blob: 330d01a65f738a71eebe0c373dbefdd90a8d67ac [file] [log] [blame]
using Lucene.Net.QueryParsers.Flexible.Core;
using Lucene.Net.QueryParsers.Flexible.Core.Nodes;
using Lucene.Net.QueryParsers.Flexible.Core.Parser;
using Lucene.Net.QueryParsers.Flexible.Core.Processors;
using Lucene.Net.QueryParsers.Flexible.Standard.Parser;
using Lucene.Net.QueryParsers.Flexible.Standard.Processors;
using Lucene.Net.Search.Spans;
using Lucene.Net.Util;
using NUnit.Framework;
using System;
namespace Lucene.Net.QueryParsers.Flexible.Spans
{
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/// <summary>
/// This test case demonstrates how the new query parser can be used.
/// <para/>
/// It tests queries likes "term", "field:term" "term1 term2" "term1 OR term2",
/// which are all already supported by the current syntax parser (
/// <see cref="StandardSyntaxParser"/>).
/// <para/>
/// The goals is to create a new query parser that supports only the pair
/// "field:term" or a list of pairs separated or not by an OR operator, and from
/// this query generate <see cref="SpanQuery"/> objects instead of the regular
/// <see cref="Search.Query"/> objects. Basically, every pair will be converted to a
/// <see cref="SpanTermQuery"/> object and if there are more than one pair they will be
/// grouped by an <see cref="OrQueryNode"/>.
/// <para/>
/// Another functionality that will be added is the ability to convert every
/// field defined in the query to an unique specific field.
/// <para/>
/// The query generation is divided in three different steps: parsing (syntax),
/// processing (semantic) and building.
/// <para/>
/// The parsing phase, as already mentioned will be performed by the current
/// query parser: <see cref="StandardSyntaxParser"/>.
/// <para/>
/// The processing phase will be performed by a processor pipeline which is
/// compound by 2 processors: <see cref="SpansValidatorQueryNodeProcessor"/> and
/// <see cref="UniqueFieldQueryNodeProcessor"/>.
/// <para/>
///
/// <see cref="SpansValidatorQueryNodeProcessor"/>: as it's going to use the current
/// query parser to parse the syntax, it will support more features than we want,
/// this processor basically validates the query node tree generated by the parser
/// and just let got through the elements we want, all the other elements as
/// wildcards, range queries, etc...if found, an exception is thrown.
///
/// <see cref="UniqueFieldQueryNodeProcessor"/>: this processor will take care of reading
/// what is the &quot;unique field&quot; from the configuration and convert every field defined
/// in every pair to this &quot;unique field&quot;. For that, a <see cref="SpansQueryConfigHandler"/> is
/// used, which has the <see cref="IUniqueFieldAttribute"/> defined in it.
///
/// <para/>
/// The building phase is performed by the <see cref="SpansQueryTreeBuilder"/>, which
/// basically contains a map that defines which builder will be used to generate
/// <see cref="SpanQuery"/> objects from <see cref="IQueryNode"/> objects.
