If creation of the global generator fails, return early with an error
message instead of trying to use the generator and crashing.
Add a CTestTestBadGenerator test to cover this case.
Reported-by: Mathieu Malaterre <malat@debian.org>
Bug-Debian: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=747306
By default, Intel compiler coverage tools generate HTML files as
reports, but the option -txtlcov can be given to codecov to output a
coverage file with LCov format.
To use Intel coverage:
* build the project with coverage flags
* run the application
* run profmerge
* run codecov
The output file will be "build_dir/CodeCoverage/SRCFILEDIR.LCOV".
Ask users to compile with -prof-dir${BUILD_DIR} instead of searching
the entire build tree recursively to find coverage files.
Use the clang RemoveCStrCalls tool to automatically migrate the
code. This was only run on linux, so does not have any positive or
negative effect on other platforms.
Casts from std::string -> cmStdString were high on the list of things
taking up time. Avoid such implicit casts across function calls by just
using std::string everywhere.
The comment that the symbol name is too long is no longer relevant since
modern debuggers alias the templates anyways and the size is a
non-issue since the underlying methods are generated since it's
inherited.
My last related commit e5e3f3d4 (CTest: filter /showIncludes output from
ninja compile launcher, 2013-12-01) filtered /showIncludes messages from
the generated xml output but they also need to be filtered in
ScrapeLog(). Otherwise they are being detected as warnings when using
compilers withs english diagnostics.
Mark unknown revisions as such and fail instead of reporting revision 0.
Otherwise CTest reports massive file updates between revisions when the
server timeouts while trying to fetch the current revision number.
The first regression resulted in endless looping due to unrun test
dependencies. The second regression prioritized all tests with dependencies
in serial test runs.
The --build-options option consumes all following arguments until either
--build-target or --test-command. Fix the logic to allow this to be
zero options.
GenerateBuildCommand now knows how to lookup CMAKE_MAKE_PROGRAM or
choose a generator-provided default build tool. Therefore the
--build-makeprogram can now be optional and simply override the
default selection when provided.
Note that with --build-nocmake we now need to load the cache in order to
make the CMAKE_MAKE_PROGRAM entry available to GenerateBuildCommand.
This avoids requiring cmGlobalGenerator::GenerateBuildCommand to produce
a string so that it can be later refactored to produce a vector with no
quoting or escaping. It also makes the ctest_build command match what
"ctest -T Build" would run in a build tree configured with the new
build_command() command behavior. It also ensures that the native build
tool used matches that selected by the configuration of the tree to be
built.
Rename cmSystemTools::FindExecutableDirectory to FindCMakeResources.
Teach it to compute the locations of cmake, ctest, cpack, ccmake, and
cmake-gui executables, and the location of CMAKE_ROOT. Provide this
information from static cmSystemTools::Get<resource>() methods.
Refactor code that needs these locations to use the new APIs.
Teach FindCMakeResources to use the OS X system API to lookup the
executable location. When running from the CMake build tree itself,
leave a file in the tree that FindCMakeResources can use to read the
location of the source tree. This avoids the need to compile the source
tree location into a binary that may be installed and used without the
source tree.
Teach the QtDialog on OS X to create a "cmake-gui" symlink in the build
tree next to "cmake" and the other tools, as is already done in the
install tree for the application bundle. This ensures a consistent set
of executables are available in one directory.
Make sure that CMakeCache.txt is the last file being removed since
the binary directory may be left in a state that is no longer
removable otherwise.
Also retry removal a couple of times which makes this more robust
on windows where file locks may temporarily prevent removal.
Instead of removing the depot name, which causes problems when looking
at the file change list in CDash, make the Update test remove a given
prefix from the files retrieved from Update.xml.
ff59365 CTest: fix dashboard issues associated with the ctest-fix-run-serial topic
7a665ae CTest: added test for RUN_SERIAL issue #14484384beff CTest: added comments that describe the basic test sorting approach
adbe00d CTest: removed redundant copy of test dependency set
1b750cb CTest: perform cycle test early
6d4d7ca CTest: consider previously failed tests before all others
e809d8c CTest: prioritize tests by their depth in the dependency graph
44017a4 CTest: handle dependent and non dependent test requirements equally
Teach the ctest_update implementation to use the p4 command-line
client to perform updates and extract the list of changes.
