| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
| |
We see the 'ellipsized' analysis flag in the output now,
as we should.
|
|
|
|
|
| |
Run diff over the output to give a meaningful error if the
comparison fails.
|
|
|
|
|
|
|
|
|
| |
When we reapply non-shape attributes, we must take care to
not add any attributes that start in the middle of an
ellipsized run, or we end up with a blue, underlined ellipsis
if there is a link anywhere inside the ellipsized text.
https://bugzilla.gnome.org/show_bug.cgi?id=735732
|
|
|
|
|
|
|
|
|
|
|
| |
Without this, it is hard to determine if you are 'inside the
ellipsis'. We need to know this in GTK+, e.g. to skip link
processing.
I decided not to turn the flags into an enum at the time to
avoid possible language binding problems.
https://bugzilla.gnome.org/show_bug.cgi?id=735719
|
|
|
|
|
|
| |
This tests works similar to the markup parser tests -
we populate a layout with known data, and then compare
a dump of its properties with expected output.
|
|
|
|
| |
This reverts commit 19a252d9a3d63e7816f9a77bab65bd161675440a.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
This test is modeled after the GMarkup parser tests in GLib.
Test file with a name ending in .markup can be dumped in
tests/markups/, and tests/markup-parse can be used to generate
the expected output files which should be in the same location
and with the extension .markup replaced by .expected.
The test is also used as an installed test.
|
|
|
|
|
|
| |
Make sure we only try to create the directory for pango.aliases if that
file didn't exist. This cleans things up a little, and avoids weird
MSBuild batch errors.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
https://github.com/GNOME/pango/commit/61aeba6257ec7691a7a5222fb69aec3cc042435b#commitcomment-7215250
|
|
|
|
|
|
|
| |
See comments.
Bug 705727 - Incorrect rendering w/ Hangul syllable composition GSUB
https://bugzilla.gnome.org/show_bug.cgi?id=705727
|
| |
|
| |
|
|
|
|
|
| |
All these years, in the win32 backend, the y_offset was inverted.
Ie. no vertical glyph positioning worked, ever... Sigh.
|
|
|
|
|
| |
The shape.c layer is doing this for clumsy backends but warns.
Fix properly.
|
| |
|
|
|
|
|
|
|
| |
We should interpolate. Added TODO. Don't have a coretext build so
don't want to break it.
Bug 733764 - Support more / free font weights
|
|
|
|
|
|
| |
This makes the weight mapping a continuous, as ooposed to a step function.
Last piece of Bug 733764 - Support more / free font weights
|
|
|
|
| |
Previous commit was incomplete.
|
|
|
|
| |
Part of Bug 733764 - Support more / free font weights
|
| |
|
| |
|
|
|
|
|
| |
Inspired by:
https://github.com/Alexpux/MINGW-packages/tree/master/mingw-w64-pango
|
|
|
|
|
| |
Based on:
https://github.com/Alexpux/MINGW-packages/blob/master/mingw-w64-pango/0008-allow-backend-selection.mingw.patch
|
|
|
|
|
| |
From:
https://github.com/Alexpux/MINGW-packages/blob/master/mingw-w64-pango/0006-give-cc-to-gir-scanner.mingw.patch
|
| |
|
|
|
|
|
| |
Needed after this:
https://github.com/behdad/harfbuzz/commit/efe74214bbb68eaa3d7621e73869b5d58210107e
|
|
|
|
|
|
| |
Be a bit more carful when copying and updating the property sheets, so that
we don't accidently change fields with '10' in them that are actually not
indicating the Visual Studio version.
|
|
|
|
|
|
|
|
|
|
|
| |
As the Visual Studio 2012/2013 are only slightly different from the Visual
Studio 2010 projects, we can provide support for them by using scripts to
copy the Visual Studio 2010 projects, and update the specific parts as
necessary. Thus, there would be little maintenance overhead for these as
only the 2010 projects need to be kept up-to-date as a result.
This might change when we do get the stack working with WinRT/Metro, but
that's going to be another totally different issue.
|
|
|
|
|
|
|
|
|
|
| |
g_once_init_leave() does not work on NULL, and just returns... and
all future calls to g_once_init_enter() will block forever.
Happens on Android, so let's just fall back to the C locale if there's
nothing else we can do.
https://bugzilla.gnome.org/show_bug.cgi?id=732276
|
| |
|
| |
|
|
|
|
|
|
| |
This reverts commit e728c4c3659cc0ead406344b4cf98c0676cb769e.
See previous commit for explanation.
|
|
|
|
|
|
| |
This reverts commit 4901fb74b3c1137d13a6b7e4e8784577c1816e47.
The removed API was added back to HarfBuzz. So, no need for this.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
There has been an API change in HB:
-hb_ot_layout_language_get_required_feature_index
+hb_ot_layout_language_get_required_feature
Commit: da132937989acb4d8ca9bd41c79f98750e7dda30
So we need to bump the dependency.
|
|
|
|
| |
Missing colon broke the g-ir-scanner.
|
|
|
|
|
|
| |
A while ago, we added a weight of 1000 to the enumeration, but
neglected to update the allowed values in the documentation.
Fix that.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently, due to the way that Visual Studio 2010+ projects are handled,
the "install" project does not re-build upon changes to the sources, as it
does not believe that its dependencies have changed, although the changed
sources are automatically recompiled. This means that if a part or more
of the solution does not build, or if the sources need some other fixes
or enhancements, the up-to-date build is not copied automatically, which
can be misleading.
Improve on the situation by forcing the "install" project to trigger its
rebuild, so that the updated binaries can be copied. This does trigger an
MSBuild warning, but having that warning is way better than not having an
up-to-date build, especially during testing and development.
|
|
|
|
|
|
| |
Follow the prevailing practice and put the installed tests
into $libdir/installed-tests/$PACKAGE instead of
$pkglibdir/installed-tests.
|
|
|
|
| |
Oops
|
| |
|
| |
|