follows:
 
      1.  Search cmake specific environment variables.  This can be skipped
      if NO_CMAKE_ENVIRONMENT_PATH is passed.
 
          <prefix>/[s]bin for each <prefix> in CMAKE_PREFIX_PATH
          CMAKE_PROGRAM_PATH
          CMAKE_APPBUNDLE_PATH
 
      2.  Search cmake variables with the same names as the cmake specific
      environment variables.  These are intended to be used on the command
      line with a -DVAR=value.  This can be skipped if NO_CMAKE_PATH is
      passed.
 
          <prefix>/[s]bin for each <prefix> in CMAKE_PREFIX_PATH
          CMAKE_PROGRAM_PATH
          CMAKE_APPBUNDLE_PATH
 
      3.  Search the standard system environment variables.  This can be
      skipped if NO_SYSTEM_ENVIRONMENT_PATH is an argument.
 
          PATH
         
 
      4.  Search cmake variables defined in the Platform files for the
      current system.  This can be skipped if NO_CMAKE_SYSTEM_PATH is
      passed.
 
          <prefix>/[s]bin for each <prefix> in CMAKE_SYSTEM_PREFIX_PATH
          CMAKE_SYSTEM_PROGRAM_PATH
          CMAKE_SYSTEM_APPBUNDLE_PATH
 
      5.  Search the paths specified after PATHS or in the short-hand
      version of the command.
 
      On Darwin or systems supporting OS X Frameworks, the cmake variable
      CMAKE_FIND_FRAMEWORK can be set to empty or one of the following:
 
          "FIRST"  - Try to find frameworks before standard
                    libraries or headers. This is the default on Darwin.
          "LAST"  - Try to find frameworks after standard
                    libraries or headers.
          "ONLY"  - Only try to find frameworks.
          "NEVER". - Never try to find frameworks.
 
      On Darwin or systems supporting OS X Application Bundles, the cmake
      variable CMAKE_FIND_APPBUNDLE can be set to empty or one of the
      following:
 
          "FIRST"  - Try to find application bundles before standard
                    programs. This is the default on Darwin.
          "LAST"  - Try to find application bundles after standard
                    programs.
          "ONLY"  - Only try to find application bundles.
          "NEVER". - Never try to find application bundles.
 
      The CMake variable CMAKE_FIND_ROOT_PATH specifies one or more
      directories to be prepended to all other search directories.  This
      effectively "re-roots" the entire search under given locations.  By
      default it is empty.  It is especially useful when cross-compiling to
      point to the root directory of the target environment and CMake will
      search there too.  By default at first the directories listed in
      CMAKE_FIND_ROOT_PATH and then the non-rooted directories will be
      searched.  The default behavior can be adjusted by setting
      CMAKE_FIND_ROOT_PATH_MODE_PROGRAM.  This behavior can be manually
      overridden on a per-call basis.  By using CMAKE_FIND_ROOT_PATH_BOTH
      the search order will be as described above.  If
      NO_CMAKE_FIND_ROOT_PATH is used then CMAKE_FIND_ROOT_PATH will not be
      used.  If ONLY_CMAKE_FIND_ROOT_PATH is used then only the re-rooted
      directories will be searched.
 
      The reason the paths listed in the call to the command are searched
      last is that most users of CMake would expect things to be found first
      in the locations specified by their environment.  Projects may
      override this behavior by simply calling the command twice:
 
          find_program(<VAR> NAMES name PATHS paths... NO_DEFAULT_PATH)
          find_program(<VAR> NAMES name)
 
      Once one of these calls succeeds the result variable will be set and
      stored in the cache so that neither call will search again.
 
  foreach
      Evaluate a group of commands for each value in a list.
 
        foreach(loop_var arg1 arg2 ...)
          COMMAND1(ARGS ...)
          COMMAND2(ARGS ...)
          ...
        endforeach(loop_var)
        foreach(loop_var RANGE total)
        foreach(loop_var RANGE start stop [step])
 
      All commands between foreach and the matching endforeach are recorded
      without being invoked.  Once the endforeach is evaluated, the recorded
      list of commands is invoked once for each argument listed in the
      original foreach command.  Before each iteration of the loop
      "${loop_var}" will be set as a variable with the current value in the
      list.
 
