Dan Sanders

> but doesn't decoding sometimes work for somewhat arbitrary inputs which would then result in information leakage There are demonstrated exploits that use concatenation; {known header} + {injected bytes} can...

> Can you elaborate on what this means? Does each "complete" piece of media data have some kind of identifiable container? Each segment is sequence of frames in a media...

I've updated the linked bug with stable results. In short, the non-stable results were roughly accurate.

I do not have data that is able to correlate MIME type with CORS state.

It's probably worth laying out the goals that WebCodes has for a flush implementation: - Flush is used primarily at end-of-stream, where flush completion is an important signal to the...

I wanted to add that, like with flush (#960), reset is used as a barrier in WebCodecs and it is useful to notify on the output side when the reset...

For WebCodecs, we do not need an abort signal. There is a lot of internal state, but chunks are not processed one-by-one. I would imagine that a Transformer that does...

> With the current semantics, we call `flush()` when the writable end is closed. I mean flush as in #960.

We currently expect that users of WebCodecs will be writing new streams code. Given that, one simple solution that is acceptable to us would be for the default implementation of...

I just wanted to check my understanding of these proposals here; with nearest interpolation I expect clamping will work, linear should be okay, but does cubic also work correctly? Are...