David Pilato 6ae6a078de Search: add format support for date range filter and queries
When the date format is defined in mapping, you can not use another format when querying using range date query or filter.

For example, this won't work:

```
DELETE /test

PUT /test/t/1
{
  "date": "2014-01-01"
}

GET /test/_search
{
  "query": {
    "filtered": {
      "filter": {
        "range": {
          "date": {
            "from": "01/01/2014"
          }
        }
      }
    }
  }
}
```

It causes:

```
Caused by: org.elasticsearch.ElasticsearchParseException: failed to parse date field [01/01/2014], tried both date format [dateOptionalTime], and timestamp number
```

It could be nice if we can support at query time another date format just like we support `analyzer` at search time on String fields.

Something like:

```
GET /test/_search
{
  "query": {
    "filtered": {
      "filter": {
        "range": {
          "date": {
            "from": "01/01/2014",
            "format": "dd/MM/yyyy"
          }
        }
      }
    }
  }
}
```

Same for queries:

```
GET /test/_search
{
  "query": {
    "range": {
      "date": {
        "from": "01/01/2014",
        "format": "dd/MM/yyyy"
      }
    }
  }
}
```

Closes #7189.
2014-10-06 15:59:18 +02:00

153 lines
6.8 KiB
Java

/*
* Licensed to Elasticsearch under one or more contributor
* license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright
* ownership. Elasticsearch 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.
*/
package org.elasticsearch.index.query;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.TermRangeQuery;
import org.elasticsearch.common.inject.Inject;
import org.elasticsearch.common.joda.DateMathParser;
import org.elasticsearch.common.joda.Joda;
import org.elasticsearch.common.lucene.BytesRefs;
import org.elasticsearch.common.xcontent.XContentParser;
import org.elasticsearch.index.mapper.FieldMapper;
import org.elasticsearch.index.mapper.MapperService;
import org.elasticsearch.index.mapper.core.DateFieldMapper;
import org.joda.time.DateTimeZone;
import java.io.IOException;
import static org.elasticsearch.index.query.support.QueryParsers.wrapSmartNameQuery;
/**
*
*/
public class RangeQueryParser implements QueryParser {
public static final String NAME = "range";
@Inject
public RangeQueryParser() {
}
@Override
public String[] names() {
return new String[]{NAME};
}
@Override
public Query parse(QueryParseContext parseContext) throws IOException, QueryParsingException {
XContentParser parser = parseContext.parser();
XContentParser.Token token = parser.nextToken();
if (token != XContentParser.Token.FIELD_NAME) {
throw new QueryParsingException(parseContext.index(), "[range] query malformed, no field to indicate field name");
}
String fieldName = parser.currentName();
token = parser.nextToken();
if (token != XContentParser.Token.START_OBJECT) {
throw new QueryParsingException(parseContext.index(), "[range] query malformed, after field missing start object");
}
Object from = null;
Object to = null;
boolean includeLower = true;
boolean includeUpper = true;
DateTimeZone timeZone = null;
DateMathParser forcedDateParser = null;
float boost = 1.0f;
String queryName = null;
String currentFieldName = null;
while ((token = parser.nextToken()) != XContentParser.Token.END_OBJECT) {
if (token == XContentParser.Token.FIELD_NAME) {
currentFieldName = parser.currentName();
} else {
if ("from".equals(currentFieldName)) {
from = parser.objectBytes();
} else if ("to".equals(currentFieldName)) {
to = parser.objectBytes();
} else if ("include_lower".equals(currentFieldName) || "includeLower".equals(currentFieldName)) {
includeLower = parser.booleanValue();
} else if ("include_upper".equals(currentFieldName) || "includeUpper".equals(currentFieldName)) {
includeUpper = parser.booleanValue();
} else if ("boost".equals(currentFieldName)) {
boost = parser.floatValue();
} else if ("gt".equals(currentFieldName)) {
from = parser.objectBytes();
includeLower = false;
} else if ("gte".equals(currentFieldName) || "ge".equals(currentFieldName)) {
from = parser.objectBytes();
includeLower = true;
} else if ("lt".equals(currentFieldName)) {
to = parser.objectBytes();
includeUpper = false;
} else if ("lte".equals(currentFieldName) || "le".equals(currentFieldName)) {
to = parser.objectBytes();
includeUpper = true;
} else if ("time_zone".equals(currentFieldName) || "timeZone".equals(currentFieldName)) {
timeZone = DateMathParser.parseZone(parser.text());
} else if ("_name".equals(currentFieldName)) {
queryName = parser.text();
} else if ("format".equals(currentFieldName)) {
forcedDateParser = new DateMathParser(Joda.forPattern(parser.text()), DateFieldMapper.Defaults.TIME_UNIT);
} else {
throw new QueryParsingException(parseContext.index(), "[range] query does not support [" + currentFieldName + "]");
}
}
}
// move to the next end object, to close the field name
token = parser.nextToken();
if (token != XContentParser.Token.END_OBJECT) {
throw new QueryParsingException(parseContext.index(), "[range] query malformed, does not end with an object");
}
Query query = null;
MapperService.SmartNameFieldMappers smartNameFieldMappers = parseContext.smartFieldMappers(fieldName);
if (smartNameFieldMappers != null) {
if (smartNameFieldMappers.hasMapper()) {
FieldMapper mapper = smartNameFieldMappers.mapper();
if (mapper instanceof DateFieldMapper) {
if ((from instanceof Number || to instanceof Number) && timeZone != null) {
throw new QueryParsingException(parseContext.index(), "[range] time_zone when using ms since epoch format as it's UTC based can not be applied to [" + fieldName + "]");
}
query = ((DateFieldMapper) mapper).rangeQuery(from, to, includeLower, includeUpper, timeZone, forcedDateParser, parseContext);
} else {
if (timeZone != null) {
throw new QueryParsingException(parseContext.index(), "[range] time_zone can not be applied to non date field [" + fieldName + "]");
}
//LUCENE 4 UPGRADE Mapper#rangeQuery should use bytesref as well?
query = mapper.rangeQuery(from, to, includeLower, includeUpper, parseContext);
}
}
}
if (query == null) {
query = new TermRangeQuery(fieldName, BytesRefs.toBytesRef(from), BytesRefs.toBytesRef(to), includeLower, includeUpper);
}
query.setBoost(boost);
query = wrapSmartNameQuery(query, smartNameFieldMappers, parseContext);
if (queryName != null) {
parseContext.addNamedQuery(queryName, query);
}
return query;
}
}