This is how I've got it: A document first passes through the UpdateChain (processors), which is document centric. Then copyFields are processed And finally the analysis in fieldTypes are processed
So you cannot use <copyField> before UpdateProcessors nor after Analysis :( -- Jan Høydahl, search solution architect Cominvent AS - www.cominvent.com On 23. feb. 2011, at 09.45, Markus Jelsma wrote: > You are right, i misread. Fields a copied first, then analyzed and then > behave > like other fields and pass the same way through the update processor. > > Cheers, > >> Markus, >> >> I searched but I couldn't find a definite answer, so I posted this >> question. >> The article you quoted talks about implementing a copyField-like operation >> using UpdateProcessor. It doesn't talk about relationship between >> the copyField operation proper and UpdateProcessors. >> >> Kuro >> >> On 2/22/11 3:00 PM, "Markus Jelsma" <markus.jel...@openindex.io> wrote: >>> Yes. But did you actually search the mailing list or Solr's wiki? I guess >>> not. >>> >>> Here it is: >>> http://wiki.apache.org/solr/UpdateRequestProcessor >>> >>>> Can fields created by copyField instructions be processed by >>>> UpdateProcessors? >>>> Or only raw input fields can? >>>> >>>> So far my experiment is suggesting the latter. >>>> >>>> ---- >>>> T. "Kuro" Kurosaka
