Condor.Search.Index

Memory based index. This module contains functions which create, update and search index. Default implementation uses algorithms for english language (stemming, stop words etc.)

Functions in this module (for performance reasons) are based on unicode strings from module Data.Text.

Basic usage:

 import Condor.Search.Index (addDocument, search)
 import Condor.Commons.Document (docFromStrings)

 let idx = addDocument emptyIndex $ docFromStrings "My document 1" "This is a document content."
 search idx "content"
 ["My document 1"]

Documentation

data Index

Inverted index

Instances

Binary Index

An instance of Binary to encode and decode an IndexParams in binary

type Term = Text

Single term. Could be normalized word

addDocument :: Document -> Index -> Index

Add document to the index. This function uses algorithms for english language to split document content into index terms.

addDocTerms :: DocName -> [Term] -> Index -> Index

Add document to the index. This function should be used if document content should be splitted into terms with custom algorithms.

emptyIndex :: Index

Create empty index. This index will be configured for english language.

docCount :: Index -> Int

Get the number of documents in the index

search :: Index -> Text -> [DocName]

Search terms given as single string in the index This function uses algorithms for english language to split query into tokens.

searchTerms :: Index -> [Term] -> [DocName]

Search terms given as array in the index. This function should be used if query should be splitted into terms with custom algorithms

termCount :: Index -> Int

Get the number of terms in the index