>> I need to aggregate data about the number of tags, tag combinations, >> keys and values. See http://osmdoc.com for the kind of data I need. > > Do something smart on line 346 for on the fly counting. Though I think > that I can generate the page that you provide maybe faster using just > group by on an existing database... MonetDB uses string deduplication, > perfect for this job ;)
I'm sorry but that does not even _remotely_ answer the question/problem I have. My problem is not the parsing of the XML file (or its performance) but the "on the fly counting" and the saving of the intermediate results until the parsing is finished. Perhaps my initial explanation was not the best. I'll try again later. No offense but I'd like to see you try generating that on-the-fly even using a column-oriented database. Still, any input is greatly appreciated! Lars _______________________________________________ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
