Triage issues for v0.4.0

[mikeboers]

There has been a ton of work done since v0.3.3. We should cut out a new release.

However, I get the feeling that there are some regressions with it. We should triage the tickets still open since v0.3.3 to make sure they can be "ignored" for a new release.

(edited to just be the issues left to address)

• Assert no blocking issues on GitHub:
  • Invalid PTS in (expected to be) automatic conditions. #202
  • Frames lost during transcode. #212
  • Passing BytesIO object to av.open causes crash. #257
• Document everything backwards incompatible in changelog:
  • Stream.encode returns a list of packets, instead of a single packet.
  • AudioFifo and AudioResampler will raise ValueError if input frames
    inconsistant pts.
  • CodecContext.rate has been removed, but proxied to VideoCodecContext.framerate
    and AudioCodecContext.sample_rate. The definition is effectively inverted from
    the old one (i.e. for 24fps it used to be 1/24 and is now 24/1).
  • Fractions (e.g. time_base, rate) will be None if they are invalid.
  • InputContainer.seek and Stream.seek will raise TypeError if given
    a float, when previously they converted it from seconds.

😨

[mikeboers]

I'm tempted to just cut a v0.4.0 right now. The tests pass as is. That is what they are there for. If there is something showstopping surely someone would have either said by now, or it should be trivial to fix in a v0.4.1.

[Schamper]

Any update on this?

[mikeboers]

No not really.

I've had a lot of personal stuff coming up. Honestly, while I've been reading all of the notifications from GitHub (as you can tell my be noticing this at all) I've been stuffing them into a "FLOSS" folder for when I eventually dig myself out.

As responsible as it may or may not be, I think I could be okay with releasing as-is, and seeing what happens. Can I get some 👍 on that?

[Schamper]

👍 from me

[tuxuser]

I would appreciate that too!

[notedit]

+1 Use pyav in my project.

[mikeboers]

So... uh... I just released v0.4.0.

A couple tests fail, I think because of FFmpeg changes in the ~6 months since I've run them.