Re: [whatwg/streams] Allow web devs to synchronize branches with tee()? (#1157)

One possible solution is to fail usage of raw tee() with these streams.

I see from the discussion of introducing tee() that it was pretty controversial when it was introduced, precisely because it makes it very easy to make patterns of buffering that are hard to reason cleanly about, and some of those patterns are footguns. (In particular, when I experimented with it, I found that tee() does not respect desiredSize on the destination stream; it enqueues stuff no matter what desiredSize is.

It's entirely feasible to write your own tee-like function that does exactly the queueing strategy that's desired, including dropping when appropriate - the standard tee operator + realtime streams + random jank seems like a Really Bad Combination.





-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/streams/issues/1157#issuecomment-895222507