Caoxiang Zhu

@jameskermode I think I might miss some other parts (or I lost something when merging). The following test doesn't pass. I will work on this later. Just post an example...

I am close to finishing. I think I need some help with the following issue. If a derived type is conflicting with python keywords, it should also be renamed. In...

A quick fix is to ignore the case when the derived type is conflicting with python keywords. Then everything else should be working and the only limitation is that people...

So far, it seems working. Great, thanks! I will try this with another legacy code which might have more issues. By the way, if you don't mind, do you know...

@jameskermode Thanks for your reply. For the linked libraries, are they required to be compiled with `-fPIC`? I got the following error with `ScaLAPACK`, which suggests recompiling with `-fPIC`. But...

You can choose the type mapping in the file `kind_map`. I am not quite sure if this will solve your question. You can check one of the examples.

I saw the same issue. For some reason, `GCC` doesn't have this issue, but `Intel` has. I was trying to set the `kind_map` to `long_int`, but it doesn't work.

@jameskermode Yes, I tested it and it is possible for `f2py`. This comes from the documentation at the end of https://numpy.org/devdocs/f2py/python-usage.html.

@jameskermode It is also possible for `f2py-f90wrap`. The "problem" happens at generating the `.fpp` file will treat `b` array as a function. ``` @property def b(self): """ Element b ftype=real...

@jameskermode By doing so, I got the following error. ```pytho In [4]: lib.library.b = [[1, 2, 3], [4, 5, 6]] --------------------------------------------------------------------------- RecursionError Traceback (most recent call last) in ----> 1...