/// </summary>
/// <seealso cref="SpansQueryConfigHandler"/>
/// <seealso cref="SpansQueryTreeBuilder"/>
/// <seealso cref="SpansValidatorQueryNodeProcessor"/>
/// <seealso cref="SpanOrQueryNodeBuilder"/>
/// <seealso cref="SpanTermQueryNodeBuilder"/>
/// <seealso cref="StandardSyntaxParser"/>
/// <seealso cref="UniqueFieldQueryNodeProcessor"/>
/// <seealso cref="IUniqueFieldAttribute"/>
public class TestSpanQueryParser : LuceneTestCase
{
private QueryNodeProcessorPipeline spanProcessorPipeline;
private SpansQueryConfigHandler spanQueryConfigHandler;
private SpansQueryTreeBuilder spansQueryTreeBuilder;
private ISyntaxParser queryParser = new StandardSyntaxParser();
public TestSpanQueryParser()
{
// empty constructor
}
public override void SetUp()
{
base.SetUp();
this.spanProcessorPipeline = new QueryNodeProcessorPipeline();
this.spanQueryConfigHandler = new SpansQueryConfigHandler();
this.spansQueryTreeBuilder = new SpansQueryTreeBuilder();
// set up the processor pipeline
this.spanProcessorPipeline
.SetQueryConfigHandler(this.spanQueryConfigHandler);
this.spanProcessorPipeline.Add(new WildcardQueryNodeProcessor());
this.spanProcessorPipeline.Add(new SpansValidatorQueryNodeProcessor());
this.spanProcessorPipeline.Add(new UniqueFieldQueryNodeProcessor());
}
public SpanQuery GetSpanQuery(/*CharSequence*/string query)
{
return GetSpanQuery("", query);
}
public SpanQuery GetSpanQuery(String uniqueField, /*CharSequence*/string query)
{
this.spanQueryConfigHandler.Set(SpansQueryConfigHandler.UNIQUE_FIELD, uniqueField);
IQueryNode queryTree = this.queryParser.Parse(query, "defaultField");
queryTree = this.spanProcessorPipeline.Process(queryTree);
return (SpanQuery)this.spansQueryTreeBuilder.Build(queryTree); // LUCENENET TODO: Find way to remove cast
}
[Test]
public void TestTermSpans()
{
assertEquals(GetSpanQuery("field:term").toString(), "term");
assertEquals(GetSpanQuery("term").toString(), "term");
assertTrue(GetSpanQuery("field:term") is SpanTermQuery);
assertTrue(GetSpanQuery("term") is SpanTermQuery);
}
[Test]
public void TestUniqueField()
{
assertEquals(GetSpanQuery("field", "term").toString(), "field:term");
assertEquals(GetSpanQuery("field", "field:term").toString(), "field:term");
assertEquals(GetSpanQuery("field", "anotherField:term").toString(),
"field:term");
}
[Test]
public void TestOrSpans()
{
assertEquals(GetSpanQuery("term1 term2").toString(),
"spanOr([term1, term2])");
assertEquals(GetSpanQuery("term1 OR term2").toString(),
"spanOr([term1, term2])");
assertTrue(GetSpanQuery("term1 term2") is SpanOrQuery);
assertTrue(GetSpanQuery("term1 term2") is SpanOrQuery);
}
[Test]
public void TestQueryValidator()
{
try
{
GetSpanQuery("term*");
fail("QueryNodeException was expected, wildcard queries should not be supported");
}
#pragma warning disable 168
catch (QueryNodeException ex)
#pragma warning restore 168
{
// expected exception
}
try
{
GetSpanQuery("[a TO z]");
fail("QueryNodeException was expected, range queries should not be supported");
}
#pragma warning disable 168
catch (QueryNodeException ex)
#pragma warning restore 168
{
// expected exception
}
try
{
GetSpanQuery("a~0.5");
fail("QueryNodeException was expected, boost queries should not be supported");
}
#pragma warning disable 168
catch (QueryNodeException ex)
#pragma warning restore 168
{
// expected exception
}
try
{
GetSpanQuery("a^0.5");
fail("QueryNodeException was expected, fuzzy queries should not be supported");
}
#pragma warning disable 168
catch (QueryNodeException ex)
#pragma warning restore 168
{
// expected exception
}
try
{
GetSpanQuery("\"a b\"");
fail("QueryNodeException was expected, quoted queries should not be supported");
}
#pragma warning disable 168
catch (QueryNodeException ex)
#pragma warning restore 168
{
// expected exception
}
try
{
GetSpanQuery("(a b)");
fail("QueryNodeException was expected, parenthesized queries should not be supported");
}
#pragma warning disable 168
catch (QueryNodeException ex)
#pragma warning restore 168
{
// expected exception
}
try
{
GetSpanQuery("a AND b");
fail("QueryNodeException was expected, and queries should not be supported");
}
#pragma warning disable 168
catch (QueryNodeException ex)
#pragma warning restore 168
{
// expected exception
}
}
}
}