29 Commits

Author SHA1 Message Date
Cosimo Lupo
d2ee826680
[ttGlyphPen] explain 'handleOverflowingTransforms' option in docstring
[skip ci]
2018-03-01 20:34:47 +00:00
Cosimo Lupo
4173315f09
[ttGlyphPen] rename option to 'handleOverflowingTransforms' 2018-03-01 20:17:48 +00:00
Cosimo Lupo
56be13915b
[ttGlyphPen] minor 2018-03-01 20:03:45 +00:00
Cosimo Lupo
aabc894d19
[ttGlyphPen] don't clamp to almost2 if asked to not decompose
In that case, let the struct.error do its job, no need to raise OverflowError ourselves here
2018-03-01 19:59:25 +00:00
Cosimo Lupo
acb1ebc793
[ttGlyphPen] add 'decomposeOverflowingTransform' option
if it's false and any transform value overflows, we raise an error instead of decomposing
2018-03-01 19:39:00 +00:00
Cosimo Lupo
4377185e29
[ttGlyphPen] further simplify 2018-03-01 19:15:36 +00:00
Cosimo Lupo
6606bbef3c
[ttGlyphPen] simplify 2018-03-01 18:50:45 +00:00
Cosimo Lupo
6a4ccd0e9f
[ttGlyphPen] check for out-of-range values before clamping 2018-03-01 18:00:41 +00:00
Denis Moyogo Jacquerye
098d9ecf6f
TTGlyphPen: handle scale limit
* decompose components with scale out of range [-2, 2)

* use closest F2Dot14 value when 0b1.11111111111111 < scale <= 2
2018-03-01 17:48:48 +00:00
Behdad Esfahbod
f3bfea12c6 Add __all__ to pens that didn't have it 2017-02-21 12:01:10 -06:00
Cosimo Lupo
9b83c377c2 [ttGlyphPen] only add empty ttProgram to simple glyphs, not to composites
Unlike simple glyphs (with contours) which must always have `instructionLength`, the field is optional in composite glyphs and depends on whether `WE_HAVE_INSTR` flag is set:
https://www.microsoft.com/typography/otspec/glyf.htm

In the previous code, the TTGlyphPen was adding an empty `program` attribute to any new glyph, including composite glyphs. It should only do it for simple glyphs.

For example, if you take Roboto-Regular.ttf from here:
https://github.com/google/roboto/releases/tag/v2.129

The font (unhinted) was compiled using the ttGlyphPen. It has 1434 composite glyphs. For each of them, a USHORT value for the number of instructions (0) is written, so the binary ttf is 2.868 KB greater than
what it should be.
2016-03-11 10:53:41 +00:00
Cosimo Lupo
87d13888ce [ttGlyphPen] check last point in qCurveTo is not None
otherwise, if points[-1] == 0, the on-curve point won't be added.
2016-03-09 17:58:48 +00:00
James Godfrey-Kittle
fd5b6de3ba [TTGlyphPen] Add component flags argument 2016-03-07 14:48:46 -08:00
James Godfrey-Kittle
f02e771527 [TTGlyphPen] Don't cast values to int
The glyf table will handle rounding when compiling, so this is
unnecessary.
2016-02-16 17:17:14 -08:00
James Godfrey-Kittle
45e496fdb8 [TTGlyphPen] Use bytes for array.array for python3 2015-11-06 11:50:39 -08:00
James Godfrey-Kittle
9ff7d62454 [TTGlyphPen] Do special case ignoring anchors
This fixes the ignoring anchors test, which led me to believe we
didn't need to special case them (which we do, and now are).
2015-11-05 15:28:19 -08:00
James Godfrey-Kittle
6dc5092e2b [TTGlyphPen] Move some code around
No effective differences, just lingering un-committed local changes.
2015-11-05 15:19:53 -08:00
James Godfrey-Kittle
0b3b83f765 [TTGlyphPen] Don't special case ignoring anchors
There's no need to special case this, it happens anyways.
2015-11-05 15:12:18 -08:00
James Godfrey-Kittle
b511b2b0d8 [TTGlyphPen] Go back to using AbstractPen
Using BasePen breaks on quadratic splines, since there's no way to
know if a on-curve point was added by the pen or in the original
glyph. Should have considered this before....
2015-11-05 13:57:53 -08:00
James Godfrey-Kittle
7767683ea4 [TTGlyphPen] Abstract out a _popPoint method 2015-11-05 13:57:06 -08:00
James Godfrey-Kittle
9f5c51897c [TTGlyphPen] Move checks for closed contours
We probably don't want to raise errors in init, which allows a drawer
to reset the pen state.
2015-11-05 13:54:51 -08:00
James Godfrey-Kittle
fa82fcd213 [TTGlyphPen] Make components with correct defaults
This seems consistent with what one sees opening up an existing TTF.
2015-11-04 17:59:57 -08:00
James Godfrey-Kittle
8c61d08d94 [TTGlyphPen] Remove last point in closed contours
When a pen adds the first point to a contour at the end to close it,
we should remove it for the TrueType outline.
2015-11-04 17:58:27 -08:00
James Godfrey-Kittle
1c73a1942a Don't use lineTo in moveTo
It shouldn't make a difference, but semantically this is now correct.
2015-11-04 17:21:17 -08:00
James Godfrey-Kittle
b83ed7ab2b Allow a TTGlyph pen to be used more than once 2015-11-04 17:19:58 -08:00
James Godfrey-Kittle
58dd55c0ae Use BasePen instead of AbstractPen in TTGlyphPen 2015-11-04 17:18:03 -08:00
James Godfrey-Kittle
8b4c3b7892 Add assertion and comments to ttGlyphPen 2015-10-06 09:14:04 -07:00
James Godfrey-Kittle
b916c5bb93 ttGlyphPen: handle qcurves with no on-curve points 2015-10-06 09:12:15 -07:00
James Godfrey-Kittle
b0c03b86d6 Add pen which draws to TrueType glyphs. 2015-10-01 15:12:20 -07:00