Add a CTest.UpdateP4 test like those that exist already for the other
version control tools. Make the test available when p4 and the p4d
server are found. During the test launch p4d in the background to
serve a repository from the test directory. Then direct the client
toward this server for the duration of the test.
We will no longer support full documentation generation from executables
and will instead generate documentation with other tools. Disable (with
a warning left behind) the command-line options:
--copyright
--help-compatcommands
--help-full
--help-html
--help-man
Drop supporting code. Drop manual sections generation from executables.
Remove internal documentation construction APIs. Drop unused sections
See Also, Author, Copyright, Compat Commands, Custom Modules.
This assumes that coverage.py has been run in such a way to produce its
standard XML output. This uses the Cobertura schema and should be somewhat
generalizable.
Add a new command line argument to ctest. This allows users to
rerun tests that failed during the previous call to ctest. This
is accomplished by analyzing the most recently modified file named
"^LastTestsFailed*" in the Testing/Temporary subdirectory of the
project's binary directory.
The commits 9db31162 (Remove CMake-language block-end command
arguments, 2012-08-13) and 77543bde (Convert CMake-language
commands to lower case, 2012-08-13) changed most cmake code
to use lowercase commands and no parameters in termination
commands. However, those changes excluded cmake code generated
in c++ by cmake.
Make a similar style change to code generated by cmake.
The output file used for memory checker runs must be unique for every test run
in parallel, so simply make them unique for every test run. Simply use the test
index to avoid collisions.
This reverts commit 6187876dea.
It was actually possible before to have paths with spaces in them, the spaces
just need to be quoted. This way spaces will work as argument separators.
10bc50e Tests: ignore Guard Malloc messages in MemChecker tests
159c3e9 Tests: add a test with custom options passed to valgrind
61ddb93 CTest: fix comment documenting cmBoundsCheckerParser class
cbdfcb0 Tests: add test for non-existent Valgrind suppression file
3b5b758 CTest: drop suppression for gcc 2.9.6 errors from default Valgrind flags
7752253 Tests: verify that memory checker output files are always present
abf1df4 Tests: remove code duplication in CTestTestMemCheck tests
f499422 CTest: remove unreachable code and CTestTestMemcheckUnknown test
dde6306 CTest: use an output file for Valgrind (#14110)
bcc0f3f Tests: create output files for all memory checkers
At this point we do not care what the current directory is, as nothing that
requires a specific directory is done. Just make sure that at the end we are
back in the same directory as at the start.
PASS_REGULAR_EXPRESSION and FAIL_REGULAR_EXPRESSION both take a list of expressions.
Stop searching if the first of those has matched the output as the result will not
change anymore.
Because of clock scew between processors or just because of someone changing
the system time the end timestamp may be before the start time. Reporting a
negative time doesn't any sense, just report zero there as it already happens
for really fast tests.
The --workaround-gcc296-bugs has been part of the default Valgrind flags since
Valgrind support was added in commit 5b232ded15
(ENH: Add initial memory check support which works for Valgrind, 2003-12-15).
The Valgrind manpage says that this option should be avoided if not really
needed as it may cause real errors to get ignored. If someone uses a compiler
that really needs the flag this flag should be set by the user explicitely.
Most users will never set any flags and probably never notice that they use a
flag they shouldn't.
The memory checker command can't be quoted at this point, because previously it
has been tested that the given file exists, which will fail if the name is
quoted. The CTestTestMemcheckUnknown test aimed to test this case, has always
failed to do so and serves no useful purpose therefore.
This allows the ctest_build command's TARGET option to name a target
in a subdirectory and still build properly with msbuild.
Add test case covering use of ctest_build() with such a TARGET.
In cmCTestHandlerCommand::InitialPass call InitializeHandler after setting
the CTest "SourceDirectory" and "BuildDirectory" configuration values
instead of before. This makes the values available to InitializeHandler
methods in subclasses, so also drop the duplicate configuration of those
values from cmCTestConfigureCommand::InitializeHandler.