      Foreach can also iterate over a generated range of numbers.  There are
      three types of this iteration:
 
      * When specifying single number, the range will have elements 0 to
      "total".
 
      * When specifying two numbers, the range will have elements from the
      first number to the second number.
 
      * The third optional number is the increment used to iterate from the
      first number to the second number.
 
  function
      Start recording a function for later invocation as a command.
 
        function(<name> [arg1 [arg2 [arg3 ...]]])
          COMMAND1(ARGS ...)
          COMMAND2(ARGS ...)
          ...
        endfunction(<name>)
 
      Define a function named <name> that takes arguments named arg1 arg2
      arg3 (...).  Commands listed after function, but before the matching
      endfunction, are not invoked until the function is invoked.  When it
      is invoked, the commands recorded in the function are first modified
      by replacing formal parameters (${arg1}) with the arguments passed,
      and then invoked as normal commands.  In addition to referencing the
      formal parameters you can reference the variable ARGC which will be
      set to the number of arguments passed into the function as well as
      ARGV0 ARGV1 ARGV2 ...  which will have the actual values of the
      arguments passed in.  This facilitates creating functions with
      optional arguments.  Additionally ARGV holds the list of all arguments
      given to the function and ARGN holds the list of argument pass the
      last expected argument.
 
  get_cmake_property
      Get a property of the CMake instance.
 
        get_cmake_property(VAR property)
 
      Get a property from the CMake instance.  The value of the property is
      stored in the variable VAR.  If the property is not found, CMake will
      report an error.  Some supported properties include: VARIABLES,
      CACHE_VARIABLES, COMMANDS, and MACROS.
 
  get_directory_property
      Get a property of the directory.
 
        get_directory_property(VAR [DIRECTORY dir] property)
 
      Get a property from the Directory.  The value of the property is
      stored in the variable VAR.  If the property is not found, CMake will
      report an error.  The properties include: VARIABLES, CACHE_VARIABLES,
      COMMANDS, MACROS, INCLUDE_DIRECTORIES, LINK_DIRECTORIES, DEFINITIONS,
      INCLUDE_REGULAR_EXPRESSION, LISTFILE_STACK, PARENT_DIRECTORY, and
      DEFINITION varname.  If the DIRECTORY argument is provided then the
      property of the provided directory will be retrieved instead of the
      current directory.  You can only get properties of a directory during
      or after it has been traversed by cmake.
 
  get_filename_component
      Get a specific component of a full filename.
 
        get_filename_component(VarName FileName
                                PATH|ABSOLUTE|NAME|EXT|NAME_WE
                                [CACHE])
 
      Set VarName to be the path (PATH), file name (NAME), file extension
      (EXT), file name without extension (NAME_WE) of FileName, or the full
      absolute (ABSOLUTE) file name without symlinks.  Note that the path is
      converted to Unix slashes format and has no trailing slashes.  The
      longest file extension is always considered.  If the optional CACHE
      argument is specified, the result variable is added to the cache.
 
        get_filename_component(VarName FileName
                                PROGRAM [PROGRAM_ARGS ArgVar]
                                [CACHE])
 
      The program in FileName will be found in the system search path or
      left as a full path.  If PROGRAM_ARGS is present with PROGRAM, then
      any command-line arguments present in the FileName string are split
      from the program name and stored in ArgVar.  This is used to separate
      a program name from its arguments in a command line string.
 
  get_property
      Get a property.
 
        get_property(<variable>
                      <GLOBAL            |
                      DIRECTORY [dir]    |
                      TARGET    <target> |
                      SOURCE    <source> |
                      TEST      <test>  |
                      VARIABLE>
                      PROPERTY <name>
                      [SET | DEFINED | BRIEF_DOCS | FULL_DOCS])
 
      Get one property from one object in a scope.  The first argument
      specifies the variable in which to store the result.  The second
      argument determines the scope from which to get the property.  It must
      be one of the following:
 
      GLOBAL scope is unique and does not accept a name.
 
      DIRECTORY scope defaults to the current directory but another
      directory (already processed by CMake) may be named by full or
      relative path.
 
