This isn't a perfect solution, but it should help increase confidence
that the transcoder in use is likely to produce an approximately
correct-sized stream.
This switches back to allowing range rather than chunked requests so that
seeking works in the web player, but does so by safer means than previous
solutions, by slightly over-estimating the transcoded size, then sending
dummy bytes to the client to fill any gap. Fixes#1117. Addresses #685.
- Add textbox in user settings page for ListenBrainz token
- Add changes to db
- Add db colume to store MusicBrainz Recording ID
- Add db colume to store ListenBrainz token
- Add test for reading id
- Add tag on testing file
- Add localization entry
Signed-off-by: Shen-Ta Hsieh <ibmibmibm.tw@gmail.com>
Signed-off-by: Andrew DeMaria <lostonamountain@gmail.com>
- Split AudioScrobblerSevice into two classes
- Change api to https in last.fm Audio Scrobbler
Signed-off-by: Shen-Ta Hsieh <ibmibmibm.tw@gmail.com>
Signed-off-by: Andrew DeMaria <lostonamountain@gmail.com>
Apache commons is providing Html-escaping, no need to reinvent the wheel:
> It supports all known HTML 4.0 entities, including funky accents. Note that the
> commonly used apostrophe escape character (') is not a legal entity and so
> is not supported).
So I manually checked that nothing is relying on escaped single-quotes,
and didn't manage to find anything that does.
This feature just generated a lot of unnecessary traffic to the server and
hurt usability because the page would refresh unexpectedly while a user was
actively looking at content. Fixes#1003, #1179, #729.
Did not remove the column for that setting in the user database, since that
would lead to complications if needing to downgrade for some reason.