Liquidsoap 2.3.2

Not an april's fool joke! 😄

This is the second bugfix release of the 2.3.x release branch of liquidsoap!

⚠️ This release includes some important fix with the long-term behavior of scripts that create dynamic sources such as crossfades and etc. ⚠️

We highly suggest to update your script to this version if you are on the 2.3.x release!

Other than that, this release includes a collection of nice bugfixes and other optimization.

🚨 While we have your attention: the 5th Liquidshop is taking place on Friday, June 13th! Please consider attending and/or presenting your project, community or topic of interest! More info: https://www.liquidsoap.info/liquidshop/5/

2.3.2 (2025-04-01) 🃏

New:

• Added support for multiple metadata fields in
  ogg and flac metadata
• Added support for track-level REM ALBUM in cue file parsing
  (#4381)

Changed:

• Added "pic" to list of excluded metadata for automatic charset conversion.
• Added settings.charset.max_string_length setting to prevent automatic charset
  conversions of strings over that length.

Fixed:

• Optimized CPU usage (#4369, #4370)
• Fixed empty initial HLS segment (#4401)
• Fixed support for duration metadata in image decoder (#4397)
• Fixed cue-out bug in cue file parsing (#4381)
• Bring back parse error location. (#4362)
• Fixed SRT encoding when restarting a stream with reverse data flow
  (#4399)
• Make sure that audioscrobbler on_track/on_end operations are
  sent to a asynchronous task queue.
• Fixed resources accumulation leading to catchup when using crossfade
  (#4419, #4410)
• Fixed source reselection logic issue that was causing crashes when using
  switch and fallback operators (#4420)
• Fixed self-sync logic with pulse audio outputs (#4429)
• Fixed script caching on windows.

Liquidsoap Rolling Release 2.3.x

This release provides liquidsoap assets before they are published as a new versioned release.

You can use it to install the latest stable code before it is published and test/prepare your production environment for it.

Rolling releases can also be useful for us to quickly detect and report bugs before the final published release!

⚠️ Warning ⚠️
Assets in this release will be deleted. If you are looking for permanent links to release assets, please head over to https://github.com/savonet/liquidsoap-release-assets/releases

For more details about our release process, please checkout https://github.com/savonet/liquidsoap#release-details
\n\n

2.3.2 (unreleased)

New:

• Added support for multiple metadata fields in
  ogg and flac metadata
• Added support for track-level REM ALBUM in cue file parsing
  (#4381)

Changed:

• Added "pic" to list of excluded metadata for automatic charset conversion.
• Added settings.charset.max_string_length setting to prevent automatic charset
  conversions of strings over that length.

Fixed:

• Optimized CPU usage (#4369, #4370)
• Fixed empty initial HLS segment (#4401)
• Fixed support for duration metadata in image decoder (#4397)
• Fixed cue-out bug in cue file parsing (#4381)
• Bring back parse error location. (#4362)
• Fixed SRT encoding when restarting a stream with reverse data flow
  (#4399)
• Make sure that audioscrobbler on_track/on_end operations are
  sent to a asynchronous task queue.
• Fixed resources accumulation leading to catchup when using crossfade
  (#4419, #4410)
• Fixed source reselection logic issue that was causing crashes when using
  switch and fallback operators (#4420)
• Fixed self-sync logic with pulse audio outputs (#4429)
• Fixed script caching on windows.

Liquidsoap 2.3.1

⚠️ Wanna help us know our users and plan the next release cycle? Fill up this form ⚠️

This is the first bugfix release of the liquidsoap 2.3.x release cycle!

So far, the 2.3.0 release has shown to be a pretty solid one! Therefore, this release only has minor bugfixes, most importantly about alsa devices and flac encoder.

Along with it are a couple of exciting features to keep things moving:

• The lastfm scrobbler implementation has been fixed and came with support for xml parsing and rendering.
• The CUE file parser for playlists has been rewritten natively and now supports many more dialects of the CUE file.

So, if you have a bunch of 90s CDs ripped with EAC that you want to scrobble to lastfm, this is your release! 😅

Also, we recommend that everyone else update to this version as well. It contains more bug fixes that may be affecting you and should be just as stable as 2.3.0!

⚠️ Warning ⚠️
Assets in this release may be deleted or updated. If you are looking for permanent links to release assets, please head over to https://github.com/savonet/liquidsoap-release-assets/releases

Full changelog:

2.3.1 (2025-02-05)

New:

• Added support for address resolution preference in SRT (#4317)
• Added global address resolution settings for SRT and Icecast (#4317)
• Added support for parsing and rendering XML natively (#4252)
• Added support for WAVE_FORMAT_EXTENSIBLE to the internal wav decoder.
• Added optional buffer_size parameter to input.alsa and output.alsa (#4243)
• Reimplemented audioscrobbler support natively using the more recent protocol (#4250)
• Added boolean getter to disable/enable normalization (#4308)

Changed:

• Make alsa I/O work with buffer size different than liquidsoap internal frame (#4236)
• Reimplemented CUE file parser in native liquidsoap script, added support for multiple files and EAC non-compliant extension (#1373, #4330)
• Make "song" metadata mapping to "title" metadata in input.harbor disabled when either "artist" or "title" is also passed. Add a configuration key to disable this mechanism. (#4235, #2676)
• output.icecast now re-sends the last metadata when connecting to the remote server unless explicitly disabled using the send_last_metadata_on_connect option (#3906)
• Add full explicit support for ipv4 vs. ipv6 resolution in SRT inputs and outputs, add global settings.srt.prefer_address and settings.icecast.prefer_address (#4317)
• Added generic SRT socket get/set API. Added new socket options, including latency and ipv6only.

Fixed:

• Fixed request resolution loop when enabling both autocue and replaygain metadata resolvers (#4245, fixed in #4246)
• Fixed flac encoding segfault (#4286, #4274)
• Fixed source last_metadata not being properly updated (#4262)
• Convert all ICY (icecast) metadata from input.http to utf8.
• Fixed inotify unwatching due to GC cleanup (#4275)
• Fixed delay initial conditions (#4281)

Liquidsoap 2.3.0

⚠️ Wanna help us know our users and plan the next release cycle? Fill up this form ⚠️

We are stoked to announce the release of liquidsoap 2.3.0! 🎉 ✨

This release brings a lot of exciting changes. As usual, we recommend to first test it in a staging environment before deploying it to production. Migration notes for this release can be found here.

A lot of hard engineering work went into this release! The largest changes should that happened during this cycle were under the hood:

Complete re-implementation of the streaming cycle and clocks

This change allowed us to prepare the codebase to use OCaml's new multicore support in future releases and get rid of design decisions that were 20 years old now. Back then, we had no idea how things would evolve! In particular, we had to account for new form of content such as raw video and encoded content.

Instead of filling a partial frame during the streaming cycle, we are now manipulating chunks of immutable content, concatenating them to build a frame on each streaming cycle. This is much more inline with the logic underlying video and encoded content where pieces of content can happen at random time, according to their presentation timestamp (PTS) or decoder timestamp (DTS).

Abstracting content this way has allowed us to finally get rid of frame breaks doubling as both track marks and partial filling marks. With the new content model, track marks and metadata are first-class content and part of the frame generation.

If you have ever encountered any of the infamous #get_frame didn't add exactly one break! fatal error, you know what we mean. This early design decision had plagued us for a while and it's wonderful to see it gone!

Along with these changes, we also reimplemented from scratch our notion of clocks, the abstract API that runs the main streaming cycles.

While clocks are still a pretty complex topic (self-sync, active vs. passive etc), we hope that this rewrite will clarify things. It was a good opportunity to clean up our implementation and use the experience gained about it in the recent years.

With these changes, clocks are now surfaced to the user via a clock method on each source. This makes it possible to directly access source clocks, animate them in your script, etc. We've also started to support synchronous streaming scripts, which should make it possible to use liquidsoap as a CLI for offline processing.

Script caching and memory optimization

Another important new feature with this release is the ability to cache the standard library and user scripts. This allows liquidsoap to run much faster. More details are available in this blog post

Incidentally, script caching also has very important benefits for memory usage. This is documented here.

This feature is probably the most important one of this release for most users!

