[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: Kmail slow
From:       Roberto Alsina <ralsina () conectiva ! com ! ar>
Date:       2000-06-08 14:07:58
[Download RAW message or body]

On Thu, 8 Jun 2000, Rik Hemsley wrote:

> #if Roberto Alsina
> > So, if you want to get the messages that are referenced in a certain
> > folder called "myfolder", which are not read, and whose subject contains
> > the word "KDE" (to give a ridiculous example) it could be this query over
> > the articles table:
> > 
> > exists article: (folders[i].name = 'myfolder') and article.read=true and
> > article.subject like '%KDE%'
> 
> I'm not doubting that the db method for storing an index is fast, but
> what is it that makes it faster than an 'index' ? Empath uses a QDict to
> store index records. This is mirrored on disk and flushed from memory
> when it hasn't been used for a while. It stores data in Qt formats
> (QString, QDateTime) so that there is no conversion overhead - the
> values can be inserted directly into a QListView. I don't know much
> about dbs, so I'm just not clear on what sort of speed advantage it
> would give over a QDict.

What is in your QDict? How many records can you reasonably put in it?

 ("\''/").__..-''"`-. .         Roberto Alsina
 `9_ 9  )   `-. (    ).`-._.`)  ralsina@unl.edu.ar
 (_Y_.)' ._   ) `._`.  " -.-'   Director técnico
  _..`-'_..-_/ /-'_.'           Conectiva SA
(l)-'' ((i).' ((!.'             Buenos Aires - Argentina
                                KDE Developer (MFCH)
"The stone age didn't end for a lack of stone" Firoz Rasul

 
>> Visit http://master.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<

[prev in list] [next in list] [prev in thread] [next in thread]