We need to remove the directory containing sh.exe from the PATH as
otherwise mingw32-make seems to use it and not cmd.exe for executing the
commands in the makefile, but the wrong directory was being removed in
appveyor.yml, having no effect.
Fix this and undo 5f2ccde269 which
compensated for it by (almost) clearing the PATH entirely which, in
turn, broke one of our unit tests (and could break other things in the
future, as having an empty PATH is rather unexpected).
Also use the full path for head program now that its directory is not in
the PATH any more. Alternative could be to use PowerShell, but while
powershell -Command "'gcc','mingw32-make','ar'|%%{ iex \"$_ --version\"|select -first 1 }"
does work, it looks really ugly, so avoid it for now.
See https://github.com/wxWidgets/wxWidgets/pull/142
See #17274.
Apparently a locale can be available at MSW level, so that
wxLocale::IsAvailable() returns true, but not supported by the MSVC CRT,
so that constructing the corresponding wxXLocale fails, which resulted
in wxXLocale unit test failures.
Fix them by checking that wxXLocale can be constructed directly instead
of using wxLocale::IsAvailable() as a proxy.
This is not ideal and perhaps wxLocale::IsAvailable() should check that
the locale is supported in wxXLocale too, but should at least allow unit
tests to pass on AppVeyor for now.
Simplify the test by using a single function instead of all the
machinery inherited from CppUnit.
Also provide more information in case of test failure.
Standard/Unix format strings are intentionally used when using these
compilers/environments (see d62c535d4c for
the former), so don't check for Windows/MSVC format strings in the test.
Allow wxPrintf("%1$s %1$s", "foo") to work.
Take into account the possibility that the number of format specifiers
and the number of actual arguments can be different.
Closes#9367.
There is no need to rebuild everything if only the documentation files
have changed.
Unfortunately there doesn't seem to be any way to do the same thing for
Travis CI yet, see https://github.com/travis-ci/travis-ci/issues/6301
These file lists contained files that were not low level (meaning "used
by wxQt/Univ" in this context) at all and it doesn't make much sense to
think about Qt-based wxUniv port anyhow, so instead of painstakingly
separating them in low level part and the rest, just rename the
variables to be less confusing.
Only define it unconditionally for wxMSW and wxOSX, not wxGTK where we
need to test that the required versions of FontConfig and Pango Fc are
available.
Also do all wxUSE_PRIVATE_FONTS-related checks in the same place, to
make things more clear and avoid similar blunders in the future.
Handle this feature as all the other ones and provide a configure switch
and a setup.h option to disable it if necessary, as it may be desirable
to do it, especially under Linux, to avoid extra dependency on pangoft2
if this functionality is unnecessary.
Don't mix "size_t" and "unsigned" and don't use "%lu" for printing
"size_t" values as this doesn't work under Win64 where sizeof(size_t) is
8 while sizeof(unsigned long) is still just 4.
Rewrite an assert using "||" as 2 asserts to allow it compile with Catch
(this wasn't detected before as this code is only compiled when using
our own vsnprintf() implementation, which is usually not the case).
This doesn't work correctly in Catch after reverting the commit that
broke compilation with MSVC 9, so we have to live with explicitly using
the values of the correct type.
ftp.wxwidgets.org seems to have gone, breaking the existing test (thanks
chris for all the fish...), so don't hardcode this URL in the test and,
without it, don't run the test by default any more and require
specifying a working FTP host (as well as an existing directory and a
file on it) when running it manually.
Rewrite the test using normal functions instead of Java-like structure
imposed by CppUnit to make it simple to skip it.
Unfortunately there is no reasonable way to implement this function for all
glibc versions as the information we need is stored in the private
_GOptionGroup struct whose layout has already changed once (in 2.44) and could
change again, so we can't rely on it.
We really need a g_option_group_get_entries() in glib itself, but the request
to add it at http://bugzilla.gnome.org/show_bug.cgi?id=431021 hasn't been
touched since 10 years, so it seems unlikely to happen.
See 99367a1530 (commitcomment-25789514)
There is no shrink_to_fit() in wxVector in this case.
Arguably, we shouldn't be building wxVector unit tests in STL build at
all as there is no point in testing the standard class, but OTOH it
could be useful for checking that the tests themselves are correct, so
keep them for now.
Output the maximal difference between the differing images pixels to
show just how far are they from each other, exactly. And show the image
file name as well, for convenience.
Also run all checks instead of stopping after the first failing one.
wxInvalidSize is a documented return value for wxDir::GetTotalSize(),
yet it was not available by including just wx/dir.h as it was declared
in wx/filename.h only.
Fix this by declaring it in wx/dir.h too.
Closes https://github.com/wxWidgets/wxWidgets/pull/609
When column resizing is finished, after HDN_ENDTRACK notification there is
also sent one (and last) HDN_ITEMCHANGING notification. We have to skip it
to prevent from sending EVT_HEADER_RESIZING after EVT_HEADER_END_RESIZE
because EVT_HEADER_END_RESIZE should be really the last one event
in the sequence of resizing events (like it's assumed in wxGrid).
Closes#16390.
Ever since the changes of 544c4a3bde
(almost 14 years ago), playing mono WAVs with wxSound completely failed
if setting the sound device to mono using SNDCTL_DSP_STEREO ioctl
failed. This doesn't look like a wise thing to do, so don't consider
this as a fatal failure, but just play mono as stereo (and even possibly
stereo as mono) instead.
This fixes the sound sample being broken out of the box on many (all?)
Linux systems.
The code was completely broken as it locked the mutex in only one thread
and then tried to unlock it in another one, which made no sense, didn't
protect against anything and resulted in errors and assert failures.
Fix this by locking and unlocking the mutex in both threads before
accessing shared data or playing sound.
Closes#17990.
