You should take a look at org.apache.solr.analysis.MappingCharFilter,
which provides a generic table-based approach for use with solr. There
are also a lot of other interesting CharFilters in the same package.
For lucene-only use, there's
org.apache.lucene.analysis.icu.ICUFoldingFilter, which looks very
promising, and I'm sure there are others I don't know about.

On 7/3/2011 8:38 AM, Jan Rothhaar wrote:
Hi everybody,

I have a pretty generic question about token filters, and I am not really sure whether it is a developer or a configuration question:

How exactly do I make lucene map letters to each other, e.g. make it treat both 'a' and 'á' as one and the same letter, or both '写' and '寫' one and the same character? I am sure this question has appeared before and there are sample implementations or sample configuration files out there, but I could not find them on my own.

I only need to map single letters (i.e. no 'oe'<=> 'ö'), but in a multi-byte charset. I have some modest experience in programming in java, but am far from being a guru.

Any help is appreciated.

Thanks in advance,


To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org

Search Discussions

Discussion Posts


Related Discussions

Discussion Navigation
viewthread | post
posts ‹ prev | 2 of 2 | next ›
Discussion Overview
groupjava-user @
postedJul 3, '11 at 12:39p
activeJul 3, '11 at 4:06p

2 users in discussion

Jan Rothhaar: 1 post Michael Sokolov: 1 post



site design / logo © 2022 Grokbase