shadowfacts/fonttools
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • 7afc6b989c uh, and the other one jvr 2005-01-24 10:18:58 +00:00
  • 79c1316d65 fixed buglet in GlyphSet support code jvr 2005-01-24 10:06:45 +00:00
  • 1ed5b4746d fixed 2.4 compat issue jvr 2005-01-24 10:05:52 +00:00
  • 90beb95b77 - rename file to fileOrPath - check for capability, not type, so XMLWriter can hanlde unicode filenames (on OS-es that support them). jvr 2005-01-17 21:34:06 +00:00
  • f34c6f3937 fix for new 2.4 hex() behavior jvr 2004-12-24 16:07:01 +00:00
  • e0912bb819 fix for new 2.4 hex() behavior jvr 2004-12-24 15:59:35 +00:00
  • 0cc0b31fde Updates from Adam Twardoch. Thanks! jvr 2004-12-14 07:55:55 +00:00
  • 28ae196229 Refactored and enhanced table order support: - Rewrote sorting function, it was really quite buggy. - Added reorderFontTables() functions, which reorders the tables in a font at the sfnt level. - TTFont.save() will now by default rewrite the font in the optimized order. This is done through a temp file since our dependency checking logic gets in the way of writing the tables in a predefined order directly (if table A depends on B, table B will always be compiled and written first, so this prevents A from showing up in the file before B). jvr 2004-11-16 10:37:59 +00:00
  • a4f6688ebe tweak & bugfix jvr 2004-11-16 09:12:30 +00:00
  • 427f9802bc make sure that a cmap subtable instance always has a language attr, so __cmp__ can't fail jvr 2004-09-26 18:32:50 +00:00
  • fba0609d54 Fix for [ 808370 ] Dumping Legendum.otf fails on 'kern' table Work around buggy kern table. jvr 2004-09-25 10:56:15 +00:00
  • 1fae9b48d3 [ 637398 ] Failure while parsing OpenType file Deal with empty Coverage table: it will be None so won't have a .glyphs attribute. jvr 2004-09-25 10:31:29 +00:00
  • 266bf21dbe workaround for bug [ 766694 ] Error from invalid date jvr 2004-09-25 10:01:12 +00:00
  • 980fae897a bug #784690: simple workaround for buggy name table jvr 2004-09-25 09:12:00 +00:00
  • a84b28d934 - Refactored XML writing, removed lots of code duplicaiton - Only output unicode names as comments if we're in fact dealing with a unicode cmap subtable (and this is -- in theory -- independent of cmap format) jvr 2004-09-25 09:06:58 +00:00
  • 175f048123 whoops, rolling back accidental #! commits jvr 2004-09-25 08:24:37 +00:00
  • 39c1918939 "Downgraded" AGL list to the "Adobe Glyph List For New Fonts", which is most appropriate here. There may be a use for the "big" AGL, but that will have to become a new module. jvr 2004-09-25 08:12:50 +00:00
  • 13325c613a [ 845172 ] Updating to Unicode 4.0.0 Instead of using a list internally, I now use a dict, since the unicode mapping is quite sparse (lots of unused slots). jvr 2004-09-25 07:47:41 +00:00
  • 6588c4e2df Patch #845551 from Anthony Fok: - two minor typos - changed copyright year in LICENSE (and it's 2004 now...) jvr 2004-09-25 07:35:05 +00:00
  • 0cd79a5642 Patch #845571 from Anthony Fok: - better exception msg in loca table - renamed "version" to "language" in cmap - made cmap 12 work (untested by me) jvr 2004-09-25 07:30:47 +00:00
  • 7718d95951 whoops, forgot one part os OS/2 version 3 support jvr 2004-09-25 07:19:27 +00:00
  • 30bcb12dcb added support for OS/2 table #3 jvr 2004-09-24 18:33:19 +00:00
  • 05e2541b49 Fixed subtle bug in curve intersection logic: due to floating point errors, sometimes a legitimate solution is ever so slightly over 1.0. Those used to be filtered out; now checking for 1.0 + 1e-10. jvr 2003-10-14 20:30:07 +00:00
  • 8e1d75b1a3 ReverseChainSingleSubst support from Yannis H. (must get that generate-otdata-from-the-docs working again) jvr 2003-09-22 13:12:55 +00:00
  • df1f145e5a bug from the spec leaked into here; pointed out by Yannis H. jvr 2003-09-22 07:09:33 +00:00
  • 7325cb480e checked in with unix line endings -- this probably needs proper review jvr 2003-09-18 07:33:01 +00:00
  • 8b8b44904e - Implemented the flex operators for T2 - Changed a whole bunch of XXX traps into NotImplementedErrors jvr 2003-09-17 17:32:11 +00:00
  • deca398915 more doctests jvr 2003-09-16 11:30:29 +00:00
