| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
| |
We were passing a format specifier that starts
with ' ', which some implementations of that API
don't like.
This change removes an extraneous space from the
output. Update affected tests.
Fixes: #659
|
|
|
|
| |
Just use g_object_ref/unref.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
If gap is zero, don't limit the inserted attributes;
that does not make sense. Spell out the different
use cases in the docs.
Testcase included.
Fixes: #653
|
|\
| |
| |
| |
| | |
Add a test for letterspacing
See merge request GNOME/pango!570
|
| |
| |
| |
| |
| | |
This tests that we turn off ligatures when letterspacing
is enabled.
|
|/
|
|
|
| |
Add API to extract the slant ratio from a PangoMatrix.
Tests included.
|
|
|
|
|
| |
Best to compare precise sizes and take rounding
out of the equation.
|
|
|
|
| |
Fixes: #648
|
|
|
|
|
|
|
| |
The Microsoft paper I was citing actually recommends
1/4 em for ASCII space. Make it so...
Update affected tests.
|
|
|
|
|
|
|
|
|
|
| |
Follow the Microsoft Typography recommendations from
https://docs.microsoft.com/en-us/typography/develop/character-design-standards/whitespace
and use half of an em for missing space.
Update affected tests.
Fixes: #647
|
|
|
|
|
| |
This test verifies the fixes for the
situation described in #646.
|
|
|
|
|
|
| |
Oops, we were not serializing the height at all.
Update all affected tests.
|
|
|
|
|
|
|
|
|
|
|
| |
When coming up with a width for a missing space,
we were just measuring the hex digits. But in
subsetted fonts, such as the ones we use for ci,
hex digits might well be missing. Take that into
account and still provide some nonzero width
for space.
Update affected tests.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It is a difficult dance between HarfBuzz and pango
which glyphs we let hb synthesize, and which ones
we need to fake ourselves.
HarfBuzz knows how to create other spaces from 0x20,
so if the font is lacking spaces, we want to tell
HarfBuzz that we don't have that glyph *except* for
0x20 where we need to fake one, and provide HarfBuzz
with a suitable width.
Update affected tests.
|
|
|
|
|
|
|
|
|
| |
This reverts commit ccb651dd2a876a4f4a4cb9351f05332173e709ba.
This broke vertical text rendering, see
https://gitlab.com/inkscape/inkscape/-/issues/2950.
Update affected tests and add a vertical test.
|
|
|
|
|
|
|
|
|
|
|
| |
HarfBuzz knows how to synthesize spaces, we just
need to tickle it in the right way.
Test included.
Update affected tests.
Fixes: #641
|
|
|
|
| |
This tests the fix in the previous commit.
|
| |
|
|
|
|
|
|
|
|
|
| |
We can't break in a tab run.
This fixes an assertion found by afl.
Testcase included.
Fixes: #638
|
|
|
|
|
|
|
|
|
|
| |
We an only use the line_width - remaining_width
shortcut if we are actually maintaining remaining_width,
i.e. not if we don't wrap lines.
Testcase included.
Fixes: #635
|
|
|
|
| |
Update affected test outputs.
|
|
|
|
|
|
|
|
|
|
| |
When we uninsert the current tab run, we need to reset
our tab state to prevent last_tab->glyphs to become a
dangling pointer.
Test included.
Fixes: #634
|
|
|
|
|
| |
This will make the font serialization useful
in GTK, where we serialize fonts in node files.
|
| |
|
| |
|
|
|
|
|
| |
Don't calculate line width the hard way when
we can do it much easier.
|
| |
|
|
|
|
| |
Use newlines to separate tabs, rather than spaces.
|
|
|
|
|
|
|
| |
This is useful for PANGO_TAB_DECIMAL.
Implement this in PangoLayout, in the
serializer, and update tests.
|
|
|
|
|
|
|
|
|
| |
Serialize tab alignment in addition to tab
position. We still accept the old format
and default alignment to left.
Update testserialize and layout tests
for this.
|
|
|
|
| |
We want to use this without initializing tests.
|
| |
|
|
|
|
|
|
|
|
| |
Even when we show ignorables, we still want to
ignore variation selectors and other ignorables
that we don't have nicks for.
Test included.
|
|
|
|
|
|
|
|
|
|
|
| |
Without this, we end up with line height being
zero in gravity north or west. The visible symptom
is that wrapped lines are drawn on top of each
other with line-height != 0.
Affected tests have been updated.
Fixes: #631
|
|
|
|
|
| |
Now that we have fonts in git, we can
reliably test this.
|
|
|
|
|
|
|
|
| |
When setting variations to "", font descriptions
would add an ugly useless " @" at the end of their
serialization. Avoid that.
Test included.
|
| |
|
|
|
|
| |
We need to copy the scale hack for bitmaps.
|
| |
|
| |
|
|\
| |
| |
| |
| |
| |
| | |
Hinted metrics fixes
Closes #626
See merge request GNOME/pango!518
|
| |
| |
| |
| |
| |
| | |
This would have caught the issue fixed in the previous commit, because
the default font choice of "serif 12" happened to not be affected by it.
Most other fonts or font sizes however were.
|
| |
| |
| |
| | |
It should work there, but it doesn't.
|
| |
| |
| |
| | |
This makes more sense.
|
| |
| |
| |
| |
| |
| |
| | |
Our new, more precise font checks rely on
data from fontconfig to identify fonts
uniquely. So make sure that we use a custom
fc fontmap, if we have one.
|
| |
| |
| |
| |
| |
| |
| |
| | |
Otherwise editing the output in an editor
is likely to make the data not match anymore,
since most editors will add a final newline.
Update all tests
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Add Cantarell, DejaVu Sans and a subset of Noto Color Emoji
to git, and make test-layout use just those fonts. This
should help with keeping the ci from breaking due to different
system fonts.
Update tests.
|
| |
| |
| |
| | |
Not used anymore.
|
| |
| |
| |
| |
| | |
These were produced with the help of
http://gitlab.gnome.org/matthiasc/layout-editor
|