Release v0.23.1 (please TEST!)

[Stypox]

Testing for regressions

Debug APK (built by our CI in #8547): NewPipe_v0.23.1_RC2_debug.zip
Release APK (signed with @Stypox' keys): NewPipe_v0.23.1_RC2_release.zip
Please report only regressions (i.e. new issues) here, not issues that were already present in the previous release!

TODO

• Merge weblate
• Release a NewPipeExtractor version actually, no, since the current NewPipeExtractor version is only temporary and needs to be reworked by NewPipeExtractor#862
• Updated extractor to latest revision and fix compile errors #8531
• Fix history view count #8336 needs to be reverted or really fixed -> Actually fix history view count #8564

NewPipeExtractor version

This version of NewPipe will ship with NewPipeExtractor version 0.23.0
-> the NPE release still hasn't been done ;-)

App changelog

New

• Support delivery methods other than progressive HTTP Support delivery methods other than progressive HTTP #8153 (thanks to @AudricV!)

  • [YouTube] Use DASH for playback, improving loading times in playback and seeking
  • [YouTube] Full playback of recently ended livestreams
  • [YouTube] Improve availability of some resolutions and formats
  • [PeerTube] Fix crash on videos with an audio-only stream
  • [PeerTube] Fix seeking on videos with only HLS streams
  • [SoundCloud] Fix playback of SoundCloud HLS-only tracks

• Add button to add a remote playlist to a local one Add button to add a remote playlist to a local one #7355

• Image preview in Android share sheet (Android 10+ only) Image preview in Android share sheet (Android 10+ only) #8143

Improved

• Improve playback parameters dialog's UI Improve playback parameter dialog's UI #7989 Fixed accidental reset of PlaybackParameterDialog on initialization #8349
• Moved subscription import/export options to (overflow) menu Moved subscription import/export options to (overflow) menu #7458

Fixed

• Fix keyboard doesn't open with disabled animations Fix keyboard showing after the search box acquiring focus #8227
• Fix crash when open NewPipe from notification bar Fix crash when open NewPipe from notification bar #8244
• Fix hiding player controls when playing from media button Fix hiding player controls when playing from media button #8315
• Fix viewed counting in history Fix history view count #8336 (TODO: ⚠️ needs to be reverted or really fixed)
• Fix RouterActivity theming (the menu that shows up when you share something to NewPipe) Fix Routeractivity theming #8332
• Fix removing only fully watched videos from playlist Fix removing only fully watched videos from playlist #8259
• Fix crash when rotating device on unsupported channels Fix crash when rotating device on unsupported channels #8192
• Make volume progress bar match system volume when we start sliding Make volume progress bar match system volume when we start sliding #8377
• Fix Add to playlist not working and cleanup RouterActivity choice handling (the menu that shows up when you share something to NewPipe) Fix "Add to playlist" not working and cleanup "RouterActivity" choice handling #8340

Localization

• Add Hindi translation of the README Add Hindi translation of the README #8355

Development

• Update ACRA library Update ACRA library #8206
• Update user agent used by the downloader Update User-Agent in Downloader #8269
• Remove unused strings.xml resources Removed unused strings.xml resources #8329
• Update Groupie and ACRA libraries Update Groupie and ACRA libraries #8407
• Update Android Gradle Plugin and Kotlin Update AGP and Kotlin #8406
• Use JvmOverloads instead of manually overloading method Use JvmOverloads instead of manually overloading method #8391
• Remove unnecessary compat method calls Remove unnecessary compat method calls. #8382
• Use AppCompatResources Use AppCompatResources #8404

[swarupbhc]

Please release this version.

[avently]

Network operations like viewing youtube videos and seeking is so much faster than before, so cool

[opusforlife2]

Livestreams often error out due to source errors, but I haven't seen this HLS one yet. Not sure if it's a regression or just another regular livestream error.

Exception

• User Action: play stream
• Request: Player error[type=ERROR_CODE_IO_UNSPECIFIED] occurred while playing https://www.youtube.com/watch?v=gvpE6615qec
• Content Country: CH
• Content Language: en-IN
• App Language: en_IN
• Service: YouTube
• Version: 0.23.1
• OS: Linux Android 10 - 29