Extend the cmGlobalGenerator::GenerateBuildCommand virtual method
signature with a "projectDir" parameter specifying the top of the
project build tree for which the build command will be generated.
Populate it from call sites in cmGlobalGenerator::Build where a
fully-generated build tree should be available.
The ctest_configure command already reads the CTEST_CMAKE_GENERATOR
variable to get the value for the cmake -G option. Read new variable
CTEST_CMAKE_GENERATOR_TOOLSET for -T.
The "ctest --build-and-test" mode already has "--build-generator" to
specify the -G option to CMake. Add a "--build-generator-toolset" option
to specify the -T value.
Note it is still possible for CTest to start more than the number of
processes specified by PARALLEL_LEVEL, but this prevents the number of
tests to start from being unbounded because of overflow.
This is for consistency throughout cmake. The cmsys version exists
becaues uses of auto_ptr types as return types does not work with
some implementations in ancient compilers.
995a35f CTest: add a check with a quoted memory checker
de8bffc CTest: add a test for CTEST_CUSTOM_MEMCHECK_IGNORE
d26c9b6 CTest: improve memory checker type detection
fcae1da CTest: add tests that simulate memcheck runs
6187876 CTest: fix pre and post test commands with spaces
95bc8aa CTest: fix usage of memory checker with spaces in path
If the checker was explicitely set with a "TypeCommand" variable (e.g.
ValgrindCommand) then we now just believe that this is valgrind, even if
nothing in the path matches "valgrind". Only when "MemoryCheckCommand" was used
we still scan the path to find out what checker we have.
Hack access to cmCTestVC::Revision on this compiler to avoid errors like
Source/CTest/cmCTestSVN.cxx, line 23: Error: cmCTestSVN::SVNInfo is not accessible from cmCTestSVN::Revision.
Source/CTest/cmCTestSVN.cxx, line 22: Error: cmCTestVC::Revision is not accessible from file level.
If the pre or post memcheck or test commands have spaces in the path these
were never escaped, leading to broken commands. This was not covered in the
test suite so it went unnoticed.
The filename was escaped in cmCTestMemCheckHandler::InitializeMemoryChecking()
and again before it was written to output in
cmCTestRunTest::ComputeArguments().
Once someone uses e.g. a valgrind path with spaces this leads to double escaping
making the memory checker fail completely because of the invalid path.
This call 'svn status' and parse the result to get the list of externals
repositories. The external repositories found are added to the
Repositories list.
- Suppress the URL, Root, Base fields from the cmCTestSVN class
- Update the code to use RootInfo instead
- LoadInfo(), GuessBase(), and a new LoadRevision() functions work on a given SVNInfo
- Use the implementation from the base class for LocalPath(), correct path is built by SVNInfo::BuildLocalPath() instead
The Repository list will contain the SVNInfo of all the repositories
(root and external ones). The RootInfo pointer will point to the
SVNInfo structure of the root repository.
c806b23 CDash now supports lots of files in coverage. So, show all files.
761d931 Do not try to run bullseye coverage if COVFILE env is empty.
5b69ce4 Update test data to match new coverage format.
1b418f1 Change GT.M Coverage Parser global
b0c07a1 Disable bullseye coverage for mumps coverage test.
0a169e6 Remove uncovered files from cache coverage data.
a7abf5e Add ability to specify more than one package directory or coverage directory.
220afca Use <TARGET_FILE> expression to run ctest so it works with Xcode and VS IDE.
62f6bce Use a script to run the test because WORKING_DIRECTORY is not in 2.8.2.
f5c5db0 Fix some warnings and a bug where it went past the length of a vector.
7955e99 Add support for Cache coverage.
a86cd33 Add virutal destructor to silence warning.
319eeb0 Add test for mumps coverage. Also refactor code to prepare for cache coverage.
72210c2 Fix line length.
dd07161 Fix warning about char* instead of const char*.
e6412e0 Add support to ctest for GTM mumps coverage.
