Hi, Jan Høydahl, Forgot to mention, the identifier I use is an existing one wrapped in Solr3.5.0., LangDetectLanguageIdentifier (http://wiki.apache.org/solr/LanguageDetection).
For the language identifier, I looked into the sc, and found that the whole content of a text is parsed before detection, which is why the end result consists of a specific language instead of multiple languages. Then I can assume, if the content is processed section by section (or even line by line), the end result shall consist of multiple languages. So the question is, can you guys plug this modification of the existing identifier into Solr? Best Regards, Bing -- View this message in context: http://lucene.472066.n3.nabble.com/Can-solr-langid-Solr3-5-0-detect-multiple-languages-in-one-text-tp3821210p3821764.html Sent from the Solr - User mailing list archive at Nabble.com.
