Recoll is a desktop-oriented full-text search tool whose core purpose is finding documents by content and file name. It is based on the Xapian search engine library, adds a text extraction layer and a Qt graphical interface, and also supports a web frontend and command-line usage. It is positioned not as a code search SaaS, but as a local/self-hosted full-text retrieval system for personal files and knowledge repositories.
Recoll runs on Linux, MS Windows, MacOS, and most Unix-like systems. It can index regular files, archive contents, email attachments, and multi-layer nested containersβfor example, a Word attachment inside a mailbox inside an archive. On the query side, it supports Boolean, phrase, proximity, wildcard searches, filtering by file type and directory, and can also be used through a GUI query builder. It offers multiple interface options, including a Qt GUI, web interface, command line, GNOME Shell Search Plugin, KDE KIO, and krunner. Format support is broad: HTML, email, OpenOffice, Office Open XML, Abiword, SVG, Gnumeric, and more can be processed; some older Office formats, PDFs, or other file types may require external tools. Chinese uses n-gram indexing by default, and results may improve if a better word segmenter is installed.
The documentation clearly states that the Linux version is free and open source under the GPL license. We did not find pricing details for Windows or MacOS, nor any commercial edition, cloud-hosted service, or paid support pricing. For self-hosting, Recoll is naturally centered on local indexing, while the web frontend can be used for remote previewing and downloading; the documentation also covers deploying the WebUI with Apache.
Its strengths are that it is open source, controllable, and capable of handling a wide range of formats and container structures, making it well suited to complex local document search. The documentation, FAQ, how-tos, and man pages are fairly complete, and there is also an issue tracker and mailing list. It can handle large datasets as well: the text mentions someone managing 11 million documents with a 250GB index. The drawbacks are that some formats depend on external helper tools, and installation and troubleshooting are more demanding than with consumer-grade search tools. News items have also mentioned upgrade, packaging, and version-specific crash issues. Chinese search quality may require additional word segmentation support.
Recoll is suitable for developers, researchers, system administrators, legal/knowledge management users, and anyone who wants to keep indexes locally rather than uploading data to the cloud. The source text does not provide information on availability from China, so this is unknown; there is also no information about payment methods. If you only need instant Windows file-name search, Everything may be an option; for cross-platform desktop full-text search, DocFetcher is worth comparing; for server-side search, consider OpenSearch, Solr, and similar tools.
β This review is compiled from public sources and does not constitute a purchase recommendation. Verify all facts on the vendor's official site. Verify on recoll.org official site.
recoll.org is an France Dev Tools provider. TG4G tracks its product information, an overall rating of 7.0/10, and a China-accessibility score of China direct-connect friendly. Click "Visit Official Site" to reach recoll.org directly.