The coverage global should be in the local namespace. This means the global
will be ^ZZCOVERAGE instead of ^COVERAGE. Change the parser to look for ^ZZCOVERAGE
instead of the old ^COVERAGE
Replace use of AppendEnv/RestoreEnv pairs with instances of
SaveRestoreEnvironment. Simplify the signature of AppendEnv and use it
in place of similar loops elsewhere. Move the RestoreEnv implementation
inside the SaveRestoreEnvironment destructor which is the only place
left that calls it.
Commit ffbe61bb (make sure english is used for output of gcov, 2008-05-10)
taught ctest_coverage to set LC_ALL=POSIX to get English output from gcov.
Use the more portable value LC_ALL=C and restore the original value when
finished.
Even if there are no lines covered in the file the gcov coverage report still
contains valueable information, the amount of uncovered lines and which exactly
they are. Set 'Covered="true"' for files we have a gcov report for even if they
have no lines covered. Otherwise CDash will neither show the uncovered line
count nor the detailed coverage report for this file.
When CTEST_EXTRA_COVERAGE_GLOB was used to collect otherwise uncovered files
'Covered="true"' was unconditionally set, so this can't be worse here.
Fix the regex meant to match errors of the form:
C:\some\dir\source.cpp(17): catastrophic error: could not open source file "some_header.h"
to make the space between ')' and ':' optional.
COVERAGE_EXTRA_FLAGS is a space separated value of extra flags
that will be passed to gcov when ctest's coverage handler invokes
gcov to do coverage analysis.
Map to CoverageExtraFlags in the CTest ini file. Use default value
of "-l" to match the coverage handler's earlier behavior from ctest
2.8.4 and earlier. The fix for related issue #11717 had added a " -p"
which was the cause of both #12415 and #12490. Here, we revert that
change to the default value, so -p is no longer there by default.
The people that care to add -p may do so in their own build trees
by appending " -p" to the new cache variable COVERAGE_EXTRA_FLAGS.
Previous code was missing some matches in the output. This commit
fixes the regular expressions used for output matching to detect
numbers reported with commas in them, too.
This deterministic ordering makes it easier for a developer
comparing the ctest output of two test runs to see what changed
from run to run.
Thanks to Fraser Hutchison for the patch.
When add_test(NAME) is called without the CONFIGURATIONS argument then
the test is intended to run in any configuration. In multi-config
generators like the VS IDE and Xcode tests created by add_test(NAME) can
only be run when testing a known configuration (otherwise there is no
way to generate the test command line). If no test command line is
known for a particular configuration, or if no configuration is given to
ctest, report the test as not run instead of silently skipping it.
Also fix CMake's own TestsWorkingDirectory test invocation to correct a
previously silent failure exposed by this change.
3e32db7 cmCTestUploadCommand::CheckArgumentKeyword should return false if not FILES
6b6f309 Add the FILES keyword to ctest_upload command
28cdd0a Don't tar/gz ctest_upload() files
fbe4356 Change 'Files' tag to 'Upload' in Upload.xml
350546d Implement ctest_upload command
Some samples of things that got unnoticed by our nightly builds:
$ JAVA_HOME= mvn
Warning: JAVA_HOME environment variable is not set.
...
$ mvn
[INFO] Scanning for projects...
[INFO] ------------------------------------------------------------------------
[ERROR] BUILD FAILURE
...
Several major Linux distributions still do not provide Git >= 1.6.5.0 in
their stable package lists. Prior to commit 1173cc4a (Update Git
submodules with --recursive, 2011-02-22) CTest was able to use older Git
versions but simply silently failed to update submodules recursively.
Instead of failing with older Git versions preserve the status quo and
add a warning in the update log. Users testing projects with recursive
submodules may simply update to a Git new enough to support them.
Using sort results in a possibly-modified sorting
when all elements are "tied" - use stable_sort instead
to preserve the original ordering of tied elements.
To avoid name clashes.
Before this commit, name clashes can occur when invoking gcov on a
project that uses the same file name in different directories. The
--preserve-paths flag ensures all file names have a complete path,
avoiding name clashes.
Since commit e4beefeb (CTest: Do not munge UTF-8 output in XML files,
2009-12-08) we validate UTF-8 encoding of build and test output as it is
written to XML files. However, in cmCTestTestHandler::CleanTestOutput
we still processed test output one byte at a time and did not recognize
multi-byte UTF-8 characters. Presence of such characters caused early
truncation.
