Files
libgit2/tests
Edward Thomson b70dd12706 packbuilder: adjust nondeterministic tests
The packbuilder tests should be deterministic. This comment suggests
that they are:

```
/*
 * By default, packfiles are created with only one thread.
 * Therefore we can predict the object ordering and make sure
 * we create exactly the same pack as git.git does when *not*
 * reusing existing deltas (as libgit2).
 *
 * $ cd tests/resources/testrepo.git
 * $ git rev-list --objects HEAD | \
 *  git pack-objects -q --no-reuse-delta --threads=1 pack
 * $ sha1sum pack-7f5fa362c664d68ba7221259be1cbd187434b2f0.pack
 * 5d410bdf97cf896f9007681b92868471d636954b
 *
 */
```

but it is disappointingly incorrect. As is evidenced by the fact that --
at least on _my_ machine -- that command does not actually produce that
output any longer.

Variations in things like the which compression library is actually
used, and its defaults, mean that we cannot accurately predict or
enforce the bytes in the packfile from one system to another.

Adjust the tests so that they do not believe that they should enforce
the bytes in the packfile. This allows broader compatibility with
zlib-compatible compression libraries, or other compression levels.
2024-03-16 12:19:16 +00:00
..
2022-02-27 23:44:19 -05:00

libgit2 tests

These are the unit and integration tests for the libgit2 projects.

  • benchmarks These are benchmark tests that excercise the CLI.
  • clar
    This is clar the common test framework.
  • headertest
    This is a simple project that ensures that our public headers are compatible with extremely strict compilation options.
  • libgit2
    These tests exercise the core git functionality in libgit2 itself.
  • resources
    These are the resources for the tests, including files and git repositories.
  • util
    These are tests of the common utility library.

Writing tests for libgit2

libgit2 uses the clar test framework, a C testing framework.

The best resources for learning clar are clar itself and the existing tests within libgit2. In general:

  • If you place a .c file into a test directory, it is eligible to contain test cases.

  • The function name for your test is important; test function names begin with test_, followed by the folder path (underscore separated), two underscores as a delimiter, then the test name. For example, a file merge/analysis.c may contain a test uptodate:

    void test_merge_analysis__uptodate(void)
    {
      ...
    }
    
  • You can run an individual test by passing -s to the test runner. Tests are referred to by their function names; for example, the function test_merge_analysis__uptodate is referred to as merge::analysis::uptodate. To run only that function you can use the -s option on the test runner:

    libgit2_tests -smerge::analysis::uptodate
    

Memory leak checking

These are automatically run as part of CI, but if you want to check locally:

Linux

Uses valgrind:

$ cmake -DBUILD_TESTS=ON -DVALGRIND=ON ..
$ cmake --build .
$ valgrind --leak-check=full --show-reachable=yes --num-callers=50 --suppressions=../libgit2_tests.supp \
  ./libgit2_tests

macOS

Uses leaks, which requires XCode installed:

$ MallocStackLogging=1 MallocScribble=1 MallocLogFile=/dev/null CLAR_AT_EXIT="leaks -quiet \$PPID" \
  ./libgit2_tests

Windows

Build with the WIN32_LEAKCHECK option:

$ cmake -DBUILD_TESTS=ON -DWIN32_LEAKCHECK=ON ..
$ cmake --build .
$ ./libgit2_tests