@borkdude so i tried working on a hook for libpython-clj: https://gist.github.com/sogaiu/b6aa01efbd4006bbcab8decf2d5889aa in this case the construct is similar to a defn. is there something i can do to make the defined name show up in the analysis so i can use that toward indexing?
if you expand to a defn that uses that name, I think it should also show up in the analysis? if it doesn't, it might be a bug or there's some reason I don't know off the top of my head
thanks
could also be metadata related, forgetting to restore a position - not sure.
thanks for the tips
ah i see rum's example tries to maintain metadata and i have not done so
@borkdude using with-meta
made a difference in whether row, col info had something numeric. thanks!
there was another piece to the puzzle. it turns out libpython-clj uses a construct similar to import-vars from potemkin (though it's not exactly the same), so i think if i write a hook for that, the indexing may work.
https://github.com/techascent/tech.parallel/blob/master/src/tech/parallel/utils.clj#L4-L27
@sogaiu clj-kondo also has support for import-vars so if you rewrite to that, may also work