Crash log

com.google.android.exoplayer2.ExoPlaybackException: Source error
	at com.google.android.exoplayer2.ExoPlayerImplInternal.handleIoException(ExoPlayerImplInternal.java:641)
	at com.google.android.exoplayer2.ExoPlayerImplInternal.handleMessage(ExoPlayerImplInternal.java:617)
	at android.os.Handler.dispatchMessage(Handler.java:103)
	at android.os.Looper.loop(Looper.java:214)
	at android.os.HandlerThread.run(HandlerThread.java:67)
Caused by: com.google.android.exoplayer2.source.hls.playlist.HlsPlaylistTracker$PlaylistStuckException
	at com.google.android.exoplayer2.source.hls.playlist.DefaultHlsPlaylistTracker$MediaPlaylistBundle.processLoadedPlaylist(DefaultHlsPlaylistTracker.java:739)
	at com.google.android.exoplayer2.source.hls.playlist.DefaultHlsPlaylistTracker$MediaPlaylistBundle.onLoadCompleted(DefaultHlsPlaylistTracker.java:584)
	at com.google.android.exoplayer2.source.hls.playlist.DefaultHlsPlaylistTracker$MediaPlaylistBundle.onLoadCompleted(DefaultHlsPlaylistTracker.java:512)
	at com.google.android.exoplayer2.upstream.Loader$LoadTask.handleMessage(Loader.java:478)
	at android.os.Handler.dispatchMessage(Handler.java:107)
	... 2 more

---

[opusforlife2]

This one isn't a livestream, but a normal video. And I haven't seen this error before either.

Exception

• User Action: play stream
• Request: Player error[type=ERROR_CODE_IO_NETWORK_CONNECTION_TIMEOUT] occurred while playing https://www.youtube.com/watch?v=_9hMQshZNNo
• Content Country: CH
• Content Language: en-IN
• App Language: en_IN
• Service: YouTube
• Version: 0.23.1
• OS: Linux Android 10 - 29

Crash log

com.google.android.exoplayer2.ExoPlaybackException: Source error
	at com.google.android.exoplayer2.ExoPlayerImplInternal.handleIoException(ExoPlayerImplInternal.java:641)
	at com.google.android.exoplayer2.ExoPlayerImplInternal.handleMessage(ExoPlayerImplInternal.java:613)
	at android.os.Handler.dispatchMessage(Handler.java:103)
	at android.os.Looper.loop(Looper.java:214)
	at android.os.HandlerThread.run(HandlerThread.java:67)
Caused by: com.google.android.exoplayer2.upstream.HttpDataSource$HttpDataSourceException: java.net.SocketTimeoutException: failed to connect to rr2---sn-cvh76nlz.googlevideo.com/173.194.154.167 (port 443) from /[blahblahIP] (port [blahblah]) after 8000ms
	at org.schabi.newpipe.player.datasource.YoutubeHttpDataSource.open(YoutubeHttpDataSource.java:389)
	at com.google.android.exoplayer2.upstream.DefaultDataSource.open(DefaultDataSource.java:258)
	at com.google.android.exoplayer2.upstream.TeeDataSource.open(TeeDataSource.java:52)
	at com.google.android.exoplayer2.upstream.cache.CacheDataSource.openNextSource(CacheDataSource.java:786)
	at com.google.android.exoplayer2.upstream.cache.CacheDataSource.open(CacheDataSource.java:599)
	at com.google.android.exoplayer2.upstream.StatsDataSource.open(StatsDataSource.java:84)
	at com.google.android.exoplayer2.source.chunk.ContainerMediaChunk.load(ContainerMediaChunk.java:124)
	at com.google.android.exoplayer2.upstream.Loader$LoadTask.run(Loader.java:412)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1167)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:641)
	at java.lang.Thread.run(Thread.java:919)
Caused by: java.net.SocketTimeoutException: failed to connect to rr2---sn-cvh76nlz.googlevideo.com/173.194.154.167 (port 443) from /[blahblahIP] (port [blahblah]) after 8000ms
	at libcore.io.IoBridge.connectErrno(IoBridge.java:191)
	at libcore.io.IoBridge.connect(IoBridge.java:135)
	at java.net.PlainSocketImpl.socketConnect(PlainSocketImpl.java:142)
	at java.net.AbstractPlainSocketImpl.doConnect(AbstractPlainSocketImpl.java:390)
	at java.net.AbstractPlainSocketImpl.connectToAddress(AbstractPlainSocketImpl.java:230)
	at java.net.AbstractPlainSocketImpl.connect(AbstractPlainSocketImpl.java:212)
	at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:436)
	at java.net.Socket.connect(Socket.java:621)
	at com.android.okhttp.internal.Platform.connectSocket(Platform.java:182)
	at com.android.okhttp.internal.io.RealConnection.connectSocket(RealConnection.java:145)
	at com.android.okhttp.internal.io.RealConnection.connect(RealConnection.java:116)
	at com.android.okhttp.internal.http.StreamAllocation.findConnection(StreamAllocation.java:186)
	at com.android.okhttp.internal.http.StreamAllocation.findHealthyConnection(StreamAllocation.java:128)
	at com.android.okhttp.internal.http.StreamAllocation.newStream(StreamAllocation.java:97)
	at com.android.okhttp.internal.http.HttpEngine.connect(HttpEngine.java:289)
	at com.android.okhttp.internal.http.HttpEngine.sendRequest(HttpEngine.java:232)
	at com.android.okhttp.internal.huc.HttpURLConnectionImpl.execute(HttpURLConnectionImpl.java:465)
	at com.android.okhttp.internal.huc.HttpURLConnectionImpl.connect(HttpURLConnectionImpl.java:131)
	at com.android.okhttp.internal.huc.DelegatingHttpsURLConnection.connect(DelegatingHttpsURLConnection.java:90)
	at com.android.okhttp.internal.huc.HttpsURLConnectionImpl.connect(HttpsURLConnectionImpl.java:30)
	at org.schabi.newpipe.player.datasource.YoutubeHttpDataSource.makeConnection(YoutubeHttpDataSource.java:704)
	at org.schabi.newpipe.player.datasource.YoutubeHttpDataSource.makeConnection(YoutubeHttpDataSource.java:548)
	at org.schabi.newpipe.player.datasource.YoutubeHttpDataSource.open(YoutubeHttpDataSource.java:383)
	... 10 more

