We already have char** typemaps and they are exact copy-pastes. No reason to have both, I tested this and diffed the generated code and it doesn't seem to be any different with this change.
Details
Diff Detail
Event Timeline
Did you mean to delete the tid_t typemap as well?
Assuming there truly are identical typemaps and these are just copies, and everything works just as well when they are removed, I have no objection
I just moved tid to the bottom because it was in the middle of the in, out, and typecheck versions of the char** typemap. So I wanted to group them all together.
Anyway, yea this didn't seem too controversial to me but I figured I'd throw it up here for review in case you knew of some quirk of swig with regards to typematching and constness. But yea, I tested it and the generated code is semantically identical.
If anything breaks it will probably be due to a swig version difference, so let me know after I commit since I'm using 3.x
Ops, sorry, didn't notice the tid_t type map was actually added back later on in the patch. My bad.
I'm going to accept this for now, with the caveat that we'll need to work through it and possibly revert if it breaks on our SWIG. Unlikely that that happens though.