182 Commits

Author SHA1 Message Date
Behdad Esfahbod
72b6102949 [instancer/L4] Fix normalizeValue for L4 solver
Imagine a font with current min/default/max of 100,700,1000. And new
setting of 100,400,1000. The current normalizeLocation will calculate
the new location for 700 to be +.33, whereas it should calculate +.5!
This is because 400 translates to -.5, so 700 will be normalized to
-1,-.5,+1 and get +.33...

We need a special normalizeLocation that is aware of the "distance"
between min/default/max, ie. the non-normalized values. Then it will be
clear that the distance from 400 to 700 is equal to 700 to 1000, and as
such 700 should be normalized to .5, not .33... I'm still trying to
figure out the case where avar is present.

Store this distance in NormalizeAxisLimit and reach it out in the
solver.

Fixes https://github.com/fonttools/fonttools/issues/3177
2023-06-21 15:09:56 -06:00
Behdad Esfahbod
94e081611c [instancer/L4] Further simplify
outGain is always zero in this branch.
2023-06-21 15:09:56 -06:00
Behdad Esfahbod
7385cbbc34 [instancer/L4] Simplify 2023-06-21 15:09:56 -06:00
Behdad Esfahbod
023ad3a363 [instancer/L4] Fix crossing calculation 2023-06-21 15:09:56 -06:00
Behdad Esfahbod
3ba7e6d706 [instancer-solver] Fix a bug
Fixes https://github.com/fonttools/fonttools/issues/3139
2023-05-29 17:39:02 -06:00
Behdad Esfahbod
33a94d820d [instancer/VarComposite] Allow instancing non-variable VarComposite axes 2023-01-30 11:26:20 -07:00
Behdad Esfahbod
e4c25d02e6 [instancer/VarComposite] Raise NotImplementedError when warranted 2023-01-30 11:26:20 -07:00
Behdad Esfahbod
0d24114ad0 [varLib/VarComposite] Plug in in a couple places 2023-01-30 11:26:20 -07:00
Behdad Esfahbod
7899dd8dbc [instancer] Limit axis ranges to that of fvar
Fixes https://github.com/fonttools/fonttools/issues/2959
2023-01-26 11:47:25 -07:00
Nikolaus Waxweiler
d584daa8fd Blacken code 2022-12-13 11:26:36 +00:00
Cosimo Lupo
dbc3b48879
[instancer] udpdate OS/2 class and post.italicAngle when default moved (L4)
Previously we were only updating these when pinning wght/wdth/slnt axes.
Now we do whenever any of these axes defaults are changed, whether or not the axes are pinned or kept

Fixes https://github.com/fonttools/fonttools/issues/2885
2022-11-09 16:33:23 +00:00
Behdad Esfahbod
672c4bacdd [instancer.featureVars] Address review comments 2022-11-04 11:55:48 -06:00
Behdad Esfahbod
392706949c [instancer.featureVars] Fix deficiency in L4 instancing
See discussion on the issue.  I think I got this right.

Fixes https://github.com/fonttools/fonttools/issues/2737
2022-11-03 14:55:01 -06:00
Behdad Esfahbod
1b6f4fb858 [instancer.featureVars] Minor adjustments 2022-11-03 14:54:47 -06:00
Behdad Esfahbod
ecfcdffe9f [instancer.featureVars] Combine two functions
Fold _limitFeatureVariationRecord() into
_instantiateFeatureVariationRecord().
2022-11-03 14:22:02 -06:00
Behdad Esfahbod
6e156a7f13 [instancer] Split featureVars code into own module
Towards fixing https://github.com/fonttools/fonttools/issues/2737
2022-11-03 14:21:27 -06:00
Behdad Esfahbod
492645239b [varLib / instancer] Remove DSIG table from font
Fixes https://github.com/fonttools/fonttools/issues/1480
2022-10-28 11:42:57 -06:00
Cosimo Lupo
b74d098115
add custom repr to AxisTriple so they print nicer in the logging output 2022-10-21 18:38:55 +01:00
Cosimo Lupo
86e1202129
[instancer] expand singles/tuples to triples upfront and use triples throughout
Also renamed AxisTent => AxisTriple because I think "tent" is more appropriate to visualize master supports (i.e. a triangle with a peak in the middle), here it's simply the new desired min, default, max values of an axis, or the axis limits.

The public facing instantiateVariableFont continues to accept a hybrid map of axis tags to single values or 2- (and now also 3-) tuples, these are normalized to an AxisLimits mapping that contains only AxisTriples and passed around to the internal methods.
2022-10-21 17:49:35 +01:00
Behdad Esfahbod
233014117b [instancer] Adjust featureVariations retaining logic
https://github.com/fonttools/fonttools/pull/2728#discussion_r975606307
2022-10-13 10:35:02 -06:00
Behdad Esfahbod
9c7ffef55e [instancer] Run black 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
e554ec1dc6 [instancer] Small fix for featureVariations instancing 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
1ff1a537e8 [instancer] Implement L4 featureVariations 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
bce11dfc85 [instancer.solver] Improve diagram for a case 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
3bdfb3e205 [instancer] Remove debug print 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
3f7b250803 [instancer] Add test for FeatureVariations limiting range 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
5ff9546531 [TupleVariation] Add __imul__ 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
fbf796a0a9 [instancer] Use minimum/maximum instead of tuple indices
One of the cases is a bugfix in fact...
2022-10-13 10:35:02 -06:00
Behdad Esfahbod
36bf633555 [instancer.solver] Format 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
b54ac734cc [instancer.solver] Add assertion 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
aabced171e [instancer.solver] Fix case3 with gain
Ouch!
2022-10-13 10:35:02 -06:00
Behdad Esfahbod
7ee8414cab [instancer.solver] Better handling of peak==axisDef 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
e08553011a [instancer.solver] Handle when peak falls on new axis default 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
528bcdf1a2 [instancer] No need to check for scalar=1
The scaleDeltas() function does it.
2022-10-13 10:35:02 -06:00
Behdad Esfahbod
d2cb561cee [instancer] L4 avar mapping 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
24cd9dc00e [instancer] Simplify avar mapping more 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
7330c78b10 [instancer] Towards L4 avar 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
d17d1355a2 [instancer] Simplify avar normalization 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
3108e52a90 [instancer] Remove redundant checks 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
e2a875b26c [instancer] Implement featureVariations L4 instancing
I *think* this is all.
2022-10-13 10:35:02 -06:00
Behdad Esfahbod
890938ef09 [instancer.solver] Try harder to fix on python 3.7 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
3c998785db [instancer.solver] Use older lru_cache instead of cache 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
8f6779d67e [instancer] Fix the rest of tests 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
53b6c4c9fe [instancer.solver] Fix negative -2 case 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
5254c9354c [instancer] Fix most tests! 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
cff162579d [instancer] Fix a few more tests 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
272e4e8999 [instancer] Fix a few more tests 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
5118e3327d [instancer] Fix some more tests 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
0947ca6b6c [instancer] Fix "drop" to some extent 2022-10-13 10:35:02 -06:00
Behdad Esfahbod
d03879607b [instancer] Make a few tests happy 2022-10-13 10:35:02 -06:00