Teach CleanTestOutput to truncate test output at the limit but without
cutting it in the middle of a multi-byte encoding. Also, stop avoiding
truncation in the middle of an XML tag like "<MyElement>" because the
'<' and '>' will be properly escaped in the generated XML anyway.
667a90a Fix sentence break in add_test documentation
96309fc Make TestsWorkingDirectory test a C file
a4a5e37 Use iostream to make Borland happy
cfe53cd Fully specify the path to old-signature add_test
017d4e9 Group adding tests with its properties
561cc33 Only test the default cwd with Makefiles
d87bae7 Simplify the _default_cwd derivation
992c74f Use --><-- markers to denote the path
5249551 Flip slashes around on Windows
0a014da Add ctype.h include for toupper()
af12f83 Fix header includes for C++ and Visual Studio
5597aa2 Rename the project to match the test
9bf4165 Add tests for WORKING_DIRECTORY arg to add_test
42de5d0 Add WORKING_DIRECTORY argument to add_test
7679f9f Rename WorkingDirectory test
d95f817 Add the WORKING_DIRECTORY property to tests
A cycle exists when the DFS returns to the root node, not just when
multiple paths lead to the same node.
Inspired-By: Alexander Esilevich <aesilevich@pathscale.com>
Restore CMake 2.6 behavoir where the dart processing apparently was run from within the tests directory instead of the root of the build tree. This addresses issue reported on the VTK Dashboard where the Testing/Temporary directory is defined as ../../../Testing/Temporary but the DartProcessing is run with respect to the root of the build tree. This causes the regression/difference images to not be located by ctest and thus not uploaded to the dashboard.
This command allows a user to quickly see the list of all available
test labels. The labels are also printed in verbose show only mode,
alongside their corresponding tests.
Refactor how cmCTestMemCheckHandler computes the memory tester command
line options to avoid encoding them in a single string just to parse
them again. The EscapeSpaces uses backslahes to escape spaces on UNIX
platforms, so replace other calls to it in CTest that are used to create
human-readable strings with simple double-quoting.
a7319cf ctest_update: Run 'git submodule' at top level
7bf8dc1 ctest_update: Support ".git file" work trees
65cb72f ctest_update: Abort if Git FETCH_HEAD has no candidates
The git submodule porcelain must be executed from the top level of the
work tree. Use 'git rev-parse --show-cdup' to find the top level
relative to the source tree. This is better than searching up the tree
for .git ourselves because it will always work the same way Git does and
thus honors settings like GIT_DISCOVERY_ACROSS_FILESYSTEM.
Commit c3781efb (Support Git upstream branch rewrites, 2010-06-08)
assumed that ".git/FETCH_HEAD" exists inside the source tree. Fix the
implementation to handle a work tree using a ".git file" to link to its
repository. Use "git rev-parse --git-dir" to locate the real .git dir.
If .git/FETCH_HEAD provides no merge candidate do not attempt to update.
Also log FETCH_HEAD lines as we parse them since they are essentially
output from the git fetch command.
38c762c Merge 'remove-CTestTest3' into ctest-file-checksum
46df0b4 Activate retry code on any curl submit failure.
8705497 Checksum test should use CMAKE_TESTS_CDASH_SERVER
d0d1cdd Mock checksum failure output for old CDash versions
af5ef0c Testing for CTest checksum
86e81b5 CTest should resubmit in the checksum failed case
d6b7107 Fix subscript out of range crash
082c87e Cross-platform fixes for checksum/retry code
e525649 Checksums on CTest submit files, and retry timed out submissions.
Previously, we were only going into the retry block
for time out conditions. But a "could not connect"
response, or really any sort of curl failure, is
also a condition where we should retry the submit
if the user has requested a retry.
Fix niggly in ExternalProject that would inadvertently
create a log file of output when a command was explicitly
set to the empty string.
Also, increase the default value for showing context
around build errors to 10 lines, up from 6. Also add
the ability to customize this setting so that users
can put it back to 6 or even lower if they want, or
show more if they want.
