Beta 1 query plans

This page lists summary task and progress notes for the beta 1 release of the MKSearch query component. This is an archive page.

More ResultsRenderer types

There are currently two result renderer types, which are very basic; a TextResultsRenderer and XhtmlResultsRenderer. At least one further type, RssResultsRenderer is required to generate an RSS feed for a standing query.

One issue for the RSS feed is the need to include the query URI itself in the output, which will require a more sophisticated renderer interface.

Task progress: Refactored the original ResultRenderer types into a more generic form that are insulated from the specifics of the Sesame API.

• Completed text, XHTML and RSS result renderers.

Results paging

The results renderer interface originally had no method for paging the results of a query nor controlling the number of results per page.

• Completed result paging for XhtmlResultRenderer

Query test documents

The query features need simple test case documents to verify query interpretation and retreival.

Task progress:

• Prepared a set of "cat and dog" test documents and verified query processing.

Dynamic query form

The main search interface will initially present a simple subject query field and search button. To enable users to search on any relevant predicate, additional search fields will be called-up through a query builder form.

Task progress

• Completed JSP QueryBuilderTag and tested manually.
• Completed JSP QueryResultTag and tested manually.
• Completed basic JSP result pages for Text, RSS and XHTML format results.

Phrase search result navigation

Where quoted phrases are used in queries, page navigation URLs should maintain the full query syntax.

Task progress

• Amended the ServletQueryBuilder class to re-instate quotes in the query URI that are extracted by the getQueryTerms method.
• Hex-encoded phrase quotes are also maintained in the RSS channel feed URL.
• Phrase query navigation complete

Outstanding issues

Extended graph queries

The query builder does not handle extended RDF graphs, where a Dublin Core qualified term is used to refine a primary search field. For example, where an indexed document has a DC.Date property with a DCTERMS.W3CDTF encoding scheme, simple date queries are not matched. If no W3C date node were present, the match is successful.

Task progress:

• Extensive amendments made to the AbstractQueryBuilder and associated classes to develop union queries.
• Union queries with optional encoding schemes completed.
• Union queries with URI objects completed.
• Various enabling modifications to the Schema and SchemaProperty implementations.
• Completed extended graph query implementation.