      TARGET scope must name one existing target.
 
      SOURCE scope must name one source file.
 
      TEST scope must name one existing test.
 
      VARIABLE scope is unique and does not accept a name.
 
      The required PROPERTY option is immediately followed by the name of
      the property to get.  If the property is not set an empty value is
      returned.  If the SET option is given the variable is set to a boolean
      value indicating whether the property has been set.If the DEFINED
      option is given the variable is set to a boolean value indicating
      whether the property has been defined such as with define_property.
      If BRIEF_DOCS or FULL_DOCS is given then the variable is set to a
      string containing documentation for the requested property.  If
      documentation is requested for a property that has not been defined
      NOTFOUND is returned.
 
  if
      Conditionally execute a group of commands.
 
        if(expression)
          # then section.
          COMMAND1(ARGS ...)
          COMMAND2(ARGS ...)
          ...
        elseif(expression2)
          # elseif section.
          COMMAND1(ARGS ...)
          COMMAND2(ARGS ...)
          ...
        else(expression)
          # else section.
          COMMAND1(ARGS ...)
          COMMAND2(ARGS ...)
          ...
        endif(expression)
 
      Evaluates the given expression.  If the result is true, the commands
      in the THEN section are invoked.  Otherwise, the commands in the else
      section are invoked.  The elseif and else sections are optional.  You
      may have multiple elseif clauses.  Note that the same expression must
      be given to if, and endif.  Long expressions can be used and the order
      or precedence is that the EXISTS, COMMAND, and DEFINED operators will
      be evaluated first.  Then any EQUAL, LESS, GREATER, STRLESS,
      STRGREATER, STREQUAL, MATCHES will be evaluated.  Then NOT operators
      and finally AND, OR operators will be evaluated.  Possible expressions
      are:
 
        if(variable)
 
      True if the variable's value is not empty, 0, N, NO, OFF, FALSE,
      NOTFOUND, or <variable>-NOTFOUND.
 
        if(NOT variable)
 
      True if the variable's value is empty, 0, N, NO, OFF, FALSE, NOTFOUND,
      or <variable>-NOTFOUND.
 
        if(variable1 AND variable2)
 
      True if both variables would be considered true individually.
 
        if(variable1 OR variable2)
 
      True if either variable would be considered true individually.
 
        if(COMMAND command-name)
 
      True if the given name is a command, macro or function that can be
      invoked.
 
        if(POLICY policy-id)
 
      True if the given name is an existing policy (of the form CMP<NNNN>).
 
        if(EXISTS file-name)
        if(EXISTS directory-name)
 
      True if the named file or directory exists.  Behavior is well-defined
      only for full paths.
 
        if(file1 IS_NEWER_THAN file2)
 
      True if file1 is newer than file2 or if one of the two files doesn't
      exist.  Behavior is well-defined only for full paths.
 
        if(IS_DIRECTORY directory-name)
 
      True if the given name is a directory.  Behavior is well-defined only
      for full paths.
 
        if(IS_ABSOLUTE path)
 
      True if the given path is an absolute path.
 
          if(variable MATCHES regex)
        if(string MATCHES regex)
 
      True if the given string or variable's value matches the given regular
      expression.
 
        if(variable LESS number)
        if(string LESS number)
        if(variable GREATER number)
        if(string GREATER number)
        if(variable EQUAL number)
        if(string EQUAL number)
 
      True if the given string or variable's value is a valid number and the
      inequality or equality is true.
 
        if(variable STRLESS string)
        if(string STRLESS string)
        if(variable STRGREATER string)
        if(string STRGREATER string)
        if(variable STREQUAL string)
        if(string STREQUAL string)
 
      True if the given string or variable's value is lexicographically less
      (or greater, or equal) than the string on the right.
 
        if(DEFINED variable)
 
      True if the given variable is defined.  It does not matter if the
      variable is true or false just if it has been set.
 
  include
      Read CMake listfile code from the given file.
 
        include(file1 [OPTIONAL] [RESULT_VARIABLE <VAR>])
        include(module [OPTIONAL] [RESULT_VARIABLE <VAR>])
 
      Reads CMake listfile code from the given file.  Commands in the file
      are processed immediately as if they were written in place of the
      include command.  If OPTIONAL is present, then no error is raised if
      the file does not exist.  If RESULT_VARIABLE is given the variable
      will be set to the full filename which has been included or NOTFOUND
      if it failed.
 
