On Tue, Feb 12, 2008 at 03:45:51PM +0100, Mario Lopez wrote: > the reversed index which takes like 20 minutes, I guess it has to do > with the plperl function, perhaps a C function for inverting would make > it up in less time.
sure. take a look at this: http://www.depesz.com/index.php/2007/09/05/indexable-field-like-something-update/ > The problem is still with the LIKE '%keyword%', my problem is that I am > not searching for Words in a dictionary fashion, suppose my "data" is > random garbage, that it has common consecutive bytes. How could I > generate a dictionary from this random garbage to make it easier for > indexing? if you are talking abount really general "random bytes", and really general "random substrings" - you're basically out of luck. but think about acceptable limitations. can you limit min-length of searched substring? check pg_trgm in contrib. as for generating dictionary - show me the data and perhaps i can think of something. depesz -- quicksil1er: "postgres is excellent, but like any DB it requires a highly paid DBA. here's my CV!" :) http://www.depesz.com/ - blog dla ciebie (i moje CV) ---------------------------(end of broadcast)--------------------------- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq