1. 27 Feb, 2019 1 commit
  2. 13 Feb, 2019 3 commits
    • Iain Lane's avatar
      installed tests: Allow tests to set environment variables · 07a1a803
      Iain Lane authored
      It's necessary sometimes for installed tests to be able to run with a
      custom environment. For example, the gsocketclient-slow test requires an
      LD_PRELOADed library to provide a slow connect() (this is to be added in
      a followup commit).
      Introduce a variable `@env@` into the installed test template, which we
      can override as necessary when generating `.test` files, to run tests
      prefixed with `/usr/bin/env <LIST OF VARIABLES>`.
      As the only test that requires this currently lives in `gio/tests/`, we
      are only hooking this up for that directory right now. If other tests in
      future require this treatment, then the support can be extended at that
    • LRN's avatar
      GLib test: Adjust to follow the implementation · 1e74b587
      LRN authored
      Canonicalization converts slashes to backslashes on Windows (most
      of the time). This is a horrible design decision, but that's what
      it does, and it's too late to change that. The test shouldn't expect
      anything else.
    • LRN's avatar
      GDate test: Disable Julian epoch test on Windows · 3ffa7ed0
      LRN authored
      Windows uses FILETIME, which starts counting from 1st Jan of year 1601 and,
      unlike time_t, can't be negative, so Windows simply has no way
      to do timestamp-math for dates before then. SYSTEMTIME (an equivalent
      of struct tm) can, obviously, represent almost arbitrary date starting
      from 1st Jan of year 0 (it's unsigned...), but GetDateFormatW() converts it
      to FILETIME at some point in its implementation, and fails.
      Unless the whole strftime() implementation of GDate is replaced by
      something that doesn't rely on WinAPI, this part of the test will
      never pass.
  3. 15 Jan, 2019 1 commit
    • Philip Withnall's avatar
      build: Drop autotools support · b3efef5b
      Philip Withnall authored
      So long, and thanks for everything. We’re a Meson-only shop now.
      glib-2-58 will remain the last stable GLib release series which is
      buildable using autotools.
      We continue to install autoconf macros for autotools-using projects
      which depend on GLib; they are stable API.
      Signed-off-by: Philip Withnall's avatarPhilip Withnall <withnall@endlessm.com>
  4. 26 Dec, 2018 1 commit
    • Vasily Galkin's avatar
      gspawn, tests: extend spawn_test, run it on win32 · 50cb4f22
      Vasily Galkin authored
      The spawn_test is enabled on win32 meson build, both msys and msvc.
      Some modifications to make it useful for auto-testing on win32:
      - use own argv0 to find helper win32-specific subprogram
      - helper subprogram and conditions changed, so testing is fully
      automated instead of manually checking contents of some MessageBoxes
      Redirection test checks "sort" output for locale-independent string
      instead of relying on "netstat" locale-dependent string.
      Also with "sort" it become usable on unix, so enabled there too.
      Currently this fails on win32 with coverage since
      some coverage-realted error output from gpawn-win32-helper
      is unexpectedly treated as executed subprocess output.
      Added test checking "sort" with error-only redirection. This also fails
      on win32 by now, due to a typo in gspawn-win32.c (checks for stdout
      redirection instead of stderr)
  5. 26 Nov, 2018 2 commits
    • Will Thompson's avatar
    • Will Thompson's avatar
      gio-test: fix leak & maybe-uninitialized warning · b45d911c
      Will Thompson authored
      GCC 8 on F29 complains:
      ../tests/gio-test.c: In function ‘main’:
      ../tests/gio-test.c:375:7: warning: ‘id’ may be used uninitialized in this function [-Wmaybe-uninitialized]
             g_free (id);
      In the normal case, when run without arguments, 'id' will be assigned
      exactly once, so all is fine. If run with argument '0', 'id' will never
      be assigned, so the warning is legit; but in that case the test program
      will never exit. If run with any argument greater than 1, 'id' will be
      assigned more than once but only the last incarnation will be freed.
      Tweak the scope of the variable to match its use, and arrange for it to
      be freed when its watch is destroyed.
  6. 06 Nov, 2018 1 commit
    • Simon McVittie's avatar
      closures test: Avoid timeout on ARM64 CPUs · cb98e373
      Simon McVittie authored
      Closures use a 16-bit atomic reference count, which is really slow
      on certain ARM64 CPUs such as the Cortex-A57 (#1316). This is
      non-trivial to solve, since the public struct field cannot be enlarged
      to 32-bit while preserving ABI, and 16-bit atomic operations would be new
      (and rather niche) API.
      Until this can be solved properly (hopefully in GLib 2.59.x), cut down
      the number of signal emission cycles and bump up the timeout in the
      Meson build system, so that builds won't time out. We can't just take
      another zero off the number of signal emission cycles, as was done in the
      original version of this patch in Debian, because if we do that it can
      result in test failures when the main thread starves the other threads.
      ARM64 CPUs are backwards-compatible with 32-bit ARM, and the same
      slowdown can be seen when building and testing 32-bit code on these
      CPUs, so check for both 32- and 64-bit ARM.
      Bug-Debian: https://bugs.debian.org/880883Co-authored-by: Iain Lane's avatarIain Lane <laney@debian.org>
      Signed-off-by: Simon McVittie's avatarSimon McVittie <smcv@debian.org>
  7. 04 Nov, 2018 2 commits
  8. 01 Nov, 2018 2 commits
  9. 31 Oct, 2018 1 commit
  10. 29 Oct, 2018 1 commit
    • Tomasz Miąsko's avatar
      mainloop-test: Fix race conditions · 88f36a1d
      Tomasz Miąsko authored
      * Wait for adder threads before deallocating crawler_array and
        context_array to avoid use after-free and data race.
      * Handle spurious wakeups around g_cond_wait.
      * Avoid starting recurser_idle without context.
      Fixes issue #1530.
  11. 27 Sep, 2018 2 commits
  12. 26 Sep, 2018 1 commit
  13. 25 Sep, 2018 1 commit
  14. 24 Sep, 2018 2 commits
  15. 23 Sep, 2018 1 commit
  16. 21 Sep, 2018 9 commits
  17. 20 Sep, 2018 1 commit
  18. 30 Jul, 2018 1 commit
  19. 18 Jul, 2018 1 commit
  20. 16 Jul, 2018 1 commit
  21. 06 Jul, 2018 1 commit
  22. 29 Jun, 2018 1 commit
  23. 14 Jun, 2018 1 commit
    • Emmanuele Bassi's avatar
      Deprecate g_type_class_add_private() · 7e5db31d
      Emmanuele Bassi authored
      It's been 4 years and 8 development cycles since we introduced
      G_ADD_PRIVATE and offset-based private data access. It is now
      time to finally deprecate the old mechanism.
      Closes: #699
  24. 12 Jun, 2018 1 commit
  25. 10 Jun, 2018 1 commit
    • Emmanuele Bassi's avatar
      Classify the tests · 16d1a3d2
      Emmanuele Bassi authored
      Meson has the ability to classify tests according to "suites", a list of
      tags. This is especially useful when we want to run specific sets of
      tests — e.g. only GLib's tests — instead of the whole test suite. It
      also allows us to classify special tests, like "slow" ones, so that we
      can only run them when needed.