This test keeps failing just because the buildbot machine are too heavily
loaded. This is not really an error, so just avoid running this test there.
This is the backport of f641dfd7 from master.
This function is available on OS X since 10.5, which is the minimum
supported version of wx 3.0.
Fixes Xcode 6.3 warning triggered by checking availability of a symbol
that wasn't marked as weak_import.
This test can't work when the in-memory representation is UTF-16, as we can't
convert surrogates to anything else in this case.
This fixes the unit tests broken since r76622, see #16298.
This is the backport of 0bb7676889 from master.
Give error if changing directory when wxFD_CHANGE_DIR or wxDIRP_CHANGE_DIR
flags are used, this is a good idea generally speaking and it also avoids
warnings about not checking chdir() return value that we got under Ubuntu.
This is the backport of 3f6741a770 from master
with the changes of cb06ecb9eb and
de27f4b08f too.
Explicitly ignore errors if we can't read from the associated FD.
This is mostly done to avoid -Wunused-result warning about not checking read()
return value under Ubuntu.
This is the backport of a74d2faf3a from master.
Relax the check in RestartBug() test too, the time can be slightly smaller
than expected (by ~2ms on disc-xp machine).
This is the backport of 7af8598903 from master.
This fixes the warning for MSVC, which warned about "forcing value to bool
'true' or 'false'", without introducing one for gcc which would complain if we
compared the value, of compatibility_iterator type, with NULL.
This backports the changes of r75445 and r75653 from master.
Try to find out why this test fails in the OS X buildbot builds.
Use macros to ensure that all checks provide information about the bitmap (or
animation) type they fail for, without having to repeat the same code for
doing it many times.
As with the previous commit, make an exception to the usual rule of not
changing 3rd party code to get rid of the annoying warnings appearing in each
buildbot libtiff rebuild.
This warning appears in each buildbot build (which is wrong on its own as
libtiff configure shouldn't forcefully regenerate its output config.h every
time neither -- but currently it does) and is annoying as it distracts from
the other warnings, so suppress it even if usually we don't do it for the
warnings in 3rd party code.
At least under GTK SetMaxLength(), tested by this test, can't be used with
multiline text controls at all, so running it fails. Keep the test for MSW
only where SetMaxLength() works with controls of all kinds.
Commit 3e67b1ef68 broke the test suite in
non-GUI case, we still need to override OnRun() then to run the tests.
Also make it more clear that m_exitcode is only used in the GUI case.
Return the program exit code from the (overridden) OnRun(), not OnExit(),
doing it from the latter didn't work since many years.
The failures in the GUI tests will be detected by buildbot now.
Previously it only handled wxALIGN_CENTRE, also the documentation switched to
wxALIGN_CENTRE_HORIZONTAL in 72e88d8ea3 and
preceding commits.
See #16943.
We can't rely on the program waking up quickly enough when running on the
(loaded) buildbot machines, better to skip this test than to have spurious
failures.
This is a backport of 5de9b53902 from trunk.
Handlers for this event expect to get wxSpinEvent but wxMSW and wxGTK1 sent an
object of base wxCommandEvent class which resulted in invalid memory access in
the handlers when using the derived class methods such as Veto().
See #16948.
Allocate our own exclusive copy of the data before modifying it, otherwise
calling ClearAlpha() would clear alpha not only for the image it was called on
but also for all the other images sharing data with it.
Don't mix signed/unsigned integers in arithmetic operations when normalizing
wxDC scale factors: variable holding GCD value should be of the same type as
variables holding devExt and logExt to avoid wrong results of /= operation
when dividend is a negative value.
Closes#16908.
This is useless at best and resulted in GTK+ warnings because we ended up
(somehow -- is there another bug lurking here?) with negative items height in
this case.
Closes#16862.
When the TZ offset specified in the RFC 822 string was equal to the local TZ
offset but the date fell in the DST period, the result was one hour off.
Fix this by converting the date to the UTC explicitly, and then converting it
back to the local TZ to ensure that the DST is taken into effect.
See #15370.
This was broken due to a wrong cast from WPARAM (i.e. 32 bit unsigned int in
Win32) value containing the ID to WORD (i.e. 16 bit unsigned short) which
truncated the ID value.
Notice that MSDN documents the requirement for the IDs to be positive, but
negative IDs seem to work, at least with Windows 7, after this fix.
Closes#16880.
Restore original list of wxPG-specific colours. wxPG-specific colours are added to the colour database when wxColourProperty is created (already existing colours are not overwritten by wxPG ones).
This is the backport of r76893 and r76894.
See #15597.
There is no need to check if referenced array with values is valid since "reference cannot be bound to dereferenced null pointer in well-defined C++ code". Moreover, conditional call of wxPGChoices::Add() methods (one with explicit parameter and one with default one) is not necessary.
This is the backport of r78343.
See #16759 (http://trac.wxwidgets.org/ticket/16759).
