On Thu, 2007-08-09 at 22:25 +0200, Jonatan Åkerlind wrote: > On tor, 2007-08-09 at 19:48 +0100, Peter Clifton wrote: > > If you want it, I can code it at some point. I already broke that > > function once recently, so am familiar with where the code is. I also > > wondered what the desired behaviour should be. > > > > What do other combinations do... > > > > >From bottom left -> top right? > > >From top right -> bottom left? > > > > Is it just the left->right'edness which determines the action? > > > > Anyone _against_ this change? > > > > Regards, > > > > Peter Clifton > > Also coming from an AutoCAD environment this is something that I like. > IIRC it was actually top left -> bottom right that selected all > completely within the box and bottom right -> top left that selected all > elements that had at least one point within the box. > > Anyway I would really like this behaviour.
If you're feeling keen, you could try the patch here: http://repo.or.cz/w/geda-gaf/pcjc2.git (Specifically: http://repo.or.cz/w/geda-gaf/pcjc2.git?a=commitdiff;h=9390040771f782987d9dd5278a81459d22d5d2ec;hp=19b6b06461a906399126803d4c46a5a42d1cc784 ) Thanks to a google search for the rectangle intersection logic: http://www.tekpool.com/?p=23 Then "simplified" / obfuscated a bit using DeMorgan's theorem by me. Trying to think the resulting expression through hurts my brain. I've added an exception to the logic, that is - the right-left select WILL NOT select locked objects, otherwise your title-block will get selected with every right-left select. I await feedback from the other developers before committing this change to the git master repository, as changing a core user interface behaviour like this could be controversial. Regards, Peter Clifton _______________________________________________ geda-user mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
