0275151490
[paint] Align deptch/edge count conditions across two backends
2023-01-09 12:27:09 -07:00
5f976d86a7
test-paint: Use %.3g for results
...
This produces more readable output.
2023-01-09 14:24:32 -05:00
ec78a486bf
Run paint tests at upem scale
...
This avoids problems with rouding.
2023-01-09 14:03:58 -05:00
839f4b6494
[ft] Fix slanting clip box
2023-01-09 12:01:46 -07:00
30adbc22d9
hb-font: Fix scale_glyph_extents() again
...
And better fix this time.
2023-01-09 11:54:44 -07:00
6c1a4bed4a
Better rounding clip boxes
2023-01-09 11:37:39 -07:00
a085efa699
test-paint: Fix font setup for ft
...
We must call hb_ft_font_set_funcs after
setting the font scale.
2023-01-09 13:11:16 -05:00
26f713a7f5
paint-test: Use larger scales
...
A scale of 20 is noise, so use 1000.
2023-01-09 12:54:06 -05:00
e886b6b8a6
[test-draw] Use a larger scale
...
A scale of 20 is in the noise category for us. Using a larger
scale makes the test pass.
2023-01-09 10:31:28 -07:00
347910fd4c
Add a draw test
...
This test compares output between ft and ot
font funcs.
2023-01-09 10:31:28 -07:00
f46dcf147b
[paint/COLR] Fix clip transform
...
Sigh. So complicated.
2023-01-08 14:35:42 -07:00
42047070dd
test-paint: More helpful output
...
No need to print ASCII chars as hex.
2023-01-08 16:08:16 -05:00
f283d4d366
[paint] Try to adjust both renderers to use same clip order
2023-01-08 12:42:34 -07:00
3fd6c0d97a
[test-paint] Remove unused FT_Library
2023-01-08 12:30:26 -07:00
28be4f8805
[test-paint] Actually run against hb-ft
2023-01-08 12:19:41 -07:00
ed7d0234e1
test-paint: Limit the precision of clip boxes
...
Update expected test results.
2023-01-08 11:57:43 -07:00
5d94eb61b8
[hb-cairo] Fix hb_cairo_glyphs_from_buffer() when utf8 missing
...
Fixes https://github.com/harfbuzz/harfbuzz/issues/4016
2023-01-08 11:56:01 -07:00
10390ec5c6
Update expected test results
2023-01-08 09:06:03 -05:00
f3ce137420
test-paint: Be more flexible for expected results
...
When generating the expected output with GENERATE_DATA=1,
Glib's test framework puts out some comments at the top
of the file. Ignore them when comparing the expected
output. This makes it possible to directly use the output
of
GENERATE_DATA=1 ./test-paint -p TESTCASE
as expected result for TESTCASE.
2023-01-08 09:01:31 -05:00
5bd6fc1ace
Comment
2023-01-07 16:32:50 -07:00
432afa9dff
[paint] Fix paint_extents usage
...
It was broken all this time :(.
The two backends do this slightly differently...
2023-01-07 16:25:30 -07:00
a63d329261
[paint-extents] Simplify transform_extents
2023-01-07 16:05:24 -07:00
e062f98260
[paint-extents] Fix transform_extents
...
Ouch!
2023-01-07 16:02:21 -07:00
70ca146033
[chafa] Re-enable truecolor mode
...
See 42bf8e3d49https://github.com/harfbuzz/harfbuzz/pull/2959#issuecomment-827056111
2023-01-07 15:48:57 -07:00
823a9b18d9
[cairo] Return COMPOSITE mode CLEAR for unknown values
...
As per the spec.
2023-01-07 14:50:30 -07:00
d0aaea2319
Update test results for paint-test
...
These were affected by the PaintComposite optimization.
2023-01-07 16:30:56 -05:00
65c3cde5da
[COLRv1] Fix scale variation
2023-01-07 14:29:18 -07:00
004cdc10f8
[open-type] More tweaks to fixed types
...
Add set_int().
2023-01-07 14:27:15 -07:00
c8486b6301
[open-type] Add to_int to fixed types
...
To make sure we don't accidentally forget to_float().
As we did recently in COLRv1 code.
2023-01-07 14:15:17 -07:00
dfd9bf8a50
[COLRv1] Fix a couple of missing to_float() calls
...
Ouch!
2023-01-07 14:05:02 -07:00
d045de78c1
[paint] Optimize PAINT_COMPOSITE
...
At the start of each paint call the current group is clear.
So we don't need to start a new group for the backdrop paint.
A paint composite really needs one group push, not two.
2023-01-07 13:56:43 -07:00
5ea5aacda9
[ft-colr] Adjust for FreeType master color-stop change
...
Part of https://github.com/harfbuzz/harfbuzz/issues/4013
2023-01-07 13:25:11 -07:00
1eb5445e75
[subset] Fix issue in hb_subset_input_override_name_table()
...
If a nameRecord with provided name_id/platform_id/encoding_id/lang_id
is not retained after subsetting, create it and insert it to
the name table. So we need to check against retained name_records
rather than name_record in the original name table.
2023-01-06 19:17:14 -07:00
30d4a73473
[hb-subset] Adjust help for instancing
2023-01-06 16:00:53 -07:00
3bcf153ad8
Change library numbering scheme
...
Fixes https://github.com/harfbuzz/harfbuzz/issues/1431
2023-01-06 15:51:35 -07:00
02948263f9
[subset] Document that CFF2 instancing is now supported
2023-01-06 13:58:29 -07:00
aba6cbe867
[hb-subset] Adjust --help-all formatting
...
Meh.
2023-01-06 13:09:10 -07:00
82c863a50b
Whitespace
2023-01-06 12:51:58 -07:00
76879c5763
[subset-cff] Minor hide num_coords again
2023-01-06 11:47:04 -07:00
576b36a31b
[cff2] Undo rounding change in draw() codepath
2023-01-06 11:40:21 -07:00
4867e0b192
[subset-cff2] Faster instancing
...
Instantiate blends during parsing. Dedups code as well.
2023-01-06 11:39:13 -07:00
3757baab2c
[subset-cff2] Better condition
...
Previous condition wasn't working for dropping axes.
2023-01-06 11:23:37 -07:00
acc6c13f05
[subset-cff] Round numbers when instancing
2023-01-06 11:20:41 -07:00
2f174f23c3
Rename
2023-01-06 11:11:14 -07:00
a59116cd8f
Oops
2023-01-06 11:09:09 -07:00
c632a164b9
[subset/cff] Support instancing
2023-01-06 11:01:25 -07:00
5153218b41
[set] Add hb_set_is_inverted()
2023-01-05 16:26:41 -07:00
e8ac0ef2fd
[face] Minor rename a variable
2023-01-05 16:20:59 -07:00
14ff747024
[set] Add tests for inverted set range iteration
2023-01-05 16:20:59 -07:00
381ac2fd78
docs: Fix a typo
2023-01-05 15:55:57 -07:00
Behdad Esfahbod
Matthias Clasen
Qunxin Liu
Khaled Hosny