Overview
Gustar provides a pragmatic, production‑ready way to add search to websites and document collections. It focuses on simple configuration, fast indexing, and readable query syntax, and integrates cleanly with existing Qddb data.
Key features
- Incremental crawls and re‑indexing.
- Compact, rank‑scored indices with stemming and stop‑word support.
- Fielded queries (e.g., title:..., path:...) and approximate phrase search.
- Fast boolean and range operators with query hints.
- Simple templates to render results with snippets and highlights.
- UNIX‑friendly: CLI tools and cron integration.
Typical workflow
- Configure your crawl roots and include/exclude rules.
- Run the crawler to build or update your index.
- Serve queries via the provided CGI or a small web handler.
- Tune ranking and snippets as needed.
Download
Grab the latest source and binaries from the Qddb FTP server:
Build instructions are included in README and platform notes in CHANGES.
Documentation
Configuration essentials
- Index roots: directories,
http://andhttps://URLs. - Ignore lists: file globs and robots.txt rules.
- Tokenizer: language, stemming, stop‑words.
- Ranking: term frequency, field boosts, freshness.
- Output: result templates, snippet length, highlight marker.
Query examples
- error handling — approximate phrase search
- title:qddb — title field
- qddb , ! gustar — boolean queries