I recognize that error message ;) You're using AnalyzingQueryParser
<http://lucene.apache.org/java/docs/api/org/apache/lucene/queryParser/analyzing/AnalyzingQueryParser.html> -
yes?
These are imo the two most obvious options:
1. Revert to standard QueryParser - it won't analyze prefix- and
wildcardqueries.
2. Keep AnalyzingQueryParser together with different analyzers for
different fields (wrapped in
org.apache.lucene.analysis.PerFieldAnalyzerWrapper).
With option #2 you could use Standard- or GermanAnalyzer for text
fields, and for filenames etc. a better choice would be something like
KeywordAnalyzer, which doesn't add or consume tokens. Remember to
always use the same analyzer(s) for indexing AND search though.
/Ronnie
Quoting Stefan Schütz <[EMAIL PROTECTED]>:
Hi,
first let me explain the situation:
We have to index an document, which contains a field "file" to store
filenames.
Sometimes filenames contain an underscore or an minus (_ or -). => e.g.
foo_bar.doc
Indexing is'nt the problem so far.
But if we now try to search for "foo_b*" the QueryParser crashes with
following exception:
org.apache.lucene.queryParser.ParseException: Cannot build PrefixQuery
with analyzer class org.apache.lucene.analysis.de.GermanAnalyzer -
token was consumed
Same problem occurs if i use the StandardAnalyzer.
To replace underscores in filenames is not a solution because we have
many other fields
which can contain underscores. :(
Is there anybody who knows this problem or knows a solution for this?
thx in advance.
Stefan
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
