[ http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-414?pag... ] Sanne Grinovero commented on HSEARCH-414: ----------------------------------------- seems nice :) but you are the DSL guru -- 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

[ http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-414?pag... ] Sanne Grinovero commented on HSEARCH-414: ----------------------------------------- On first impact, I got a bit confused about "term().on(" ; you mean a TermQuery right? So wouldn't monthQb.term().on( "mythology_ngram" ).matches( "snobored" ).createQuery(); be more readable as monthQb.termQuery( "mythology_ngram" ).matches( "snobored" ).create(); sorry I have the vague idea we talked about something like that, but forgot the reasons if any; term().on() doesn't seem bad either, just a matter of taste :) I'll give it a try and get you some better feedback. Tests are very readable, are you going to add some using booleanQuery to show a composite query? I've been looking into the new Lucene's features to create efficient RangeQuery on number/dates, that definitely needs some helpers - just a thought for future. Is it picking the right analyzer automatically (from field name) ? -- 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

[ http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-414?pag... ] Emmanuel Bernard updated HSEARCH-414: ------------------------------------- Fix Version/s: (was: 3.2.0.Beta3) 3.3.0 -- 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

[ 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

[ http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-414?pag... ] Emmanuel Bernard updated HSEARCH-414: ------------------------------------- Fix Version/s: (was: 3.3.0) 3.3.0.Alpha1 -- 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