Then you're probably going to write your own Collector if you need to see each document and do something different with it. Do be aware that you _really_ need to be sure you get your values from docValues fields. If you use the simple get(docId).getField() method the stored fields will be read from disk and decompressed in order to be returned which will be a performance killer. docValues fields _may_ be automatic if you use useDocValuesAsStored=true for the fields you want to examine.
WARNING: I haven't been into the code for useDocValuesAsStored so I'm not at all clear whether it's really automatic at the collector level. Best, Erick On Tue, Mar 14, 2017 at 1:15 AM, alexpusch <a...@getjaco.com> wrote: > Single field. I'm iterating over the results once, and need each doc in > memory only for that single iteration. I need different fields from each doc > according to the algorithm state. > > > > -- > View this message in context: > http://lucene.472066.n3.nabble.com/Iterating-sorted-result-docs-in-a-custom-search-component-tp4324497p4324818.html > Sent from the Solr - User mailing list archive at Nabble.com.
