semlix 3.1.0 documentation¶

semlix is based on Whoosh, which was created by Matt Chaput. semlix extends semlix with semantic search capabilities while maintaining full backward compatibility.

You can view the project and report issues on the semlix GitHub page.

Contents¶

Release notes
- semlix 3.1 release notes
- semlix 3.0 release notes
- semlix 2.x release notes
- semlix 1.x release notes
- semlix 0.3 release notes
Quick start
- A quick introduction
- The Index and Schema objects
- The IndexWriter object
- The Searcher object
Introduction to semlix
- About semlix
- What is semlix?
- What can semlix do for you?
- Getting help with semlix
Glossary
Designing a schema
- About schemas and fields
- Built-in field types
- Creating a Schema
- Modifying the schema after indexing
- Dynamic fields
- Advanced schema setup
How to index documents
- Creating an Index object
- Clearing the index
- Indexing documents
- Merging segments
- Deleting documents
- Updating documents
- Incremental indexing
- Clearing the index
How to search
- The Searcher object
- Results object
- Scoring and sorting
- Highlighting snippets and More Like This
- Filtering results
- Which terms from my query matched?
- Collapsing results
- Time limited searches
- Convenience methods
- Combining Results objects
Semantic Search
- Overview
- Quick Start
- Components
- Advanced Usage
- Performance Considerations
- Migration Guide
- API Reference
BM25 Index
- Overview
- Quick Start
- Components
- Advanced Features
- Configuration
- Performance Tuning
- Migration
- Compatibility
- Limitations
- Examples
- Performance Comparison
- See Also
Unified Index
- Overview
- Quick Start
- Searching
- Components
- Advanced Features
- Configuration
- Migration
- Performance
- Examples
- Best Practices
- Troubleshooting
- See Also
Migration Guide
- Overview
- Migration Scenarios
- Quick Start
- Detailed Migration
- Migration Strategies
- Performance Considerations
- Compatibility
- Data Integrity
- Common Issues
- Examples
- Best Practices
- See Also
Parsing user queries
- Overview
- Using the default parser
- Common customizations
- Advanced customization
The default query language
- Overview
- Individual terms and phrases
- Boolean operators
- Fields
- Inexact terms
- Ranges
- Boosting query elements
- Making a term from literal text
Indexing and parsing dates/times
- Indexing dates
- Parsing date queries
- About time zones and basetime
- Date parser notes
- Limitations
Query objects
About analyzers
- Overview
- Using analyzers
- Advanced Analysis
Stemming, variations, and accent folding
- The problem
- Stemming
- Variations
- Lemmatization
- Character folding
Indexing and searching N-grams
- Overview
Sorting and faceting
- Overview
- Sorting
- Grouping
- Facet types
- MultiFacet
- Missing values
- Using overlapping groups
- Using a custom sort order
- Expert: writing your own facet
How to create highlighted search result excerpts
- Overview
- Requirements
- How to
- The character limit
- Customizing the highlights
- Highlighter object
- Speeding up highlighting
- Using the low-level API
Query expansion and Key word extraction
- Overview
- Usage
- Expansion models
“Did you mean… ?” Correcting errors in user queries
- Overview
- Pulling suggestions from an indexed field
- Pulling suggestions from a word list
- Merging two or more correctors
- Correcting user queries
Field caches
- Customizing cache behaviour
- Creating a custom caching policy
Tips for speeding up batch indexing
- Overview
- StemmingAnalyzer cache
- The limitmb parameter
- The procs parameter
- The multisegment parameter
Concurrency, locking, and versioning
- Concurrency
- Locking
- Versioning
Indexing and searching document hierarchies
- Overview
- Using nested document indexing
- Using query-time joins
semlix recipes
- General
- Analysis
- Searching
- Shortcuts
- Sorting and scoring
- Results
- Global information
semlix API
- analysis module
- codec.base module
- collectors module
- columns module
- fields module
- filedb.filestore module
- filedb.filetables module
- filedb.structfile module
- formats module
- highlight module
- support.bitvector module
- index module
- lang.morph_en module
- lang.porter module
- lang.wordnet module
- matching module
- qparser module
- query module
- reading module
- scoring module
- searching module
- sorting module
- spelling module
- support.charset module
- support.levenshtein module
- util module
- writing module
Technical notes
- How to implement a new backend
- filedb notes

Indices and tables¶