macOS/TagFetcher: improve Opus matching reliability (AcoustID POST + MusicBrainz fallback + safer fingerprinting) by OpsKonekt · Pull Request #2002 · strawberrymusicplayer/strawberry

OpsKonekt · 2026-02-13T20:26:41Z

Summary

This PR improves Tag Fetcher reliability for .opus files (notably files downloaded via yt-dlp) on macOS, while keeping the app self-contained (no external CLI dependency at runtime).

Problem

Some files failed in Tag Fetcher with errors like:

AcoustID HTTP 400 / invalid fingerprint
"No decoded audio samples were produced by GStreamer"
No useful fallback when AcoustID returned no recording IDs, especially when embedded tags were missing

What changed

1) AcoustID request robustness

Switched lookup request from query-style GET to form POST (application/x-www-form-urlencoded)
Improved API/network error parsing and diagnostics
Kept request semantics identical (format, client, duration, meta, fingerprint)

2) MusicBrainz fallback when AcoustID fails

Added metadata search fallback path in TagFetcher:
- If AcoustID returns no MBIDs, perform a MusicBrainz recording search
Added MusicBrainzClient::StartSearchRequest(...)
Added query normalization/tokenization and result scoring (title/artist/album/duration heuristics)
Preserved existing MBID path behavior

3) Better handling of missing tags

If title/artist tags are missing, derive search metadata from filename
Supports common filename patterns like track-number + artist + title

4) Better user diagnostics

Track selection error page now shows technical details from backend failures
Keeps details readable as plain text

5) Fingerprinting hardening + memory safety

Hardened GStreamer decode path in Chromaprinter
Added explicit in-memory PCM cap to avoid unbounded growth
Added safer pad checks and clearer failure reasons

6) macOS bundle/runtime fixes

Improved bundled GStreamer plugin discovery on startup
Added plugin/runtime packaging adjustments needed for matroska/webm/opus flow
Added libgstriff bundling when required by libgstmatroska
Registry filename now includes executable timestamp to avoid stale cache reuse across rebuilds

Validation

Build: cmake --build build-m1-full -j8 (passes)
Manual tests on macOS Sonoma 14.6.1 (Apple Silicon):
- Tag Fetcher with .opus files with and without embedded tags
- AcoustID success path
- MusicBrainz fallback path
- Error diagnostics visibility in UI

Commit breakdown

musicbrainz: add metadata-search fallback when AcoustID fails
ui: show technical diagnostics in tag fetcher error page
chromaprinter: harden decoding and cap in-memory PCM usage
macos: bundle and discover GStreamer runtime dependencies reliably

Bumps [vmactions/freebsd-vm](https://github.com/vmactions/freebsd-vm) from 1.4.1 to 1.4.2. - [Release notes](https://github.com/vmactions/freebsd-vm/releases) - [Commits](vmactions/freebsd-vm@v1.4.1...v1.4.2) --- updated-dependencies: - dependency-name: vmactions/freebsd-vm dependency-version: 1.4.2 dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com>

jonaski

Thanks, since there is a lot of changes here addressing different things, could you open a PR for each? This way it's easier to review.
I suggest to split into 5 different PR's similar to your commits:

Improved error handling
Chromaprint pipeline changes
Additional MusicBrainz fallback to tag search
The gstreamer environmental variable changes (but I'm not sure we need these)
macgstcopy

You should be able to create 4 separate branches and cherry-pick the relevant commit to each branch.
Also, please rebase the commit messages to include capitalized "ClassName:" as prefix if the changes only affect one class, otherwise the commit title should be without a prefix.
And was a AI tool used for any of these changes?

jonaski · 2026-02-18T22:34:57Z

This commit shouldn't be included here.

jonaski · 2026-02-18T22:36:12Z

 libgstisomp4
 libgstlame
+# Needed for WebM/Matroska containers that often wrap Opus streams.
+libgstmatroska


Matroska isn't currently built, it needs to be enabled in https://github.com/strawberrymusicplayer/strawberry-macos-dependencies/
how did you test this?

jonaski · 2026-02-18T22:37:14Z

  fi
 done

+# libgstmatroska depends on libgstriff, which is not always pulled automatically.


Then we need to investigate why, this seems like a workaround, isn't libgstriff-1.0.0.dylib linked?

jonaski · 2026-02-18T22:38:00Z

    Song original_song_;
    bool pending_;
    QString progress_string_;
+    // Carries technical diagnostics returned by TagFetcher for the error page.


Don't think we need too many comments like this, since no other variables are commented.

jonaski · 2026-02-18T22:38:49Z

  return GST_FLOW_OK;

 }
-


Avoid committing formatting changes in the same PR

jonaski · 2026-02-18T22:40:46Z

-  QString gst_registry_filename = StandardPaths::WritableLocation(StandardPaths::StandardLocation::AppLocalDataLocation) + QStringLiteral("/gst-registry-%1-bin").arg(QCoreApplication::applicationVersion());
+  const QFileInfo executable_info(QCoreApplication::applicationFilePath());
+  const qint64 executable_stamp = executable_info.exists() ? executable_info.lastModified().toSecsSinceEpoch() : 0;
+  // Include executable timestamp to avoid reusing stale GStreamer registries across rebuilds.


We only distribute a version once, so this shouldn't be necessary as far as I can tell. During development you can easily delete these manually. Is there a specific reason this is needed?

jonaski · 2026-02-18T22:52:12Z

Validation

Build: cmake --build build-m1-full -j8 (passes)

Manual tests on macOS Sonoma 14.6.1 (Apple Silicon):

Tag Fetcher with .opus files with and without embedded tags

AcoustID success path

MusicBrainz fallback path

Error diagnostics visibility in UI

I'm confused by this, does this mean you tested everything?
And, is it tested on macOS only, or Linux and Windows too?

OpsKonekt and others added 5 commits February 13, 2026 21:12

musicbrainz: add metadata-search fallback when AcoustID fails

1311792

ui: show technical diagnostics in tag fetcher error page

4b4ed52

chromaprinter: harden decoding and cap in-memory PCM usage

5aa49f4

macos: bundle and discover GStreamer runtime dependencies reliably

b0a7aa6

jonaski requested changes Feb 18, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

macOS/TagFetcher: improve Opus matching reliability (AcoustID POST + MusicBrainz fallback + safer fingerprinting)#2002

macOS/TagFetcher: improve Opus matching reliability (AcoustID POST + MusicBrainz fallback + safer fingerprinting)#2002
OpsKonekt wants to merge 5 commits intostrawberrymusicplayer:masterfrom
OpsKonekt:fix/macos-opus-tagfetcher-fallback

OpsKonekt commented Feb 13, 2026

Uh oh!

jonaski left a comment •

edited

Loading

Uh oh!

jonaski Feb 18, 2026

Uh oh!

jonaski Feb 18, 2026

Uh oh!

jonaski Feb 18, 2026 •

edited

Loading

Uh oh!

jonaski Feb 18, 2026

Uh oh!

jonaski Feb 18, 2026

Uh oh!

jonaski Feb 18, 2026

Uh oh!

jonaski commented Feb 18, 2026 •

edited

Loading

Validation

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

OpsKonekt commented Feb 13, 2026

Summary

Problem

What changed

1) AcoustID request robustness

2) MusicBrainz fallback when AcoustID fails

3) Better handling of missing tags

4) Better user diagnostics

5) Fingerprinting hardening + memory safety

6) macOS bundle/runtime fixes

Validation

Commit breakdown

Uh oh!

jonaski left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jonaski Feb 18, 2026

Choose a reason for hiding this comment

Uh oh!

jonaski Feb 18, 2026

Choose a reason for hiding this comment

Uh oh!

jonaski Feb 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jonaski Feb 18, 2026

Choose a reason for hiding this comment

Uh oh!

jonaski Feb 18, 2026

Choose a reason for hiding this comment

Uh oh!

jonaski Feb 18, 2026

Choose a reason for hiding this comment

Uh oh!

jonaski commented Feb 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Validation

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

jonaski left a comment •

edited

Loading

jonaski Feb 18, 2026 •

edited

Loading

jonaski commented Feb 18, 2026 •

edited

Loading