Internal autocue

Autocue, the method for computing the perfect crossfade has also been polished in this release. We have now switched to our internal implementation, which is available when enabling our ffmpeg support.

You should be able to enable it by adding this to your script:

enable_autocue_metadata()

With this, all requests should resolve with the appropriate metadata so that any crossfade produces perfect transitions!

Remember that this feature requires CPU-heavy computations on each track!

A lot of testing went into this feature so we are really thankful to our users for the support and patience!

Other new features

Here's a list of other notable changes in this release:

• Frame duration was dropped to 0.02s. This should enhance streams latency and was made possible thanks to the new streaming implementation. See #4033 for a discussion about the trade-offs involved with this setting.
• New support for video size-independent placement of video elements. This makes it possible to design video streams with different elements whose placement is independent of the final video resolution. See this blog post for an example of it.
• Support for NDI output. We always try to extend the scope of applications that can be interfaced with liquidsoap. In the future, we would like to add support for NDI inputs and extend the supported formats.

The rest of the changelog is included below.

As with each release, we want to thank all our users for their patience, early testing and collaboration to make this release possible. It's a testament to their help that we are able to ship this release with such a level of confidence despite all the major changes that happened in the code base.

Special thanks go to: @vitoyucepi, @gAlleb, @RM-FM, @Moonbase59, @Russsgithub, @gilou, Xogium, the whole AzuraCast and Radio France team and everyone who helped reporting, fixing and confirming bugs.

Thanks y'all!

2.3.0 (2024-11-27)

New:

• Rewrote the streaming API to work with immutable frame content. This
  should greatly impact impredictable side-effect of the previous models w.r.t.
  track marks, content sharing and more. This also impacts multiple operators
  behavior. Mostly, things should be roughly the same with differences around
  behaviors related to track marks (source.on_track and etc). (#3577)
• Added script caching layer for faster script startup time. See: https://www.liquidsoap.info/blog/2024-06-13-a-faster-liquidsoap/ for details (#3924, #3949, #3959 and #3977)
• Rewrote the clock/streaming loop layer. This prepares our streaming system to
  support multicore when the OCaml compiler is mature enough to allow it. Clocks
  are now attached to sources via their clock methods. Returned value is a stripped
  down clock variable. Users can use the clock function to retrieve the full
  methods, e.g. s = sine(); c = clock(s.clock). This value has advanced functions
  for clock control such as start/stop, ticks and self_sync to check for
  self-sync. (#3781)
• Allow frames duration shorter than one video frames, typically values under 0.04s.
  Smaller frames means less latency and memory consumption at the expense of
  a higher CPU usage (#3607)
• Change default frame duration to 0.02s (#4033)
• Optimized runtime (#3927, #3928, #3919)
• Added NDI output support (#4181)
• Added finally to execute code regardless of whether or not an exception is raised
  (see: #3895 for more details).
• Added support for Spinitron submission API (#4158)
• Removed gstreamer support. Gstreamer's architecture was never a good fit for us
  and created a huge maintenance and debugging burden and it had been marked as
  deprecated for a while. Most, if not all of its features should be available using
  ffmpeg. (#4036)
• Removed taglib support. It is superseded by the internal ocaml-metadata module
  and taglib, with its dependency on the C++ runtime library, has been causing issues
  with binary builds portability and crashes with the (not yet supported) OCaml 5
  compiler. (#4087)
• Added video.canvas to make it possible to position video elements independently
  of the rendered video size (#3656, blog post)
• Added cover manager from an original code by @vitoyucepi (#3651)
• Added non-interleaved API to %ffmpeg encoder, enabled by default when only
  one stream is encoded.
• Allow trailing commas in record definition (#3300).
• Added metadata.getter.source.float (#3356).
• BREAKING: Added duration and ticks to metadata available when computing HLS segment names (#4135)
• Added optional main_playlist_writer to output.file.hls and
  derivated operator (#3484)
• Added is_nan, is_infinite, ceil, floor, sign and round (#3407)
• Added %track.drop to the %ffmpeg encoder to allow partial encoding
  of a source's available tracks (#3480)
• Added let { foo? } = ... pattern matching (#3481)
• Added metadata.replaygain method to extract unified replay gain value from metadata (#3438).
• Added metadata.parse.amplify to manually parse amplify override metadata.
• Added compute parameter to file.replaygain to control gain calculation (#3438).
• Added compute parameter to enable_replaygain_metadata to control replay gain calculation (#3438).
• Added copy: protocol (#3506)
• Added file.touch.
• Added support for sqlite databases (#3575).
• Added string.of_int and string.spaces.
• Added list.assoc.nullable.
• Added source.cue (#3620).
• Added string.chars (#4111)
• Added atomic file write operations.
• Added new macos_say speech synthesis protocol. Make it the default implementation for the say:
  protocol on macos.
• Added settings.request.timeout to set the request timeout globally.

Changed:

• Reimplemented request.once, single and more using source.dynamic. Removed experiment
  flag on source.dynamic. The operator is considered stable enough to define advanced sources
  but the user should be careful when using it.
• Mute SDL startup messages (#2913).
• int can optionally raises an error when passing nan or infinity, int(infinity)
  now returns max_int and int(-infinity) returns min_int. (#3407)
• Made default font a setting (#3507)
• Changed internal metadata format to be immutable (#3297).
• Removed source.dump and source.drop in favor of safer request.dump and request.drop.
  source.{dump, drop} can still be implemented manually when needed and with the proper
  knowledge of what's going on.
• Allow a getter for the ...

Liquidsoap Rolling Release 2.2.x

This release provides liquidsoap assets before they are published as a new versioned release.

You can use it to install the latest stable code before it is published and test/prepare your production environment for it.

Rolling releases can also be useful for us to quickly detect and report bugs before the final published release!

⚠️ Warning ⚠️
Assets in this release will be deleted. If you are looking for permanent links to release assets, please head over to https://github.com/savonet/liquidsoap-release-assets/releases

For more details about our release process, please checkout https://github.com/savonet/liquidsoap#release-details

2.2.5 (unreleased)

New:

• Added various ffmpeg timestamps when exporting ffmpeg metadata from filters.
• Added enable_autocue_metadata and autocue: protocol to automatically compute cue points and crossfade parameters (#3753, @RM-FM and @Moonbase59)
• Added db_levels method to blank.* sources (#3790)
• Added excluded_metadata_resolvers to request.create to make it possible to selectively disable specific metadata resolvers when resolving requests.

Changed:

• Allow to disable http.* url normalization. Add warning when url normalization changes the url (#3789)

Fixed:

• Fixed source.drop and source.dump clock initialization.
• Fixed bogus report of non-monotonous PTS content when using raw ffmpeg content.
• Fixed streaming errors when disconnecting input.harbor.
• Fixed issues with rendered id3v2 frame that contain binary data (#3817)

Liquidsoap 2.2.5

This is the fifth release of the 2.2.x release cycle, following the initial 2.2.0 release.

This release comes with some minor bugfixes and changed feature and one big chunk of new functionality around crossfading: autocue.

We will post a longer article about this on our blog. The gist of the functionality is that it should allow to compute excellent crossfade transition parameters in your sources, at the expense of an increased CPU usage.

Users can register their own autocue implementations and there is a default one that is available if you have compiled liquidsoap with its ffmpeg support.

In this case, you can try to add:

enable_autocue_metadata()

To your script and see what happens!

---

2.2.5 (2024-05-01) (Mayday!)

New:

• Added enable_autocue_metadata and autocue: protocol to automatically compute cue points and crossfade parameters (#3753, #3811, @RM-FM and @Moonbase59)
• Added various ffmpeg timestamps when exporting ffmpeg metadata from filters.
• Added db_levels method to blank.* sources (#3790)
• Added excluded_metadata_resolvers to request.create to make it possible to selectively disable specific metadata resolvers when resolving requests.
• Normalized expected API from autocue, allow multiple implementation and adapt cross/crossfade to work with it out of the box with workaround for short tracks.
• Added private and swapped memory reporting when compiled with mem_usage.
• Added priorities to metadata deocoders, allows finer-control of metadata overriding. (#3887)

Changed:

• Allow to disable http.* url normalization. Add warning when url normalization changes the url (#3789)
• Add namespace and optional source IDs to mix.

Fixed:

• Prevent request metadata from overriding root metadata (#3813)
• Fixed source.drop and source.dump clock initialization.
• Fixed bogus report of non-monotonous PTS content when using raw ffmpeg content.
• Fixed streaming errors when disconnecting input.harbor.
• Fixed issues with rendered id3v2 frame that contain binary data (#3817)
• Fixed memory leaks with SRT listen socket polling callbacks.
• Fixed %ffmpeg copy muxer logic with some audio/video streams (#3840)
• Fixed duration metadata calculation in the presence of cue_in/cue_out metadata.

Liquidsoap 2.2.4

This is the fourth follow-up release from the 2.2.x branch. This release comes with important bug fixes related to memory consumption and audio quality.

You can refer to the liquidsoap blog for more details about this release: https://www.liquidsoap.info/blog/2024-02-07-liquidsoap-v2.2.4/

As usual, we recommend any user on the 2.2.x branch to upgrade to this version. However, this time, it is recommended to read about the non-trivial changes it brings and to test it with your staging or pre-prod environment!

Full release changelog:

2.2.4 (2024-02-04)

New:

• Added support for id3v2 metadata in output.*.hls when using %mp3, %shine or %fdkaac encoders (#3604)
• Added option to set preferred address class (ipv4, ipv6 or system default) when resolving hostnames in http transports and output.icecast
• Added self_sync option to input.srt to accommodate for streams in file mode (#3684)
• Added curve parameter to fade functions and liq_fade_{in,skip,out}_curve metadata override (#3691)
• Added delay parameter to fade functions to make it possible to add delay before fade happens. Add liq_fade_{in,skip,out}_delay metadata override.
• Added single_track option to allow sequence to play each source until they are unavailable while keeping track marks.

Changed:

• cue_cut operator has been removed. Cueing mechanisms have been moved to underlying file-based sources. See migration notes for more details.

Fixed:

• Fix pop/click at the end of fade out/in (#3318)
• Fix audio/video synchronization issues when decoding live streams using ffmpeg.
• Fix issues with TLS connecting clients not being properly timed out (#3598)
• Make sure reconnection errors are router through the regulat on_error callback in output.icecast (#3635)
• Fixed discontinuity count after a restart in HLS outputs.
• Fixed file header logic when reopening in output.file (#3675)
• Fixed memory leaks when using dynamically created sources (input.harbor, input.ffmepg, SRT sources and request.dynamic)
• Fixed invalid array fill in add (#3678)
• Fixed deadlock when connecting to a non-SSL icecast using the TLS transport (#3681)
• Fixed crash when closing external process (#3685)

Liquidsoap 2.2.3

This is the third and, hopefully, near final point release of the liquidsoap 2.2.x release cycle.

The most important changes in this release are:

• A fix involving shared active sources leading to infinite loop and CPU usage going to 100%. This impacted in particular scripts doing recording of a streamer being sent via another output. See: #3537 for such an example.
• Several SRT-related fixes w.r.t file descriptor leaks and the caller connection not being properly processed asynchronously.

This release should be safe to upgrade to for any script on the 2.2.x branch. But, as usual, don't forget to test it with a staging environment first!

2.2.3 (2023-12-14)

New:

• Added register_telnet option to all outputs to
  prevent telnet commands registration (#3545)
• Added file.touch

Changed:

• BREAKING: setting.playlists has been renamed
  setting.playlist, settings.playlist.mime_types
  now refers to a full list of records (#3561)

Fixed:

• Cleanup telnet commands on output shutdown,
  prevent commands registration for internal
  outputs (#3545, fixes: #3541)
• Allow overwrite in youtube-dl/yt-dlp call,
  remove use of unix-specific process call.
  (#3563)
• Fixed playlist parsers not being registered
  when evaluatin top-level script (#3553)
• Fixed infinite loop on some active sources
  (#3560, fixes: #3544, and #3537)
• Fixed fd leak on failed connection in srt
  caller mode (#3583)
• Fixed high CPU usage on failed connection
  in srt caller mode (#3583, #3573)
• Prevent outputs from keeping the last 10 cover
  metadata (#3504)

Liquidsoap 2.2.2

This release is the second bugfix release of the 2.2.x release cycle. The most important fix is related to pop/clicks heard during crossfade (#3318) and the ffmpeg decoder failing with files and streams with unknown codecs.

This is the first release since we improved developers tools so feel free to also take this opportunity to update your developer setup!

As usual, we recommend trying this version with a staging/preview system before putting it on production. Beside that, this release should improve the stability of all scripts running on the 2.2.x release branch.

We want to thank all contributors, bug reporters and patient users who have helped us bring all these improvements!

2.2.2 (2023-11-02)

New:

• Added string.escape.html (#3418, @ghostnumber7)
• Add support for getters in arguments of blank.detect (#3452).
• Allow float in source content type annotation so that it possible to write: source(audio=pcm(5.1))

Changed:

• Trim urls in input.ffmpeg by default. Disable using trim_url=false (#3424)
• Automatically add HLS-specific ffmpeg parameters to %ffmpeg encoder (#3483)
• BREAKING: default on_fail removed on playlist (#3479)

Fixed:

• Allow channel_layout argument in ffmpeg encoder to set the number of channels.
• Improved support for unitary minus, fix runtime call of optional methods (#3498)
• Fixed map.metadata mutating existing metadata.
• Fixed reloading loop in playlists with invalid files (#3479)
• Fixed main HLS playlist codecs when using mpegts (#3483)
• Fixed pop/clicks in crossfade and source with caching (#3318)
• Fixed pop/clicks when resampling using libsamplerate (#3429)
• Fixed gstreamer compilation. Remember that gstreamer features are DEPRECATED! (#3459)
• Fixed html character escaping in interactive.harbor (#3418, @ghostnumber7)
• Fixed icecast not reconnecting after erroring out while closing connection in some circumstances (#3427)
• Fixed parse-only mode (#3423)
• Fixed ffmpeg decoding failing on files with unknown codecs.
• Fixed a crash due to wait_until timestamp being in the past when using posix-time2
• Make sure that temporary files are always cleaned up in HLS outputs (#3493)

Liquidsoap 2.2.1

This is the first bugfix release from the liquidsoap 2.2.x release cycle! It contains important fixed in the HLS outputs and muxer operator. This release should be safe to upgrade from 2.2.0 with some minor breaking changes on the HLS output on_file_change callback to account for the atomic file operations.

2.2.1 (2023-09-05)

Changed:

• BREAKING: on HLS outputs, on_file_change events are
  now "created", "updated" and "deleted", to better
  reflect the new atomic file operations (#3284)
• Added compact argument to the http.response.json function. http.response.json will produce minified JSON by
  default. Added a newline symbol to the end of the JSON data produced by http.response.json. (#3299)
• Bumped internal ogg decoder to make sure that it is used over the ffmpeg decoder whenever possible.
  FFmpeg has issues with metadata in chained streams which needs to be fixed upstream. Unfortunately,
  input.http can only use the ffmpeg decoder at the moment.
• Cleanup output.file encoding and file handling logic (#3328)
• Added ratio to source.{dump,drop} to make it possible to control its CPU peaks.
• Enhanced clock error reporting (#3317)

Fixed:

• Fixed slow memory leak in muxer operator (#3372, #3181, #3334)
• Fixed discontinuity logic error in HLS outputs after a restart.
• Fixed HTTP response status in output.harbor (#3255)
• Make sure main HLS playlist is regenerated after being
  unlinked (#3275)
• Fixed hard crash on icecast disconnection errors.
• Fix output.harbor encoder header when encoding with
  %ogg, %vorbis and etc. (#3276)
• Fixed quality argument parsing in ffmpeg encoders (#3267)
• Make all HLS file write atomic (#3284)
• Allow seek and cue operators to work with muxed sources
  using a single underlying source (#3252)
• Fixed export of cover art metadata (#3279)
• Remove use of stereo: protocol in say: protocol:
  this is now handled automatically by the decoder and generates latency via
  high CPU usage peak.
• Fixed output.file reopening with flac encoding (#3328)