Currently PyUnicode_CopyCharacters doesn't check arguments thoroughly. This could lead to undefined behaviour or crash in debug mode. For example, from_start > len(from), how_many < 0. Another case is that when how_many > len(from), it will choose len(from) but this can still fail since from_start can > 0. The doc of it is also not perfect, it does not necessarily return 0 on success.

Added comments on Rietveld. I don't know whether tests for this function are needed. It is public, but is not a part of stable API.

v2 applies Serhiy's suggestions.

New changeset 13addd71b751 by Serhiy Storchaka in branch '3.5':
Issue #28379: Added sanity checks and tests for PyUnicode_CopyCharacters().
https://hg.python.org/cpython/rev/13addd71b751

Thank you for your contribution.

Thanks Serhiy! But sorry I think I have made a mistake. In unicode_copycharacters we don't need PyUnicode_READY since it has been done in argument parse. Could you remove it?

New changeset 1be8cd7cee92 by Serhiy Storchaka in branch 'default':
Issue #28379: Removed redundant check.
https://hg.python.org/cpython/rev/1be8cd7cee92

We don't need to remove it for 3.5 and 3.6?

This is just a cleaning up of not very important code.

Fine. :-)