You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, makeotf combines the chain targets for the following feature into the SinglePos lookups {A:-40, B:-40, C:-40} and {A:-111, B:-40}. The duplicated entry for B seems redundant. Instead of putting B into both SinglePos chain targets with the same value, makeotf could emit {A:-40, B:-40, C:-40} and {A:-111} to save some space.
feature test {
pos X [A-B]' -40 B' -40 A' -40 Y;
pos X A' -111 Y;
pos X B' -40 A' -111 [A-C]' -40 Y;
} test;
This is definitely a contrived test case, and I’m not sure how wasteful the current encoding is in practice. Just noticed this while working on fonttools/fonttools#514.
The text was updated successfully, but these errors were encountered:
Currently, makeotf combines the chain targets for the following feature into the SinglePos lookups {A:-40, B:-40, C:-40} and {A:-111, B:-40}. The duplicated entry for B seems redundant. Instead of putting B into both SinglePos chain targets with the same value, makeotf could emit {A:-40, B:-40, C:-40} and {A:-111} to save some space.
This is definitely a contrived test case, and I’m not sure how wasteful the current encoding is in practice. Just noticed this while working on fonttools/fonttools#514.
The text was updated successfully, but these errors were encountered: