[Erfouni]

If you pass PlyData.read a file that is not a PLY file and whose first bytes are not ASCII (for example a PNG, or any binary file by mistake), it fails with a bare UnicodeDecodeError:

import io
from plyfile import PlyData
PlyData.read(io.BytesIO(b'\x89PNG\r\n\x1a\n'))
# UnicodeDecodeError: 'ascii' codec can't decode byte 0x89 ...

A non-PLY ASCII file already gives a clean PlyHeaderParseError ("expected 'ply'"), so this makes the binary case behave the same way. _PlyHeaderLines._decode does a strict s.decode('ascii'); I wrapped it to raise PlyHeaderParseError on UnicodeDecodeError. Since _decode is also used while reading the rest of the header, this covers non-ASCII bytes mid-header too. Valid files are unaffected. Added a test.

---

📚 Documentation preview 📚: https://python-plyfile--91.org.readthedocs.build/en/91/

[dranjan]

Hey @Erfouni, thanks for the contribution! This shouldn't take long to review. From a cursory glance, I didn't see anything concerning. Also thanks for adding the test.

[dranjan]

Looks good to me. It occurred to me that we're not using the raise from idiom, but that's separate from this pull request.