On Thu, 2 May 2013 16:58:01 +0200 "Jan Slodicka" <j...@resco.net> wrote:
> Hi Dan > > > What are your settings for pragmas "cache_size", "journal_mode" and > > "synchronous"? > > cache_size/synchronous - default values > > Don't remember, which journal_mode was used for testing. Should be > WAL, but I might have been lazy to write needed code. The source code > was meanwhile modified, but I can write it again if necessary. Maybe the delay at the end of the operation was an automatic checkpoint. I think that would explain why sqlite3_interrupt() and the others did not help. Maybe you can run the checkpoint in a background thread after rebuilding the FTS index or something. If you don't mind locking the database, you could also try the rebuild in rollback mode. It might be a little slower overall, but it might also be more responsive as far as sqlite3_interrupt() goes. _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
