#mysql - Sat 3 Mar 2007 between 23:36 and 23:54

NY Lost Funds



^MichaelAh, that could be a problem i guess
victori_does not give you accurate results? thats nice
WillWell, I guess it's more "approximation"
victori_I want my db to give me whatever numbers it chooses /sarcasm
WillSarcasm is not allowed in #mysql.
^Michael:(
c0m--anyone have any experience using transactions ?
victori_I can see the benefit of switching to mysql for a small little project that relies on an intensive database
myisam table for the >500k row table with text match
sounds like a decent fit
WillI wouldn't use MySQL's Fulltext indexing
victori_why?
WillUse a different engine, like Lucene. It's better/ faster
victori_currently I am using postgres and tsearch2 , however postgresql has slow aggregates specifically count, which I am considering of hopping to mysql for this specific task.
will: really its that bad?
I am currently using tsearch2 on this tableset http://ircmine.com/statistics
fairly fast and index is updated realtime via trigger
http://jayant7k.blogspot.com/2006/06/benchmarking-results-of-mysql-lucene.html
ya I see what your talking about
that is bad
WillI've used it in production. It's not bad. For me, it's easy to use and setup, and works for my needs.
Why do people call it postgres anyways... unless they are really using postgres...
victori_well my database grows fairly quickly and for it to work efficiently, the whole thing needs to be in ram, thats not realistic.
sphinx looks like a decent option , maybe lucene if I want to deal with the java platform'
WillJust use grep
victori_you can do that?
WillYeah, export to CSV
Gatoelhowhat's the difference between tinytext and varchar[255] ?

Page: 4 11 18 25 32 

IrcArchive

NY Lost Funds