manticoresearch を Homebrew, Nix, zypper でインストール

プロジェクトリンク

• ホームページ
• リポジトリ
• 上流ドキュメント
• パッケージマネージャページ

コマンドとエイリアス

• index_converter
• indexer
• indextool
• searchd
• spelldump
• wordbreaker

プロジェクトの歴史

The official README says Manticore Search was forked from Sphinx 2.3.2 in 2017. The public GitHub repository was created in June 2017, and the README still carries the historical line from Sphinx to Manticore while presenting Manticore as its own database and search server.

The project is implemented primarily in C++ and keeps the search-daemon heritage visible through commands such as `searchd`, `indexer`, and configuration in `manticore.conf`. Over time, the README has expanded the positioning from full-text search to a broader search database with real-time indexing, SQL, JSON/HTTP APIs, vector search, replication, backup tooling, and client libraries.

採用の歴史

The README names Craigslist, Socialgist, PubChem, Rozetka, and others as users of Manticore for search and stream filtering. It also points to official package repositories, Docker images, Homebrew installation, cloud offerings, a forum, Slack, Telegram, and language clients, indicating adoption as both server software and developer-facing infrastructure.

By 2026 the repository had over ten thousand GitHub stars and hundreds of forks. That is a significant footprint for an open-source search engine outside the Elasticsearch/OpenSearch mainstream.

使われ方

Users run Manticore as a service or container, configure `searchd`, create tables, insert data, and query through MySQL-compatible SQL or HTTP JSON APIs. The manual documents explicit startup with `searchd --config /etc/manticoresearch/manticore.conf`, Linux service management, and Windows startup with `searchd -c manticore.conf`.

The README emphasizes full-text operators, ranking, filtering, faceting, geospatial search, hybrid search, KNN/vector search, spell correction, stream filtering, replication, backup and restore, data import from databases or files, and clients for languages including PHP, Python, JavaScript, TypeScript, Java, Elixir, Go, and Rust.

パッケージ好きにとっての重要性

Manticore matters to package maintainers because it is not just a CLI: it ships a daemon, indexer tools, config files, data directories, service integration, protocol compatibility, optional authentication, and language clients. Its Sphinx lineage also makes it part of the long-running history of self-hosted full-text search engines.

タイムライン

• 2017: Manticore Search forked from Sphinx 2.3.2 according to the official README.
• 2017: Public GitHub repository created.
• 2025: release-15.1.0 published.
• 2026: 27.1.5 release published.

Related projects

• Sphinx Search is the upstream project from which Manticore was forked.
• Manticore Columnar Library is listed by the README as supporting columnar storage.
• Official clients exist for PHP, Python, Python asyncio, JavaScript, TypeScript, Java, Elixir, Go, and Rust.

ソース

• https://github.com/manticoresoftware/manticoresearch
• https://github.com/manticoresoftware/manticoresearch#readme
• https://github.com/manticoresoftware/manticoresearch/releases/tag/27.1.5
• https://manual.manticoresearch.com/
• https://manual.manticoresearch.com/Security/Authentication_and_authorization
• https://manual.manticoresearch.com/Starting_the_server/Manually