On Fri, 2004-08-20 at 11:01, Douglas Soares de Andrade wrote: > Hi ! > > > First of all, (nearly all) the data for the whole dictionary is > > transfered via the network socket to the IDE. Secondly, these values are > > added to the debug browser. This needs some CPU cycles. I just tested it > > on my machine with a dictionary of 8000 integers and I recognized a > > little delay but it was bearable. But this depends on the power of your > > machine. > > My machine is a Athlon XP, with 256 MB, running slackware 10 and kde 3.3. > > > Not without redesigning this part. If we introduce watchvariables, you > > could enter the variable names your interested in. Today, all data is > > accessible. I am unsure, which one is the better way. > > I would like to select the variable i want to see the values. I dont have to > see the 8000 itens dict, only the others variables...
How about wildcard filters? The default could be *; but you could remove it and add other filters. I agree this would be a great addition. I've had a python app with Numeric arrays with thousands of array entries. Eric would 'hang' when it hit a breakpoint with these variables in scope. I haven't had to debug this app recently, but the ability to remove these from the list would have been a great improvement. -- Michael Andrews <[EMAIL PROTECTED]> _______________________________________________ PyKDE mailing list [EMAIL PROTECTED] http://mats.imk.fraunhofer.de/mailman/listinfo/pykde