shadowfacts/fonttools
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
6,255 Commits 1 Branch 0 Tags
Commit Graph

4354 Commits

Author SHA1 Message Date
Behdad Esfahbod
05f95f0f58 [otBase] Try ExtensionLookup if other offset-overflow methods fail 
If offset overflow happens other than between Lookup and SubTable, or
within SubTable, then we don't know how to fix it currently.  Now we
try to put that lookup in an Extension lookup anyway, since that will
isolate it and sharing with other lookups won't happen, hopefully
avoiding the overflow.

This is really a bandaid... What we want is the 99proof branch to
be finished and replace our current packing algorithm.

Fixes https://github.com/fonttools/fonttools/issues/1296
 2018-07-25 11:06:44 -07:00
Behdad Esfahbod
642cb03296 Remove unused line of code 2018-07-25 11:06:44 -07:00
Cosimo Lupo
8991f2c097
Merge pull request #1300 from anthrotype/explicit-subtable-break 
support explicit "subtable" breaks in kerning lookups
 2018-07-25 11:58:26 +01:00
Cosimo Lupo
3734b6c036
cffLib/specializer: minor; use < instead of <= maxstack - 1 2018-07-25 11:56:32 +01:00
Kunihiko Sakamoto
41a21cce27 [cffLib.specializer] Leave 1 stack free area for subroutinizer use 2018-07-25 17:29:27 +09:00
Cosimo Lupo
9456ae9485
feaLib: support explicit 'subtable' statements in PairPos lookups 
Fixes https://github.com/fonttools/fonttools/issues/1279
 2018-07-24 17:13:52 +01:00
Cosimo Lupo
d3229d5777
feaLib: ensure languagesystems using 'DFLT' script tag precede all others 
https://github.com/fonttools/fonttools/pull/1292#issuecomment-407427544

this is to match makeotf's behavior

https://github.com/adobe-type-tools/afdko/pull/498/files#diff-17824930b3cdfb707ae56e1f42bccec6R2305
 2018-07-24 16:17:23 +01:00
Cosimo Lupo
1740d18584
Merge pull request #1292 from silnrsi/bugfix2 
More depths to allowing languages in DFLT
 2018-07-24 15:24:50 +01:00
Cosimo Lupo
248ae41e9d
loggingTools: fix CapturingLogHandler to work with py37 
we can't set the logger.level directly because in py37 the logger has a _cache
attribute and that needs to be invalidated by calling setLevel method.

also, I noticed we weren't restoring the original 'disabled' attribute value...

And we need to also set/restore the 'propagate' attribute.
 2018-07-24 12:17:59 +01:00
Cosimo Lupo
5a1ccf1640
Merge pull request #1297 from anthrotype/fix-mark-base-overflow 
atttempt to fix offset overflows in MarkBasePos subtables
 2018-07-23 12:58:39 +01:00
Cosimo Lupo
94c8bcc333
Bump version: 2.3.1 → 2.3.2.dev0 2018-07-21 19:18:11 +01:00
Cosimo Lupo
902fbc29d6
Release 2.3.1 2018-07-21 18:43:21 +01:00
Cosimo Lupo
e51f81dce4
Bump version: 2.3.0 → 2.3.1.dev0 2018-07-21 18:36:31 +01:00
Cosimo Lupo
6072fe1802
plistlib: rename some private functions 
to better match the plist element names they are meant to construct
 2018-07-21 18:34:27 +01:00
Cosimo Lupo
d7ac0ad359
subset: write default file extension based on --flavor or sfntVersion 
Fixes https://github.com/fonttools/fonttools/issues/1298
 2018-07-21 18:28:33 +01:00
Cosimo Lupo
ab3e069a25
test_plistlib: parametrize tests with use_builtin_types option 2018-07-20 19:25:19 +01:00
Cosimo Lupo
5dcf204729
plistlib: add use_builtin_types option 
Fixes https://github.com/unified-font-object/ufoLib/issues/162
 2018-07-20 19:20:33 +01:00
Cosimo Lupo
6e8e431dee
otTables: try to fix MarkBasePos subtable overflows 
Fixes https://github.com/googlei18n/fontmake/issues/450
 2018-07-19 17:52:00 +01:00
Cosimo Lupo
2798bf6d1f
otTables: print subtable type when can't split in fixSubTableOverFlows 2018-07-19 17:50:43 +01:00
Cosimo Lupo
0602e1fe7b
Release 2.3.0 2018-07-16 15:08:35 +02:00
Cosimo Lupo
bced5b62b9
Bump version: 2.2.5.dev0 → 2.3.0.dev0 2018-07-16 15:01:32 +02:00
Cosimo Lupo
42a3428d0f
writeLayerContents: make sure bytes strings in layerOrder are decoded 
looks like defcon LayerSet.newLayer does not ensure the 'name' argument is a unicode
string, so serializing the layercontents.plist with the new ufoLib.plistlib may
lead to the layer name being encoded as a <data> element instead of a <string>.