  • f184f75486 Added lots of doco and doctests. jvr 2003-09-16 11:01:40 +00:00
  • 0b0fec3b08 debogofied doc string, added another one jvr 2003-09-16 10:14:48 +00:00
  • d3ee2d4319 - Properly support the pen protocol for open sub paths in Type 1 jvr 2003-09-16 09:48:28 +00:00
  • cc580ac0f8 Added Pen for reportlab.graphics. jvr 2003-09-15 12:26:10 +00:00
  • 27f108844e this should have been part of the previous path by some fixes from klchxbec jvr 2003-09-11 07:11:35 +00:00
  • fbff58cd78 Correctly deal with the TT-no-on-curve special case. jvr 2003-09-09 23:29:45 +00:00
  • 23cb200532 Factored out the SuperBezier and TT-implied-point algorithms, as the may be useful separately from pens. jvr 2003-09-07 09:41:28 +00:00
  • 40cde70f16 - added endPath() method to the Pen protocol, as a counterpart for closePath() for *open* sub paths. This allows pen implementations to reliably detect the end of a sub path. - improved various doc strings. jvr 2003-09-06 16:00:03 +00:00
  • 49b5521f2e ensure that the current point is always set correctly jvr 2003-09-05 14:18:26 +00:00
  • 3a6aa2359e tweaked doc string jvr 2003-09-02 19:23:13 +00:00
  • 81383d6efc more acks jvr 2003-09-01 16:10:22 +00:00
  • 5aaeac3333 Ha, a reserved field got eaten. Noticed by Yannis Haralambous. jvr 2003-09-01 15:09:29 +00:00
  • bb6041f834 *** empty log message *** jvr 2003-08-30 06:46:56 +00:00
  • fc3879e463 - renamed all l,t,r,b tuff to xMin, yMin, xMax, yMax - added ome more doc strings - added some minimal test code jvr 2003-08-29 19:29:46 +00:00
  • 6628513fe4 T2: I'm not sure if the seac-variant of the endchar operator may be combined with actual outlines, but if it is, we need to do the closePath before the components are added. jvr 2003-08-29 08:05:39 +00:00
  • 382df6c42a added deprecated endchar/seac support for T2 charstrings jvr 2003-08-28 20:43:06 +00:00
  • 934fe5fb8f - added support for quadratic contours that have NO on-curve points. - more doco and comments. jvr 2003-08-28 19:30:46 +00:00
  • 53d7523c1c more doco and comments jvr 2003-08-28 19:03:29 +00:00
  • 1cff4cb190 workaround for buggy 2.2 mac support jvr 2003-08-28 18:23:43 +00:00
  • 3430abf7ab whitespace nits jvr 2003-08-28 18:14:33 +00:00
  • bafa66e665 Another patch from rroberts. He writes: """It adds full support for cmap format 2, which is what the Adobe CJK fonts use for the Mac cmap subtable.""" jvr 2003-08-28 18:04:23 +00:00
  • ceb3449e2b whoops, forgot to add the .pens subpacke to the distutils script :-( noticed by rroberts. jvr 2003-08-28 17:59:52 +00:00
  • 4162de350e Is the point inside or outside the outline? jvr 2003-08-28 08:51:45 +00:00
  • bfadfe33db - Fixed ZeroDivisionError in solveCubic(). The solution is mathematically dubious (I don't think 0.0/0.0 == 0.0...) but the result seems to be correct. - Documented that soleCubic() and solveQuadratic() are not guaranteed to return the roots in order, and nor that they are guaranteed to not return duplicate roots. jvr 2003-08-26 19:20:33 +00:00
  • d028b7be3c Set .width in _TTGlyph.__init__ after all: these are just this wrapper objects, _TTGlyphSet doesn't cache them, so setting .width in .draw() is confusing to say the least. jvr 2003-08-26 19:00:38 +00:00
  • e028b38340 new Cocoa=specific drawing pen jvr 2003-08-26 18:20:27 +00:00
  • 6f8f5ded55 fixed AttributeError in _TTGlyphSet.keys() jvr 2003-08-26 18:19:11 +00:00
  • d453a23b38 typo pointed out by Adam T. jvr 2003-08-26 12:02:48 +00:00
  • 4ca01eab4a output don't-edit note jvr 2003-08-25 21:19:47 +00:00
  • 6d9e8e8831 document that this file is generated jvr 2003-08-25 21:19:09 +00:00
  • 77bc20a387 add rroberts to the Acknowledgements section; updated some years jvr 2003-08-25 21:16:12 +00:00
  • 7d4b693627 add the generic getGlyphSet() API to T1Font as well. jvr 2003-08-25 17:53:29 +00:00
  • 1c9917bcf4 comment typo fix, reflow jvr 2003-08-25 13:20:38 +00:00
  • 808c7885bd lots of stuff jvr 2003-08-25 13:18:23 +00:00
