lucene-java-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
Subject svn commit: r610961 - in /lucene/java/trunk/contrib: queries/README.txt xml-query-parser/README.htm
Date Thu, 10 Jan 2008 22:38:39 GMT
Author: buschmi
Date: Thu Jan 10 14:38:38 2008
New Revision: 610961

Rename README files to uppercase letters


Added: lucene/java/trunk/contrib/queries/README.txt
--- lucene/java/trunk/contrib/queries/README.txt (added)
+++ lucene/java/trunk/contrib/queries/README.txt Thu Jan 10 14:38:38 2008
@@ -0,0 +1,27 @@
+This module contains a number of filter and query objects that add to core lucene.
+The "MoreLikeThis" class from the "similarity" module has been copied into here.
+If people are generally happy with this move then the similarity module can be deleted, or
at least a 
+"Moved to queries module..." note left in its place.
+==== FuzzyLikeThis  - mixes the behaviour of FuzzyQuery and MoreLikeThis but with special
+of fuzzy scoring factors. This generally produces good results for queries where users may
provide details in a number of 
+fields and have no knowledge of boolean query syntax and also want a degree of fuzzy matching.
The query is fast because, like
+MoreLikeThis, it optimizes the query to only the most distinguishing terms.
+==== BoostingQuery - effectively demotes search results that match a given query. 
+Unlike the "NOT" clause, this still selects documents that contain undesirable terms, 
+but reduces the overall score of docs containing these terms.
+==== TermsFilter -  Unlike a RangeFilter this can be used for filtering on multiple terms
that are not necessarily in 
+a sequence. An example might be a collection of primary keys from a database query result
or perhaps 
+a choice of "category" labels picked by the end user.
+Mark Harwood

Added: lucene/java/trunk/contrib/xml-query-parser/README.htm
--- lucene/java/trunk/contrib/xml-query-parser/README.htm (added)
+++ lucene/java/trunk/contrib/xml-query-parser/README.htm Thu Jan 10 14:38:38 2008
@@ -0,0 +1,35 @@
+	<body>
+		<h1>XML based query syntax
+		</h1>
+		<p>
+			This module contains:
+			<ul>
+				<li>a modular Lucene Query Parser where queries are expressed as XML</li>
+				<li>JUnit test</li>
+				<li>DTD schemas and <a href="docs/index.html">generated documentation</a>
+				<li>Example XML queries</li>
+				<li>Test index (subset of Reuters 21578)</li>
+			</ul>
+		</p>
+		<p>
+			The original motivation for creating this package was outlined and discussed <a href="">here</a>.
+		</p>
+		<p>
+			Parser support includes:
+			<ul>
+				<li>"Span" queries</li>
+				<li>Simple caching for filters</li>
+				<li>"Like this" queries</li>
+				<li>Boolean, Term, and UserInput (parsed with existing query parser)</li>
+				<li>BoostingQuery - a class that can downgrade scores for hits on
+				certain terms rather than the hard-line approach taken by BooleanClause.Occurs.MUST_NOT</li>
+				<li>FilteredQuery, RangeFilter, DuplicateFilter and "TermsFilter" for non-sequential
+				<li>"FuzzyLikeThis" a new query which is a cross between "LikeThis" and "fuzzy"
but with
+					better scoring of fuzzy terms than standard fuzzy queries</li>
+				<li>A modular design with expandable support for new query/filter types</li>
+			</ul>
+		</p>
+		<p>This code is dependent on the "queries" contrib module although the "CoreParser"
can be compiled with just Lucene core if required</p>
+	</body>
\ No newline at end of file

View raw message