You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-user@lucene.apache.org by Markus Jelsma <ma...@openindex.io> on 2017/10/03 13:17:48 UTC
RE: Moving to Point, trouble with IntPoint.newRangeQuery()
Ok, i have stripped down the QParser to demonstrate the problem. This is the basic test with only one document in the index:
public void testPointRange() throws Exception {
assertU(adoc("id", "8", "digest1", "-1820898630"));
assertU(commit());
assertQ(
req("q", "{!qdigest field=digest1}", "debug", "true", "indent", "true"),
"//result/doc[1]/str[@name='id'][.='8']");
}
The following parse() implementation passes (because it simply uses LuceneQParser syntax):
@Override
public Query parse() throws SyntaxError {
QParser luceneQParser = new LuceneQParser("digest1:[-1820898630 TO -1820898630]", localParams, params, req);
return luceneQParser.parse();
}
But when i switch to a BooleanQuery with just one RangeQuery, it fails:
@Override
public Query parse() throws SyntaxError {
BooleanQuery.Builder builder = new BooleanQuery.Builder();
Query pointQuery = IntPoint.newRangeQuery("digest1", -1820898630, -1820898630);
builder.add(pointQuery, Occur.SHOULD);
return builder.build();
}
I might be overlooking things but i really don't so the problem with the second parse() impl.
What am i doing wrong?
Many thanks,
Markus
-----Original message-----
> From:Chris Hostetter <ho...@fucit.org>
> Sent: Tuesday 26th September 2017 18:52
> To: Solr-user <so...@lucene.apache.org>
> Subject: Re: Moving to Point, trouble with IntPoint.newRangeQuery()
>
>
> : I have a QParser impl. that transforms text input to one or more
> : integers, it makes a BooleanQuery one a field with all integers in
> : OR-more. It used to work by transforming the integer using
> : LegacyNumericUtils.intToPrefixCoded, getting a BytesRef.
> :
> : I have now moved it to use IntPoint.newRangeQuery(field, integer,
> : integer), i read (think javadocs) this is the way to go, but i get no
> : matches!
>
> As a general point: if you want to do this in a modular way, you should
> fetch the FieldType from the IndexSchema and use the
> FieldType.getRangeQuery(...) method.
>
> That said -- at a quick glance, knowing how your schema is defined, i'm
> not sure why/how your IntPoint.newRangeQuery() code would fail.
>
> Maybe add a lower level test of the QParser directly and assert some
> explicit properties on the Query objects? (class, etc...)
>
>
>
> -Hoss
> http://www.lucidworks.com/
>