      If a module is specified instead of a file, the file with name
      <modulename>.cmake is searched in the CMAKE_MODULE_PATH.
 
  list
      List operations.
 
        list(LENGTH <list> <output variable>)
        list(GET <list> <element index> [<element index> ...] <output variable>)
        list(APPEND <list> <element> [<element> ...])
        list(FIND <list> <value> <output variable>)
        list(INSERT <list> <element_index> <element> [<element> ...])
        list(REMOVE_ITEM <list> <value> [<value> ...])
        list(REMOVE_AT <list> <index> [<index> ...])
        list(REMOVE_DUPLICATES <list>)
        list(REVERSE <list>)
        list(SORT <list>)
 
      LENGTH will return a given list's length.
 
      GET will return list of elements specified by indices from the list.
 
      APPEND will append elements to the list.
 
      FIND will return the index of the element specified in the list or -1
      if it wasn't found.
 
      INSERT will insert elements to the list to the specified location.
 
      REMOVE_AT and REMOVE_ITEM will remove items from the list.  The
      difference is that REMOVE_ITEM will remove the given items, while
      REMOVE_AT will remove the items at the given indices.
 
      REMOVE_DUPLICATES will remove duplicated items in the list.
 
      REVERSE reverses the contents of the list in-place.
 
      SORT sorts the list in-place alphabetically.
 
      NOTES: A list in cmake is a ; separated group of strings.  To create a
      list the set command can be used.  For example, set(var a b c d e)
      creates a list with a;b;c;d;e, and set(var "a b c d e") creates a
      string or a list with one item in it.
 
      When specifying index values, if <element index> is 0 or greater, it
      is indexed from the beginning of the list, with 0 representing the
      first list element.  If <element index> is -1 or lesser, it is indexed
      from the end of the list, with -1 representing the last list element.
      Be careful when counting with negative indices: they do not start from
      0.  -0 is equivalent to 0, the first list element.
 
 
  macro
      Start recording a macro for later invocation as a command.
 
        macro(<name> [arg1 [arg2 [arg3 ...]]])
          COMMAND1(ARGS ...)
          COMMAND2(ARGS ...)
          ...
        endmacro(<name>)
 
      Define a macro named <name> that takes arguments named arg1 arg2 arg3
      (...).  Commands listed after macro, but before the matching endmacro,
      are not invoked until the macro is invoked.  When it is invoked, the
      commands recorded in the macro are first modified by replacing formal
      parameters (${arg1}) with the arguments passed, and then invoked as
      normal commands.  In addition to referencing the formal parameters you
      can reference the values ${ARGC} which will be set to the number of
      arguments passed into the function as well as ${ARGV0} ${ARGV1}
      ${ARGV2} ...  which will have the actual values of the arguments
      passed in.  This facilitates creating macros with optional arguments.
      Additionally ${ARGV} holds the list of all arguments given to the
      macro and ${ARGN} holds the list of argument pass the last expected
      argument.  Note that the parameters to a macro and values such as ARGN
      are not variables in the usual CMake sense.  They are string
      replacements much like the c preprocessor would do with a macro.  If
      you want true CMake variables you should look at the function command.
 
  make_directory
      Deprecated.  Use the file(MAKE_DIRECTORY ) command instead.
 
        make_directory(directory)
 
      Creates the specified directory.  Full paths should be given.  Any
      parent directories that do not exist will also be created.  Use with
      care.
 
  mark_as_advanced
      Mark cmake cached variables as advanced.
 
        mark_as_advanced([CLEAR|FORCE] VAR VAR2 VAR...)
 
      Mark the named cached variables as advanced.  An advanced variable
      will not be displayed in any of the cmake GUIs unless the show
      advanced option is on.  If CLEAR is the first argument advanced
      variables are changed back to unadvanced.  If FORCE is the first
      argument, then the variable is made advanced.  If neither FORCE nor
      CLEAR is specified, new values will be marked as advanced, but if the
      variable already has an advanced/non-advanced state, it will not be
      changed.
 
