Greg, You could use copyField to copy the column in question to 6 fields, one for each of your 6 languages, and hope they none of the analyzers do something reasonable without crashing. Or apply the white-space tokenizer and hope for the best?
If the column has long enough text, you could try a language detector. My company, Basis Technology, sells one, and it can plug into Solr easily. http://www.basistech.com/language-identification/ On 2/22/11 11:50 AM, "Greg Georges" <greg.geor...@biztree.com> wrote: >Hello all, > >I have just gone through the mailing list and have set up my different >field type analysers for my 6 different languages in my shema.xml. Here >is my question. I am using the dataimporthandler to import data from my >database into my index. In my table, the documentname column's data can >be in any of the 6 languages. Lets say I want to index this data and >apply the different language analysers for certain cases, what would be >the best way in my case. The real problem is that I do not know the >language of the string in the documentname column once I create my index, >therefore I cannot apply the correct field type. Should I create a custom >transformer? > >Thanks > >Greg ---- T. "Kuro" Kurosaka, 415-227-9600x122, 617-386-7122(direct)
