|
From: Ralf J. <ral...@gm...> - 2010-12-12 21:32:56
|
Hi again, I now spent about two hours attempting to work around this silly bug. It turned out that just chunking the data was not enough, as removing the old data (which is done while processing the first chunk) is horribly slow: eXist barely manages to remove a collection with 9k documents before this bug (or whatever it is) kicks in. But I think I got enough hacks together now. Getting data into eXist is as much of a pain as it is a joy to work on the data once it is in :( Also, I am under the impression that adding/removing to a collection scales quite badly with the number of documents in that collection (where I assumed that removing the whole collection was approximately linear with their size, and adding almost constant). Is that right? Is this worth a bug report and/or further investigation? Kind regards, Ralf Jung Am Sonntag 12 Dezember 2010, um 12:56:06 schrieb Ralf Jung: > Hi, > > > Split before & use ant task should be good. Ant have a way to reindex > > (look at "[Exist-open] Index corruption -- la lutte continue" thread ). > > This is not about a search index, but about a sort index created with > http://demo.exist-db.org/exist/functions/sort that is crated as part of the > XQuery doing the import. > Also, I never used Ant at all, so it would be some completely new framework > to learn... I think I'll try to do the splitting as part of the > pre-processing. Or live with the error messages. > > Kind regards, > Ralf Jung > > --------------------------------------------------------------------------- > --- Oracle to DB2 Conversion Guide: Learn learn about native support for > PL/SQL, new data types, scalar functions, improved concurrency, built-in > packages, OCI, SQL*Plus, data movement tools, best practices and more. > http://p.sf.net/sfu/oracle-sfdev2dev > _______________________________________________ > Exist-open mailing list > Exi...@li... > https://lists.sourceforge.net/lists/listinfo/exist-open |