Change types of local variables, or casting, or re-arrange
expressions to get rid of "conversion may alter value" warnings
as seen on recent dashboard submissions from londinium.kitware.
Git's diff-tree format has no '\n'-terminated blank line at the end of
its commit message body block if there are no diff lines. Instead the
message body is terminated by '\0' and there is no diff section. Teach
CTest to parse the format in this case.
Use 'git fetch' followed by 'git reset' to update the source tree. This
is better than 'git pull' because it can handle a rewritten upstream
branch and does not leave local modifications. After fetch, parse
FETCH_HEAD to find the merge head that 'git pull' would choose to track
the upstream branch. Then reset to the selected head.
In the normal fast-forward case the behavior remains unchanged.
However, now local modifications and commits will be erased, and
upstream rewrites are handled smoothly. This ensures that the upstream
branch is tested as expected.
Commit 67277bac (Teach ctest_update about Git submodules, 2010-05-04)
accidentally logged "git submodule update" with the prefixes "pull-out"
and "pull-err". Fix it to use "submodule-out" and "submodule-err"
instead.
Add the <Email>...</Email> element in Update.xml for each commit
reported. This field was defined by Dart but never really used.
Distributed version control systems use author name and email
instead of a user id, so now it makes sense to use this field.
Git does not automatically checkout the matching version of a submodule
when it checks out a new version of the parent project in the work tree.
If the submodule reference changed in the parent project then we were
reporting the submodule path as a local modification. Work around the
problem in ctest_update using "git submodule update" after "git pull".
For projects with no submodules this is a no-op. See issue #10662.
Also add a submodule to the test project for CTest.UpdateGIT to test the
work-around.
Report in <UpdateReturnStatus> when the update command line returns
non-zero. Otherwise the failure may be silently ignored since the error
was previously recorded only in the local log file.
The SaveRestoreEnvironment helper object makes sure that the
original environment is restored immediately after the
StartProcess call returns rather than waiting for the end
of the test. When tests are executed in parallel, there is
no guarantee about the ordering of EndTest calls relative
to StartTest calls. In fact, it would be odd for them to
be nested nicely. Therefore, to avoid the corruption of
the calling ctest's environment, the original environment
must be restored before ForkProcess returns.
Allow the user to set the CMake variable CTEST_COST_DATA_FILE, which will be used to store the cost data from test runs. If not set, defaults to the original location in the build tree Testing/Temporary dir.
We use 'git diff-index' to detect local modifications after pull. On
some filesystems the work tree timestamps of a few files may be dated
after the index, making them appear as locally modified. We address the
problem by using 'git update-index --refresh' to refresh the index and
avoid false local modifications.
Our internal path processing methods assume no trailing slashes, but bzr
adds trailing slashes to updated directories. This can lead to empty
entries in Update.xml files. We address the problem by stripping the
slashes as soon as they are parsed.
The improved text gives very clear information when either the CMakeLists.txt or CTestConfig.cmake file is missing. Hopefully, it makes it easier for those poor future souls who encounter these messages to solve their problems quickly.
If APPEND is given to ctest_start, it will read the tag from the current existing Testing/TAG file rather than creating a new one based on the current time stamp. This allows a developer to run several dashboard scripts in a row, all of which will share the same tag/stamp/buildid when they finally get submitted to CDash. Now you can split the running of build phases and test phases for the same dashboard row into multiple scripts.
Previously we produced commit times formatted like
1261403774 -0500
which is what the Git plumbing prints. Now we use a human-readable
format like
2009-12-21 15:28:06 -0500
which is still easy to machine-parse.
Previously we escaped quotes in <UpdateCommand>...</UpdateCommand>
values using '"'. This is not necessary because the value is in
xml CDATA and not an xml attribute.
A Subversion revision is unique across the entire repository, but work
trees typically correspond only to a subdirectory below the root path.
In order to specify the version of the source code that was tested,
CTest now submits a <SVNPath> element in Update.xml that specifies the
directory of the repository that corresponds to the work tree. In
combination with the revision number this uniquely specifies the tested
source. See issue #7541.