Aditya, Char filters are applied prior to tokenization, so they can affect tokenization, but I can't think of any tokenization changes that accent stripping would cause.
Token filters can be re-ordered to achieve certain objectives. For example, if you want to use a stemmer that only recognizes lowercase terms, you could put a lowercasing filter in front of it. In your case, if you use a char filter to do the accent stripping, and you use a stemmer, you won't be able to order it after stemming, because char filters always precede tokenization, which always precedes token filtering. Stripping accents before stemming could be a problem, though, if your stemmer assumes properly accented words in order to function properly; in that case, you'd want to use a token filter to do the accent stripping instead, and place it after your stemmer. There may be other reasons you'd want to choose one over the other that I'm not thinking of, but primarily it's about choosing processing order to affect further stages in the pipeline. If you don't think order matters, then you should be fine choosing either one. Steve On Jun 13, 2013, at 8:17 PM, adityab <aditya_ba...@yahoo.com> wrote: > this might be a dumb question. But can you please point me some key > difference between ASCIIFolding Filter and Character Filter using a map > File. > thanks > Aditya > > > > -- > View this message in context: > http://lucene.472066.n3.nabble.com/Best-way-to-match-umlauts-tp4070256p4070398.html > Sent from the Solr - User mailing list archive at Nabble.com.