      It does nothing in script mode.
 
  math
      Mathematical expressions.
 
        math(EXPR <output variable> <math expression>)
 
      EXPR evaluates mathematical expression and return result in the output
      variable.  Example mathematical expression is '5 * ( 10 + 13 )'.
      Supported operators are + - * / % | & ^ ~ << >> * / %.  They have the
      same meaning as they do in c code.
 
  message
      Display a message to the user.
 
        message([SEND_ERROR | STATUS | FATAL_ERROR]
                "message to display" ...)
 
      By default the message is displayed in a pop up window (CMakeSetup),
      or in the stdout of cmake, or the error section of ccmake.  If the
      first argument is SEND_ERROR then an error is raised, and the generate
      phase will be skipped.  If the first argument is FATAL_ERROR, all
      processing is halted.  If the first argument is STATUS then the
      message is displayed in the progress line for the GUI, or with a -- in
      the command line cmake.
 
  option
      Provides an option that the user can optionally select.
 
        option(<option_variable> "help string describing option"
                [initial value])
 
      Provide an option for the user to select as ON or OFF.  If no initial
      value is provided, OFF is used.
 
  remove
      Deprecated.  Use the list(REMOVE_ITEM ) command instead.
 
        remove(VAR VALUE VALUE ...)
 
      Removes VALUE from the variable VAR.  This is typically used to remove
      entries from a vector (e.g.  semicolon separated list).  VALUE is
      expanded.
 
  return
      Return from a directory or function.
 
        return()
 
      Returns from a directory or function.  When this command is
      encountered, it caused process of the current function or directory to
      stop and control is return to the caller of the function, or the
      parent directory if any.  Note that a macro is not a function and does
      not handle return like a function does.
 
  separate_arguments
      Split space separated arguments into a semi-colon separated list.
 
        separate_arguments(VARIABLE)
 
      Convert the value of VARIABLE to a semi-colon separated list.  All
      spaces are replaced with ';'.  This helps with generating command
      lines.
 
  set
      Set a CMAKE variable to a given value.
 
        set(<variable> <value> [[CACHE <type> <docstring> [FORCE]] | PARENT_SCOPE])
 
      Within CMake sets <variable> to the value <value>.  <value> is
      expanded before <variable> is set to it.  If CACHE is present, then
      the <variable> is put in the cache.  <type> and <docstring> are then
      required.  <type> is used by the CMake GUI to choose a widget with
      which the user sets a value.  The value for <type> may be one of
 
        FILEPATH = File chooser dialog.
        PATH    = Directory chooser dialog.
        STRING  = Arbitrary string.
        BOOL    = Boolean ON/OFF checkbox.
        INTERNAL = No GUI entry (used for persistent variables).
 
      If <type> is INTERNAL, then the <value> is always written into the
      cache, replacing any values existing in the cache.  If it is not a
      cache variable, then this always writes into the current makefile.
      The FORCE option will overwrite the cache value removing any changes
      by the user.
 
      If PARENT_SCOPE is present, the variable will be set in the scope
      above the current scope.  Each new directory or function creates a new
      scope.  This command will set the value of a variable into the parent
      directory or calling function (whichever is applicable to the case at
      hand) If VALUE is not specified then the variable is removed from the
      parent scope.
 
        set(<variable> <value1> ... <valueN>)
 
      In this case <variable> is set to a semicolon separated list of
      values.
 
      <variable> can be an environment variable such as:
 
        set( ENV{PATH} /home/martink )
 
      in which case the environment variable will be set.
 
  set_directory_properties
      Set a property of the directory.
 
        set_directory_properties(PROPERTIES prop1 value1 prop2 value2)
 
      Set a property for the current directory and subdirectories.  If the
      property is not found, CMake will report an error.  The properties
      include: INCLUDE_DIRECTORIES, LINK_DIRECTORIES,
      INCLUDE_REGULAR_EXPRESSION, and ADDITIONAL_MAKE_CLEAN_FILES.
 
