queue->tail->next cannot be non-NULL, as pushing onto the end of the
queue is handled by the call to g_queue_push_tail_link() above.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
This makes it a bit clearer that we expect the queue to be empty as a
result of calling g_queue_clear_full(), rather than as a result of any
of the later cleanup.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
This makes the g_list_insert_before() follow more closely the guidelines
for GLib, which is to avoid implicit pointer boolean value and to prefer
for over while to improve readability.
This adds two new helpers that allow for inserting pre-allocated GList
elements to the queue similar to existing helpers. This may be advantagous
in some situations such as statically allocated GList elements.
Use Meson’s support for running scan-build (Clang’s static analyser)
against the build, so that we get static analysis of each pipeline. Add
it in a new pipeline stage, after code coverage, so that we don’t waste
resources on it unless the unit tests pass — a static analysis build
takes perhaps 10× as long as a normal GCC build.
https://mesonbuild.com/howtox.html#use-clang-static-analyzer
Currently, the static analysis results are uploaded as artifacts, but
the job will always succeed (regardless of whether there are any bugs
found in the analysis).
Currently, a large number of reports are outputted by the analyser,
which need to be fixed before we can gate the pipeline on it.
Furthermore, in order to get scan-build to exit with a non-zero status
if any bugs are found, we need to depend on Meson ≥ 0.49.0, which
contains the fix https://github.com/mesonbuild/meson/issues/4334,
allowing us to add the following to .gitlab-ci.yml:
```
variables:
# Exit with a non-zero status if any bugs are found
SCANBUILD: "scan-build --status-bugs"
```
Signed-off-by: Philip Withnall <withnall@endlessm.com>
Fedora 27 was EOL'ed on November 2018.
We move to Fedora 28 because the Android NDK requires Python 2 and
probably other things, and bumping to Fedora 29 is going to be more
painful.
The Fedora image we use contains MinGW bits that ought to go into
their own Docker container. This avoids having a massive Docker image
that gloms everything and is harder to update.
While we're splitting off, we can also update to Fedora 29, as we can
rely on Fedora packagers doing their job and ensuring that the MinGW
cross-compilation toolchain still works.
The Fedora image we use contains Android bits that ought to go into
their own Docker container. This avoids having a massive Docker image
that gloms everything and is harder to update.
We reuse the same Docker image we used for Fedora, to avoid regressing.
The GHashTable code ignores the duplicated-branches GCC warning, but we
need to do a compiler and version check, as either non-GCC compatible
compilers, or older versions of GCC will warn about the unknown pragma
or diagnostic.
If we don't do this while turning warnings into error, we're going to
fail the build unnecessarily.
Apparently, the documentation of g_strcanon() was not really cristal
clear, so this new code sample try to make it clear the fact that we
are working on the given string and not a copy. Moreover, it provides
a way to keep the original string at once.
Fix#29
Using --GTestSkipCount 0 is the same as omitting it. A skip count
greater than the number of tests is the same as equalling the number
of tests: they are all skipped.
Signed-off-by: Simon McVittie <smcv@collabora.com>
The -p option is documented, and can be used to select and repeat
test-cases. This is particularly useful when debugging a single
failure among a large number of test-cases, or when debugging a
test-case that you suspect influences another test-case by leaking
global state.
Until now, -p was only supported with GLib's default (GLib-specific)
textual output format, and not with the standardized TAP format that
we are now encouraging. If we are considering making TAP the new default
(see glib#1619) it should get feature-equivalence with the current
default.
Because -p allows test-cases to be re-ordered and repeated, and an entry
in the test_paths list can match any number of test-cases (including
zero), we don't know ahead of time how many test-cases we are going to
run. TAP allows the "plan" to be deferred to the end, exactly to support
situations like this.
Signed-off-by: Simon McVittie <smcv@collabora.com>
The undocumented --GTestSkipCount option is internal to the deprecated
gtester tool and rather obscure, but it's straightforward to support
by making G_TEST_LOG_SKIP_CASE produce TAP output similar to what already
happened when we emitted G_TEST_LOG_STOP_CASE with result
G_TEST_RUN_SKIPPED. I might as well do that while I'm looking at the
interaction between the --tap, -p and -s options.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Using commands:
glib/gen-unicode-tables.pl -both 12.0.0 path/to/UCD
tests/gen-casefold-txt.py 12.0.0 path/to/UCD/CaseFolding.txt \
> tests/casefold.txt
tests/gen-casemap-txt.py 12.0.0 path/to/UCD/UnicodeData.txt \
path/to/UCD/SpecialCasing.txt > tests/casemap.txt
plus some manual additions of the new G_UNICODE_SCRIPT_* symbols to
gunicode.h, guniprop.c and glib/tests/unicode.c.
Using UCD release https://www.unicode.org/Public/zipped/12.0.0/UCD.zip.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
Fixes: #1713
The network-available property can be asserted by querying the NMState
describing the current overval network state, instead of the
NMConnectivityState. The advantage of the NMState is that is reflects
immediately the network state modification, while the connectivity
state is tested at a fixed frequency.
