-------- Original-Nachricht -------- > Datum: Thu, 5 Nov 2009 22:45:54 +1300 > Von: "Yashasvi A.C." <yashasviac@xxxxxxxxx> > > On Thu, Nov 5, 2009 at 8:06 PM, Ingo Weinhold <ingo_weinhold@xxxxxx> > wrote: > > > > > On 2009-11-05 at 03:46:59 [+0100], Rene Gollent <anevilyak@xxxxxxxxx> > > wrote: > > > On Wed, Nov 4, 2009 at 8:39 PM, Rene Gollent <anevilyak@xxxxxxxxx> > > wrote: > > > > far more than the 4 entries in your folder. The other problem is > that > > > > last_modified is not relative, it's a timestamp, which is to say you > > > > probably want something like "(%now% - last_modified) < 10000000" as > > > > > > On second thought, you might want to rework that as "last_modified > > > > (%now% - 10000000)"..I'm not sure how intelligent BFS is at optimizing > > > queries, and in case it doesn't do any such thing this form should be > > > friendlier for the indices. > > > > I don't think that would work either. The expression could be > > "last_modified > %-10 secs%", though. > > The "last_modified > (%non% - 10000000)" predicate doesn't really restrict > the files that were modified in the last 10 seconds. So, even if I click > the > button once again after 15 seconds, the previous messages are still > listed. Particularly with the "%non%" in the query predicate is invalid. Under BeOS the "query" command line tool fails. Under Haiku it doesn't fail and instead returns a whole lot of files. Apparently a bug. > Also, as Rene mentioned, all the files are being checked. So, I put in a > simple filter on the file name (pretty much checking if the string starts > with Message and checking the character after that is a digit) to get the > files that I needed. Maybe I could try the predicate that Ingo suggested > and > check if it gives me the files within the last 10 seconds. I just tried it myself under Haiku and there seems to be an issue with the time zone not being considered correctly (in GMT+1 I have to specify "%+50 mins%" to get the files modified at most ten minutes ago). I'm using an older Haiku ATM, though, so maybe the bugs have been fixed already. I vaguely recall having seen BFS query code related commits recently. CU, Ingo