Surely it's best to have a specific analyzer for each language? Would support for multiple Analyzers with a single index require a different IndexWriter for each Analzser/language? Would you then need to manage the disk access of these regarding locking etc, so two IndexWriter's can not do so at the same time?
-----Original Message----- From: Olivier Jaquemet [mailto:[EMAIL PROTECTED] Sent: 06 September 2005 13:21 To: java-user@lucene.apache.org Subject: Re: Multiple Language Indexing and Searching Gusenbauer Stefan wrote: >I think nutch uses ngramj for language classification but i don't know >what type of saving language information they use. In our application >for example i save the language in an extra field in the document >because lucene is supporting multiple fields with the same names we >would be able to handle different languages. but for now we don't need it > > But then, if you do so, you do not benefit from any specialized Analyzer you could use for each language, do you? Then again, maybe it's not that interesting to use specialized analyzers for each language?. --------------------------------------------------------------------- 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]
