[hibernate-issues] [Hibernate-JIRA] Issue Comment Edited: (HSEARCH-414) Add support for a Lucene query builder using a fluent API

[ http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-414?pag... ] Emmanuel Bernard edited comment on HSEARCH-414 at 4/16/10 11:19 AM: -------------------------------------------------------------------- I've made significant changes (as in break pretty much everything :) ). Let me know what you think It looks cleaner now. You have several well defined layers: - chose the type of query (exact, fuzzy, wildcard) - (optional) customize query level operations (boost, constant score, filter) - select the targeted field - (optional) customize field level operations (boost, ignoreAnalyzer) - select the text to match - create the Lucene query I plan to add multi field queries as well. Questions: - does the query level vs field level distinction make sense? - fuzzy() or approximate() ? Examples: {code} final QueryBuilder monthQb = fts.getSearchFactory() .buildQueryBuilder().forEntity( Month.class ).get(); Query //term query based on several words query = monthQb.exact().onField( "mythology" ).matches( "colder darker" ).createQuery(); //fuzzy search with custom threshold and prefix query = monthQb .fuzzy() .threshold( .8f ) .prefixLength( 1 ) .onField( "mythology" ) .matches( "calder" ) .createQuery(); //wildcard query query = monthQb .wildcard() .onField( "mythology" ) .matches( "mon*" ) .createQuery(); //combined query, January and february both contain whitening but February in a longer text query = monthQb .bool() .should( monthQb.exact().onField( "mythology" ).matches( "whitening" ).createQuery() ) .should( monthQb.exact().onField( "history" ).matches( "whitening" ).createQuery() ) .createQuery(); {code} was (Author: emmanuel): I've made significant changes (as in break pretty much everything :) ). Let me know what you think It looks cleaner now. You have several well defined layers: - chose the type of query (exact, fuzzy, wildcard) - (optional) customize query level operations (boost, constant score, filter) - select the targeted field - (optional) customize field level operations (boost, ignoreAnalyzer) - select the text to match - create the Lucene query I plan to add multi field queries as well. Questions: - does the query level vs field level distinction make sense? - fuzzy() or approximate() ? Examples: {code} final QueryBuilder monthQb = fts.getSearchFactory() .buildQueryBuilder().forEntity( Month.class ).get(); Query //term query based on several words query = monthQb.exact().onField( "mythology" ).matches( "colder darker" ).createQuery(); //fuzzy search with custom threshold and prefix query = monthQb .fuzzy() .threshold( .8f ) .prefixLength( 1 ) .onField( "mythology" ) .matches( "calder" ) .createQuery(); //wildcard query query = monthQb .wildcard() .onField( "mythology" ) .matches( "mon*" ) .createQuery(); //combined query, January and february both contain whitening but February in a longer text query = monthQb .bool() .should( monthQb.exact().onField( "mythology" ).matches( "whitening" ).createQuery() ) .should( monthQb.exact().onField( "history" ).matches( "whitening" ).createQuery() ) .createQuery(); {code} -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira