someone in the mood to confirm a bug with me? I think this is new. Seeing this with F115 (on linux) on a clean, new profile. The code didn't change.
I used to be able to listen to a big audio file in a streaming manner when loaded from goldfire/howler.js library.
The new behaviour is: firefox will download the whole file before starting to play.
Example: (25MB bytes down will occur) https://jsfiddle.net/ugnt03jo/1/ -> click Run in upper left to render html
- if you have traffic stats on your system you'll see bytes trickling in the moment you render the html - not when you click the Play button. This is acknowledged in the devtools network-tab after it is fetched in full
- until fetched in full, the Play button will not stream the audio file
#firefox
https://bugzilla.mozilla.org/show_bug.cgi?id=1842432 is a regression introduced by https://bugzilla.mozilla.org/show_bug.cgi?id=1817997
I wonder how the fix at https://hg.mozilla.org/releases/mozilla-esr115/rev/ea6e2e85fec6
is improving the outcome of that gnarly cpp at https://searchfox.org/mozilla-central/rev/ea550f0d1afaf7747ebd988325ea10cf4e3e4de4/dom/media/ChannelMediaDecoder.cpp#385
my intuition is DownloadProgressed() never gets to fire CanPlayThrough() because ComputePlaybackRate() never comes back with aDuration.isInfinite() negated. Someone with a cpp background explaining that function would help me heaps