Skip to content

Conversation

@goodmami
Copy link
Owner

@goodmami goodmami commented Jan 5, 2026

This reduces Synset.ili to only have a string identifier or None, since this the main usage of them, and all ILI class modeling and higher-level functionality goes to a wn.ili module. This thus follows Larry Wall's maxim easy things should be easy, and hard things should be possible.

This change makes it more difficult to get at proposed ILIs (e.g., synsets don't indicate whether they have one or not), but not without justification. Proposed ILIs cannot be used in interlingual queries, and arguably their only use is for lexicographers and CILI maintainers, not everyday users of Wn.

This change also makes it so ILI metadata is on the ILI definition rather than ILI objects themselves. This is more accurate according to where the metadata comes from in WN-LMF.

This helps avoid cyclic imports for upcoming wn.ili module.
It also makes the _core module a bit more manageable.
Resolves #282

This commit also cleans up a bit of linting and type-checking issues.
@goodmami goodmami merged commit 2d3db19 into main Jan 5, 2026
14 checks passed
@goodmami goodmami deleted the gh-282-ili-module branch January 5, 2026 06:56
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants