Closed francesco-ballarin closed 3 weeks ago
You need is_pointer_v, etc.
What would be the best way to explain this change in the changelog @oremanj ?
You need is_pointer_v, etc.
Hopefully after four attempts I've got it right :(
I originally used conjunction because I was worried about potentially forming an invalid expression if T was not a pointer type, but on second thought I don’t think it helps or matters.
Possible changelog entry: “nanobind no longer prevents casting to a C++ container of pointers T*
where T
is a type with a user-defined type caster if the caster seems to operate by extracting a T*
from the Python object rather than a T
.” with a link to the issue that prompted this PR for more context.
I’m mostly offline until Monday and will respond to this in more depth at that time if needed.
Thanks @oremanj , I've added what you suggested to the changelog.
Merged, thanks!
Follow up to discussion #605.
I confirmed with my downstream project that the patch does indeed fix the regression introduced in nanobind 2.0. Please note that, even though the commit has my name and email address, the actual patch was provided by @oremanj.