Hello Hossman, sorry for my late response.
For this specific case, you are right. It makes more sense to do such work "on the fly". However, I am only testing at the moment, what one can do with Solr and what not. Is the UpdateProcessor something that comes froms Lucene itself or from Solr? Thanks! hossman wrote: > > > : Is there a way to prepare a document the described way with Lucene/Solr, > : before I analyze it? > : My use case is to categorize several documents in an automatic way, > which > : includes that I have to "create" data from the given input doing some > : information retrieval. > > As Ryan mentioned earlier: this is what the UpdateRequestProcessor API > is for -- it allows you to modify Documents (regardless of how they were > added: csv, xml, dih) prior to Solr processing them... > > http://old.nabble.com/Custom-Analyzer-Tokenizer-works-but-results-were-not-saved-to27026739.html > > Personally, i think you may be looking at your problem from the wrong > dirrection... > > : >> Imagine you would analyze, index and store them like you normally do > and > : >> afterwards you want to set, whether the document belongs to the > expensive > : >> item-group or not. > : >> If the price for the item is higher than 500$, it belongs to the > : >> expensive > : >> ones, otherwise not. > > ...for a situation like that, i wouldn't attempt to "classify" the docs as > "expensive" or "cheap" when adding them. instead i would use numeric > ranges for faceting and filtering to show me how many docs where > "expensive" or "cheap" at query time -- that way when the ecomony tanks i > can redifine my definition of "expensive" on the fly w/o needing to > reindex a million documents. > > > > -Hoss > > > -- View this message in context: http://old.nabble.com/Custom-Analyzer-Tokenizer-works-but-results-were-not-saved-tp27026739p27109760.html Sent from the Solr - User mailing list archive at Nabble.com.