      ADDITIONAL_MAKE_CLEAN_FILES is a list of files that will be cleaned as
      a part of "make clean" stage.
 
  set_property
      Set a named property in a given scope.
 
        set_property(<GLOBAL                            |
                      DIRECTORY [dir]                  |
                      TARGET    [target1 [target2 ...]] |
                      SOURCE    [src1 [src2 ...]]      |
                      TEST      [test1 [test2 ...]]>
                      [APPEND]
                      PROPERTY <name> [value1 [value2 ...]])
 
      Set one property on zero or more objects of a scope.  The first
      argument determines the scope in which the property is set.  It must
      be one of the following:
 
      GLOBAL scope is unique and does not accept a name.
 
      DIRECTORY scope defaults to the current directory but another
      directory (already processed by CMake) may be named by full or
      relative path.
 
      TARGET scope may name zero or more existing targets.
 
      SOURCE scope may name zero or more source files.
 
      TEST scope may name zero or more existing tests.
 
      The required PROPERTY option is immediately followed by the name of
      the property to set.  Remaining arguments are used to compose the
      property value in the form of a semicolon-separated list.  If the
      APPEND option is given the list is appended to any existing property
      value.
 
  site_name
      Set the given variable to the name of the computer.
 
        site_name(variable)
 
 
  string
      String operations.
 
        string(REGEX MATCH <regular_expression>
                <output variable> <input> [<input>...])
        string(REGEX MATCHALL <regular_expression>
                <output variable> <input> [<input>...])
        string(REGEX REPLACE <regular_expression>
                <replace_expression> <output variable>
                <input> [<input>...])
        string(REPLACE <match_string>
                <replace_string> <output variable>
                <input> [<input>...])
        string(COMPARE EQUAL <string1> <string2> <output variable>)
        string(COMPARE NOTEQUAL <string1> <string2> <output variable>)
        string(COMPARE LESS <string1> <string2> <output variable>)
        string(COMPARE GREATER <string1> <string2> <output variable>)
        string(ASCII <number> [<number> ...] <output variable>)
        string(CONFIGURE <string1> <output variable>
                [@ONLY] [ESCAPE_QUOTES])
        string(TOUPPER <string1> <output variable>)
        string(TOLOWER <string1> <output variable>)
        string(LENGTH <string> <output variable>)
        string(SUBSTRING <string> <begin> <length> <output variable>)
        string(STRIP <string> <output variable>)
        string(RANDOM [LENGTH <length>] [ALPHABET <alphabet>]
                <output variable>)
 
      REGEX MATCH will match the regular expression once and store the match
      in the output variable.
 
      REGEX MATCHALL will match the regular expression as many times as
      possible and store the matches in the output variable as a list.
 
      REGEX REPLACE will match the regular expression as many times as
      possible and substitute the replacement expression for the match in
      the output.  The replace expression may refer to paren-delimited
      subexpressions of the match using \1, \2, ..., \9.  Note that two
      backslashes (\\1) are required in CMake code to get a backslash
      through argument parsing.
 
      REPLACE will replace all occurrences of match_string in the input with
      replace_string and store the result in the output.
 
      COMPARE EQUAL/NOTEQUAL/LESS/GREATER will compare the strings and store
      true or false in the output variable.
 
      ASCII will convert all numbers into corresponding ASCII characters.
 
      CONFIGURE will transform a string like CONFIGURE_FILE transforms a
      file.
 
      TOUPPER/TOLOWER will convert string to upper/lower characters.
 
      LENGTH will return a given string's length.
 
      SUBSTRING will return a substring of a given string.
 
      STRIP will return a substring of a given string with leading and
      trailing spaces removed.
 
      RANDOM will return a random string of given length consisting of
      characters from the given alphabet.  Default length is 5 characters
      and default alphabet is all numbers and upper and lower case letters.
 