Thanks Miguel for noticing this.
 2018-07-15 21:32:13 +01:00
Cosimo Lupo
e15d56a2d3
test_etree: add test for pretty_print=True with ordered dict 2018-07-15 13:55:01 +01:00
Cosimo Lupo
bc7494b7f0
test_etree: start adding some unit tests to etree module 2018-07-15 12:32:17 +01:00
Cosimo Lupo
0c0d46e53b
etree: pretty_print is False by default (like in lxml) 2018-07-15 12:31:11 +01:00
Cosimo Lupo
6c6b94a580
etree: restrict py2 bytes to ASCII only 2018-07-14 20:06:22 +01:00
Cosimo Lupo
abe74f1edc
aarg! fix writeDataFileAtomically not removing empty file... 
this had been broken for years, since commit 66e5ae02
 2018-07-14 19:58:38 +01:00
Cosimo Lupo
b44e3d5087
validators: fixup doctests after adding unicode_literals 2018-07-14 19:42:16 +01:00
Cosimo Lupo
0951e2063e
ensure all modules do from __future__ import absolute_import, unicode_literals 2018-07-14 19:25:18 +01:00
Cosimo Lupo
83e8da2206
etree: only write self-closing <tag/> when element.text is None 
like lxml does
 2018-07-14 17:58:44 +01:00
Cosimo Lupo
1a94fbc121
etree: in invalid xml chars allow surrogates for 'narrow' pythons 2018-07-14 14:31:52 +01:00
Cosimo Lupo
5f61bcfcb4
etree: simplify regex for invalid xml chars 2018-07-14 14:07:42 +01:00
Cosimo Lupo
eea1766d9a
etree: fix valid XML regex for narrow UCS-2 only pythons 
fixes https://ci.appveyor.com/project/adrientetar/ufolib/build/1.0.489/job/epryi911juu5lqdl#L72
 2018-07-14 13:41:19 +01:00
Cosimo Lupo
d1fa34e72c
test_plistlib.py: mute harmless ResourceWarning on py27 2018-07-14 12:22:42 +01:00
Cosimo Lupo
59ac1aa357
test_plistlib: use ufoLib.etree 2018-07-13 18:50:02 +01:00
Cosimo Lupo
a9a37818cb
glifLib: use ufoLib.etree 2018-07-13 18:49:21 +01:00
Cosimo Lupo
a99a6d6ee5
plistlib: make signledispatch optional; use ufoLib.etree 2018-07-13 18:48:21 +01:00
Cosimo Lupo
d87c4edfe1
ufoLib.etree: add shim module that exports ElementTree API 
works with both lxml and xml.etree backends

adds some missing things from built-in etree, such as the ability to
use an OrderedDict for attributes, support for pretty_print argument
to add indentation, etc.
 2018-07-13 18:44:11 +01:00
Cosimo Lupo
49a88d6cbf
test_plistlib: must open plist as binary file 
fixes https://ci.appveyor.com/project/adrientetar/ufolib/build/1.0.484/job/9em92wjinypwn2fu#L399
 2018-07-12 14:10:48 +01:00
Cosimo Lupo
b743ccd2b6
utils: remove testing warning from doctest; fails with py27 
I hate doctests :(
 2018-07-12 13:54:19 +01:00
Cosimo Lupo
72a99726f7
test_plistlib: test deprecated readPlist/writePlist etc. as well 2018-07-12 13:47:23 +01:00
Cosimo Lupo
d664f31a50
plistlib: restore old readPlist/writePlist etc., with deprecation warning 2018-07-12 13:37:38 +01:00
Cosimo Lupo
c6bc1cda0a
add utils.py module with @deprecated decorator 2018-07-12 13:33:21 +01:00
Cosimo Lupo
a526b7170c
tests: fix expected results after Unicode 11 update 
fixes https://github.com/fonttools/fonttools/issues/1291
 2018-07-12 11:35:16 +01:00
Cosimo Lupo
452c85ecef
Update Blocks, Scripts and ScriptExtensions for Unicode 11 
I run: python3 MetaTools/buildUCD.py
 2018-07-12 11:35:16 +01:00
Cosimo Lupo
460c08286d
test_plistlib: add test for pretty_print=False 2018-07-11 18:33:10 +01:00
Cosimo Lupo
8011e828dd
glifLib: call plistlib.totree with intent_level=2 
since 'lib' element is itself contained in 'glyph' root element
 2018-07-11 18:32:37 +01:00
Cosimo Lupo
790876da04
plistlib: justify multiline base64 data; add 'pretty_print' option 
we right-justifiy binary data text to max 76 chars (min 16) per line
similarly to the way the stdlib plistlib module does (although they
use tabs for indentation, equivalent to 8-spaces).

We write the plist doctype ourselves, otherwise lxml always adds a
newline to it (even wen pretty_print is False).
 2018-07-11 18:27:12 +01:00
Cosimo Lupo
a8124a0eef
plistlib: use double-quotes in xml declaration 
for consistency with glifLib
 2018-07-11 13:29:54 +01:00
Cosimo Lupo
61c19acace
Add test_plistlib.py 
this is basically the same cpython plistlib test suite,
rewritten as pytest-style tests
 2018-07-11 12:48:14 +01:00