Increase sanity level of string handling

hynek / pem

PEM file parsing in Python.

https://pem.readthedocs.io/

MIT License

159 stars 37 forks source link

Increase sanity level of string handling #24

Closed mithrandi closed 8 years ago

mithrandi commented 8 years ago

This PR makes the PEM objects always use bytes internally. In addition:

An as_bytes method is added for always getting bytes.
__str__ still returns a native string as required by Python.
__repr__ doesn't explode on Python 3 (fixes #23)
Constructing the PEM object types with either unicode or bytes now works on both Py 2/3, but the unicode path is deprecated.

hynek commented 8 years ago

Hm since PEM is about convenience, I'd say that deprecating Unicode construction is not desirable.

mithrandi commented 8 years ago

Hmm, I guess I don't really have strong feelings either way! Reverted the deprecation.

mithrandi commented 8 years ago

I have no idea what the heck that Travis failure is about, but I'm pretty sure it's not my fault :grin:

hynek commented 8 years ago

yeah it was just Travis being a distributed system

hynek commented 8 years ago

Code looks OK but you’ll have to fix the docs still. Currently it says They have *no public API* except that they can be transformed using``str(obj)``into a PEM string. which is now patently wrong. :)

mithrandi commented 8 years ago

Okay, pushed a change, did I miss anything else that needs to be updated?

hynek commented 8 years ago

I think all that’s missing is a link to this PR and we’re done!

mithrandi commented 8 years ago

Ah, yes. Added :)

hynek commented 8 years ago

Thanks!