currently if the user defines two identical lookups we will assign each a unique lookup id (although the tables themselves will be deduplicated during table packing).
This is the source of some difference in GSUB ttx for oswald.
This issue is imprecise; duplicate lookups can be meaningful (because of lookup order) and the real issue in the GSUB diff had more to do with how we combine rules in contextual lookups, which is improved in #842.
currently if the user defines two identical lookups we will assign each a unique lookup id (although the tables themselves will be deduplicated during table packing).
This is the source of some difference in GSUB ttx for oswald.