Petteri Hintsanen <[email protected]> writes: > Yoni Rabkin <[email protected]> writes: > >> Works for me on both 29.1 and 28.2. >> >> I only got one error on one machine, and with one MP3 track. But on >> closer inspection of that track it really doesn't start with "ID3". > > Thanks for testing. Did you get sensible playing times? That's the > main new feature, after all. > > I have found out that just checking for zero length tracks in the cache > might reveal problematic cases: > > (maphash (lambda (key bdata) > (when (eq (alist-get 'type bdata) 'file) > (let ((name (alist-get 'name bdata)) > (playtime (or (alist-get 'info-playing-time bdata) 0))) > (when (zerop playtime) > (message "%s\t%d" name playtime))))) > emms-cache-db) > > For me, most of the zero length tracks are MP3s without an ID3v2 tag, or > some other files than {Ogg, Opus, FLAC, MP3}. All the others have quite > correct info-playing-time.
After feeding the new code all of the track I could muster, the only outliers are in a single album of ogg files which all return a 0 playing time on all versions of info-native, new and old. Nothing immediately stands out to me about the files in this album in terms of how the header looks, but I haven't checked closely. In any case, it isn't a regression. -- "Cut your own wood and it will warm you twice"