      The following characters have special meaning in regular expressions:
 
          ^        Matches at beginning of a line
          $        Matches at end of a line
          .        Matches any single character
          [ ]      Matches any character(s) inside the brackets
          [^ ]      Matches any character(s) not inside the brackets
          -        Matches any character in range on either side of a dash
          *        Matches preceding pattern zero or more times
          +        Matches preceding pattern one or more times
          ?        Matches preceding pattern zero or once only
          |        Matches a pattern on either side of the |
          ()        Saves a matched subexpression, which can be referenced in the REGEX REPLACE operation. Additionally it is saved in the special CMake variables CMAKE_MATCH_(0..9).
 
  variable_watch
      Watch the CMake variable for change.
 
        variable_watch(<variable name> [<command to execute>])
 
      If the specified variable changes, the message will be printed about
      the variable being changed.  If the command is specified, the command
      will be executed.  The command will receive the following arguments:
      COMMAND(<variable> <access> <value> <current list file> <stack>)
 
  while
      Evaluate a group of commands while a condition is true
 
        while(condition)
          COMMAND1(ARGS ...)
          COMMAND2(ARGS ...)
          ...
        endwhile(condition)
 
      All commands between while and the matching endwhile are recorded
      without being invoked.  Once the endwhile is evaluated, the recorded
      list of commands is invoked as long as the condition is true.  The
      condition is evaluated using the same logic as the if command.
 
  write_file
      Deprecated.  Use the file(WRITE ) command instead.
 
        write_file(filename "message to write"... [APPEND])
 
      The first argument is the file name, the rest of the arguments are
      messages to write.  If the argument APPEND is specified, then the
      message will be appended.
 
      NOTE 1: file(WRITE ...  and file(APPEND ...  do exactly the same as
      this one but add some more functionality.
 
      NOTE 2: When using write_file the produced file cannot be used as an
      input to CMake (CONFIGURE_FILE, source file ...) because it will lead
      to an infinite loop.  Use configure_file if you want to generate input
      files to CMake.
 
------------------------------------------------------------------------------
Properties
 
  CMake Properties - Properties supported by CMake, the Cross-Platform Makefile Generator.
 
This is the documentation for the properties supported by CMake.  Properties
can have different scopes.  They can either be assigned to a source file, a
directory, a target or globally to CMake.  By modifying the values of
properties the behaviour of the build system can be customized.
 
------------------------------------------------------------------------------
Compatibility Commands
 
  CMake Compatibility Listfile Commands - Obsolete commands supported by CMake for compatibility.
 
This is the documentation for now obsolete listfile commands from previous
CMake versions, which are still supported for compatibility reasons.  You
should instead use the newer, faster and shinier new commands.  ;-)
 
------------------------------------------------------------------------------
Standard CMake Modules
 
------------------------------------------------------------------------------
Copyright
 
Copyright (c) 2002 Kitware, Inc., Insight Consortium.  All rights reserved.
 
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
 
Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.
 
Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
 
The names of Kitware, Inc., the Insight Consortium, or the names of any
consortium members, or of any contributors, may not be used to endorse or
promote products derived from this software without specific prior written
permission.
 
Modified source versions must be plainly marked as such, and must not be
misrepresented as being the original software.
 
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER AND CONTRIBUTORS ``AS IS''
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHORS OR CONTRIBUTORS BE LIABLE FOR
ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
------------------------------------------------------------------------------
See Also
 
The following resources are available to get help using CMake:
 
  Home Page
      http://www.cmake.org
 
      The primary starting point for learning about CMake.
 
  Frequently Asked Questions
      http://www.cmake.org/Wiki/CMake_FAQ
 
      A Wiki is provided containing answers to frequently asked questions.
 
  Online Documentation
      http://www.cmake.org/HTML/Documentation.html
 
      Links to available documentation may be found on this web page.
 
  Mailing List
      http://www.cmake.org/HTML/MailingLists.html
 
      For help and discussion about using cmake, a mailing list is provided
      at cmake@cmake.org.  The list is member-post-only but one may sign up
      on the CMake web page.  Please first read the full documentation at
      http://www.cmake.org before posting questions to the list.
 
Summary of helpful links:
 
  Home: http://www.cmake.org
  Docs: http://www.cmake.org/HTML/Documentation.html
  Mail: http://www.cmake.org/HTML/MailingLists.html
  FAQ:  http://www.cmake.org/Wiki/CMake_FAQ
</pre>

Revision as of 13:04, 24 April 2018