  • 8df8f6385e TTFont now has a getGlyphSet() method, which will return a generic GlyphSet. A GlyphSet is simply a dict-like object mapping glyph names to glyphs. The glyphs in a GlyphSet have a .draw(pen) method and a .width attribute. This provides a generic interface for drawing glyphs or extracting outlines, and works both for CFF-based fonts and TT fonts. jvr 2003-08-25 13:15:50 +00:00
  • 731ed49dbc add some more dict-like stuff to the glyf table jvr 2003-08-25 12:23:01 +00:00
  • ce522410f2 small nits jvr 2003-08-25 07:37:25 +00:00
  • 489d76a340 Refactored outline extraction for CharStrings. The interface to T{1,2}OutlineExtractor is not backwards compatible and this change basically makes them private classes: CharStrings now have a .draw() method that takes a Pen object (see fontTools.pens.*), so you never have to deal with the extractor objects yourself. Only lightly tested. jvr 2003-08-24 19:56:16 +00:00
  • 3d24a106ba don't use macfs, it's deprecated jvr 2003-08-24 19:52:03 +00:00
  • 382102645a small tweak jvr 2003-08-24 17:23:34 +00:00
  • 9920ad5286 added and tweaked some asserts jvr 2003-08-24 16:25:12 +00:00
  • 6385a4e9e1 remove trailing whitespace jvr 2003-08-24 16:17:11 +00:00
  • 2bb6cd3506 ugh, lineTo != moveTo... jvr 2003-08-24 09:48:49 +00:00
  • 82ef2a52c9 wrapped some long lines jvr 2003-08-23 20:24:42 +00:00
  • b369ef33fc Pen stuff, see http://just.letterror.com/cgi-bin/wypy?PenProtocol Only lightly tested, component support is not tested at all. jvr 2003-08-23 20:19:33 +00:00
  • b47e3ba3ae some fixes from klchxbec jvr 2003-08-22 20:21:08 +00:00
  • aa2723c559 VORG support by rroberts. jvr 2003-08-22 20:02:25 +00:00
  • ed10151701 Lots of CID work by rroberts. jvr 2003-08-22 19:53:32 +00:00
  • 700df03131 - attempted to sort tables in order recommended by spec. TODO: need to fix table dependency order to complete this. (Read: would you mind posting a bug report regarding this?) jvr 2003-08-22 19:44:08 +00:00
  • f509c0f070 support for CEF fonts: don't depend on the head table being available jvr 2003-08-22 19:38:37 +00:00
  • 07dd0e4504 recalc numGlyphs upon writing jvr 2003-08-22 19:34:48 +00:00
  • d407973545 update macUtils to current day MacPython jvr 2003-08-22 18:56:01 +00:00
  • 5149e18031 attempt to set the value for fRectWidth 'more correctly' jvr 2003-08-22 18:53:29 +00:00
  • 0bb3ba5222 some modernizations jvr 2003-08-22 18:52:22 +00:00
  • 2e838ce766 Jaguar Python 2.2 workaround jvr 2003-08-22 18:50:44 +00:00
  • 9e58c90401 this module has been included in so many (in house) packages that it's time it gets a more central place. jvr 2003-08-22 14:56:48 +00:00
  • efaae5245c - optimized a couple of invariant expressions - made sure solveCubic() also works when called with integer arguments jvr 2003-06-29 19:25:05 +00:00
  • 9524c7bdd3 splitLine(): make sure the split is between the end points jvr 2003-06-29 18:32:00 +00:00
  • 05b4b4a271 new module bezierTools.py jvr 2003-06-29 18:25:07 +00:00
  • 182a7ba7be two new functions jvr 2003-06-29 18:18:54 +00:00
  • b19141e489 avoid FSSpec on MacOS jvr 2003-06-07 15:15:51 +00:00
  • 5910f97146 add default _attrs value so pickling AFM objects doesn't blow up jvr 2003-05-24 12:50:47 +00:00
  • 59afba7684 updated for MacPython2.3 jvr 2003-05-24 12:34:11 +00:00
  • e73b8a53a5 Fix for bug #691744; calc_mac_epoch_diff() was broken when the timezone was GMT (or perhaps other situations; it's not entirely clear). jvr 2003-02-23 19:40:07 +00:00
  • 924e4e274e cmap format 12 support, donated by rroberts: thanks! jvr 2003-02-08 10:45:23 +00:00
  • a87cde236e second try to work around bogus stringOffset value jvr 2003-01-25 18:20:22 +00:00
  • b71d0f46f6 renamed stringoffset to stringOffset as per spec jvr 2003-01-25 11:15:42 +00:00
  • aabca6d793 gracefully handle bogus stringOffset values (thanks to Anthony Fok) jvr 2003-01-25 11:14:59 +00:00
  • 1332d3e549 clean up checkSumAdjustment XML output: suppress trialing 'L' jvr 2003-01-10 22:34:13 +00:00
  • 26fc63c854 fix from Owen Taylor that fixes my previous patch; thanks! jvr 2003-01-10 22:23:46 +00:00
  • df22c27d92 detab table string literal jvr 2003-01-03 21:29:23 +00:00
  • 130c1a0d9c make two fields unsigned, to conform to the spec but also to suppress Python 2.3 warnings for hex(negativenumber). jvr 2003-01-03 21:23:04 +00:00