You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
At the first glance it seems like this might be a considerably cleaner, more modern alternative to ctypesgen.
To my understanding, it delegates parsing / AST creation to a clang binding, which should be more future proof given newer C standards, and there's no need for any custom lex/yacc.
Disclaimer: I haven't tested ctypeslib yet, but plan to do so eventually.
The text was updated successfully, but these errors were encountered:
mara004
changed the title
ctypeslib: possible alternative to ctypesgen
ctypeslib: possible alternative to ctypesgen?
Feb 21, 2024
While it's definitely interesting, I'm not entirely happy with the output -- IMHO pypdfium2-ctypesgen's is better, at least for our use case.
I believe some features might be missing, like precise symbol control or proper dependency handling.
Not to mention that we have to specify system include dirs manually, there are some strange templates, and no __slots__ on structs.
However, it also offers some cute things ctypesgen doesn't, like taking over the C declaration as docstring, or doxygen comments.
It happens that I just came across trolldbois/ctypeslib.
At the first glance it seems like this might be a considerably cleaner, more modern alternative to ctypesgen.
To my understanding, it delegates parsing / AST creation to a clang binding, which should be more future proof given newer C standards, and there's no need for any custom lex/yacc.
Disclaimer: I haven't tested
ctypeslib
yet, but plan to do so eventually.The text was updated successfully, but these errors were encountered: