lastgenre: Global and artist-based genre blacklist

[JOJ0]

Description

Forbids a list of genres or regex patterns configurable for an artist or globally (*). It uses a simple custom file format that could look like this:

fracture:
     ^(heavy|black|power|death)?\s?(metal|rock)$|\w+-metal\d*$
    progressive metal
gilles peterson:
    samba
    bossa nova
*:
    electronic

This configuration gives a practical example within: Artist "Fracture" is a Drum And Bass producer but a also a Metal band exists with the same name. last.fm returns genres for both artists. Removing Metal variants (Power Metal, Heavy Metal, ...) from the global whitelist file would result in not being able to fetch proper genres for any other Metal band in the user's library anymore.

A mixture of regex and simple genre names is possible since regex patterns are precompiled and if that fails, escaped literal strings are added to the blacklist.

This PR also includes some cleanup in the setup() method which got rather longish. It moves all file loading code to separate helper methods.

Notes

After several experiments with YAML and INI as the blacklist file format I came to the conclusion that it's just too much PITA trying to fit regex patterns within these formats and I gave up looking for the perfect format that doesn't stand in the way of userfriendliness and allowing all kinds of regex patterns withouth special escaping and formatting tricks.....I'm back from that hell and decide that a very simple custom format suits this usecase best.

This feature idea also came up during this discussion: #5721 (comment) and is on my personal wish list for ages.

My personal roadmap for the lastgenre plugin: #5915

To Do

• Documentation.
• Changelog.
• Tests.

[github-actions]

Thank you for the PR! The changelog has not been updated, so here is a friendly reminder to check if you need to add an entry.

[codecov]

Codecov Report

❌ Patch coverage is 95.52239% with 3 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (master@e3574a7). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
beetsplug/lastgenre/__init__.py	95.52%	1 Missing and 2 partials ⚠️

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[github-actions]

Thank you for the PR! The changelog has not been updated, so here is a friendly reminder to check if you need to add an entry.

[Copilot]

Pull Request Overview

This PR implements a blacklist feature for the lastgenre plugin to forbid specific genres globally or per-artist using regex patterns or plain text matching. It uses a custom file format to maximize compatibility with complex regex patterns and provides both global (*) and artist-specific blacklisting capabilities.

• Adds blacklist functionality with custom file format supporting regex patterns and plain genre names
• Refactors the setup() method by extracting file loading logic into separate helper methods
• Updates genre filtering to check blacklist patterns before accepting genres

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.

File	Description
beetsplug/lastgenre/init.py	Core implementation of blacklist functionality, file parsing, and genre filtering
test/plugins/test_lastgenre.py	Comprehensive test coverage for blacklist patterns, file format parsing, and error handling
docs/plugins/lastgenre.rst	Documentation for the new blacklist feature with examples and configuration options

[Copilot]

The file is opened in binary mode ('rb') but then decoded line by line. This could fail if the file contains multi-byte Unicode characters that span across buffer boundaries. Consider opening the file in text mode with explicit encoding or using a proper text decoder.

Suggested change

	with open(bl_filename, "rb") as f:
	for lineno, line in enumerate(f, 1):
	line = line.decode("utf-8").lower()
	with open(bl_filename, "r", encoding="utf-8") as f:
	for lineno, line in enumerate(f, 1):
	line = line.lower()

[JOJ0]

I'm not sure if I should follow this advice. I tried to orientate my code on how the simple text file whitelist.txt is loaded. Our guides only tell that paths to store in the database should be made sure to be bytes https://beets.readthedocs.io/en/latest/contributing.html#handling-paths

So actually I coud safely follow this advice. Some guidance please @wisp3rwind or @snejus

Thanks :-)

[Copilot]

Calling .lower() on the entire line before parsing could interfere with regex patterns that depend on case sensitivity. The case conversion should be done after separating the artist name from the patterns, and only applied to the artist name.

Suggested change

	line = line.decode("utf-8").lower()
	line = line.decode("utf-8")

[Copilot]

The _filter_valid_genres method calls _is_forbidden for every genre, which performs regex matching against potentially many patterns. Consider caching compiled regex patterns or implementing early termination strategies for better performance with large genre lists.