Should parsing an empty document yield specific error?

[na-g]

@nils-ohlmeier, what should happen when an empty document is parsed?

[nils-ohlmeier]

Well we do have a test verifying that it returns an error https://github.com/mozilla/webrtc-sdp/blob/master/src/lib.rs#L1297

What would be arguments to make this a success instead?

[na-g]

I have updated the question to ask if we want to yield a specific error (which would be matchable) if the document is empty.

[nils-ohlmeier]

Yes that probably makes sense to improve.