ENH: support bytearray and unicode (#3)

[westurner]

• ENH: support {bytearray, unicode} #3
• TST: manual test bencode.dumps with Python 2
• TST: manual test bencode.dumps with Python 3
• TST: test that a .torrent file at least loads (with Transmission)
  • is this handling of utf8 unicode strings consistent with which BEP?
    • https://wiki.theory.org/BitTorrentSpecification#Metainfo_File_Structure
      " All character string values are UTF-8 encoded."
• TST: write test cases for bytearrays
• TST: write test cases for unicodes
• PRF: port to _fast.c

[notpushkin]

Fails on Python 3, NameError: name 'unicode' is not defined :(

[westurner]

Whoops. I think that should only be for Python 2 because in Python 3 all
str are unicode.

On Wednesday, November 16, 2016, Alexander Pushkov notifications@github.com
wrote:

Fails on Python 3, NameError: name 'unicode' is not defined :(

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
#4 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AADGy77ug9XCD5PdA461pQNqM3AtSJzmks5q-vfTgaJpZM4Kwxlh
.

[notpushkin]

The error actually happens in the elif t is unicode line. We can safely substitute unicode for str (in py3k) here (as unicode_to_bytes accepts str in python 3), so I suggest this simple fix: Songbee@0bb114b

[westurner]

Good call @iamale

[westurner]

• TST: What should I add to TEST_DATA to test for unicode strings?
• ENH,TST: IDK Python C-API well enough to port this from _pure.py to _fast.c

[westurner]

... TIL "All character string values are UTF-8 encoded." https://wiki.theory.org/BitTorrentSpecification#Metainfo_File_Structure

[westurner]

is this correct?

[notpushkin]

Also I think it would be nice to decode unicode strings to unicode/str as well (although it might be a bit trickier).