The Lucid query parser is designed as a replacement for the existing Apache Lucene and Solr query parsers to produce good results with queries in any format, extracting the maximum possible information from the user entry to produce the best matches. The user (or system administrator) does not need to indicate what type of query it is: query interpretation is "modeless". The system will never produce an error message in response to a query and will always make the best interpretation possible. The system will match documents with alternative word forms (for example, singulars/plurals; this is on by default), expand synonyms, spell-check queries and handle queries in various languages (English by default) as configured by the administrator.
Features
The basic features of the Lucid query parser include all those of the traditional Apache Lucene and Apache Solr DisMax query parsers and is designed to work well for users accustomed to the search syntax used by popular Web search engines.
The Lucid query parser also includes a number of features not found in some or all of these other query parsers, including:
- Relational operators (comparative operators) - '<', '<=', '==', '!=', '>=', '>'
- More flexible range searches
- Implicit AND operator when no explicit operators are present
- The ALL pseudo-field to search across all fields
- Automatic bigram and trigram relevancy boosting
- Support for natural language queries
- Support for a wide variety of date formats
- Support for advanced proximity operators - NEAR, BEFORE, AFTER
- Support for multi-word synonyms
- Enhanced support for hyphenated terms
- Case-insensitive field names
- Sticky field names
- Good results even in the presence of query syntax errors with automatic error recovery
- Intelligent out-of-the-box settings for best results
- Wide range of configuration options for special needs
Simply put, the Lucid query parser is capable of handling existing Lucene and Solr queries and many Web search engine queries.