---

[opusforlife2]

Oh, and I don't know if this is a regression. YT Shorts thumbnails are now fit to width, cutting off the top and bottom portions, instead of fit to height, with black bars on the side, like in the current release.

Edit: I wasn't specific enough. This is for list item thumbnails. The ones that are shown on the video details page are fine.

[AudricV]

Ended age-restricted livestreams cannot be watched anymore.

Reason: the client used to bypass age-restricted contents only provides a DASH and a HLS manifest, which are incomplete, and no audio or video stream seems to be provided (issue was reproduced on video PXvs97_7z2I, now rencoded as a video and non-age restricted) -> a StreamExtractException exception with the message Could not get any stream is thrown.

In order to fix this issue, we have to parse the DASH manifest (again, and so reintroduce DashMpdParser in the extractor) to extract Streams objects and:

• calculate the average duration between two segments;
• extract the base streaming URL;
• build an ItagItem;
• make YoutubePostLiveStreamDvrDashManifestCreator accepts "parameters and values" using / instead of & and = (edit: we can also convert the URL to use query parameters instead of a full path);
• and probably more.

Screenshot of the streamingData object returned in this case for the client used to bypass age-restricted contents:

[AudricV]

@opusforlife2

Livestreams often error out due to source errors, but I haven't seen this HLS one yet. Not sure if it's a regression or just another regular livestream error.

That's the exception (PlaylistStuckException) for which coefficient was slightly increased. There is nothing we can do about this, and is related to a server-side issue (the error means that the manifest has been not updated since a while, if I am not wrong). It can also happen on the end of brodcasts, because YouTube doesn't end properly HLS manifests sometimes.

This one isn't a livestream, but a normal video. And I haven't seen this error before either.

Were you listening to the content or were you watching it? The TimeoutExceptions may happen in background because of battery optimizations, especially the agressive ones of OEMs, which can do sometimes strange behaviors on networks state. I got this issue probably at most three times row since I use #8153 (for two months at least and maybe more) on my phone running EMUI. If you were watching the content, then that's probably a network/server issue.

We can also increase the connection timeout, if that's needed.

Oh, and I don't know if this is a regression. YT Shorts thumbnails are now fit to width, cutting off the top and bottom portions, instead of fit to height, with black bars on the side, like in the current release.

That's a YouTube change on short thumbnails (which are returned without "borders" (so not anymore in the 16/9 format) on which we can't really do something, except that we can remove the sqp and/or the rs parameters to get original thumbnails from the image URL to get again a 16/9 format.

The only issue is that thumbnails which have normally black bars but don't have them using these parameters will get black bars, which isn't something we'd like to have I think.

Also, shorts thumbnail URLs seem to use a _2 in the end of the thumbnail name.

Finally, contrary to videos, there is now only one thumbnail returned for shorts:

In conclusion, this change is not a regression at all (and is also reproducible on the current release for me).

[opusforlife2]

If you were watching the content, then that's probably a network/server issue. We can also increase the connection timeout, if that's needed.

I was watching. I was thinking that there might be need somewhere in the code to request YT to refresh the connection, instead of just increasing the timeout? If timeout is the only option, then yeah, increasing that might be good. Although I think we might need to test if that helps at all.

[opusforlife2]

The only issue is that thumbnails which have normally black bars but don't have them using these parameters will get black bars, which isn't something we'd like to have I think.

Why? If it's just Shorts thumbnails displaying this behaviour, they will always be in 16:9 aspect ratio, and should have black bars.

Are you saying the proposed code change will affect non-Shorts thumbnails as well, and there is no way to prevent that?

[Donkey-Doug]

FYI, this completely fixes the intermittent, constant buffering issue for me.

[Stypox]

@opusforlife2 according to what @AudricV explained above, thumbnails are now extracted at their original size, instead of at the forced-16:9 size like before (which included black bars for shorts). So the extractor has only improved and there is no regression there. The only problem is that thumbnails in NewPipe are not rendered in an optimal way, cropping them instead of showing them whole. I'll open a PR about this.