Hold down control while selecting text by clicking and dragging. I believe this is still supposed to do a column select, but it just makes mousepad get stuck endlessly using 100% of the CPU.
I don't think column select is supported (is that the same as block/rectangular selection?), but indeed something weird happens. It doesn't get stuck at all here, but it does peg my CPU(s) at around 30-35% (as show in "Task Manager") until clicking anywhere else and then it goes back to normal.
> It doesn't get stuck at all here, but it does peg my CPU(s) at around 30-35% 30-35% could essentially mean 100% of one CPU on a quad core machine... > I don't think column select is supported (is that the same as block/rectangular selection?) Yes, I think we're talking about the same thing. But I did some more testing and: - I guess it must be supported, because it generally does work if you ctrl-click at the current cursor location, as long as this is not at the end of the line. i.e. click to place the cursor where you want the start or end of the selection to be, then ctrl-click and drag from the exact same location. (You can also start at or off the end of the line if you click to put the cursor at the end of the line, and then click and drag to the left until you are within the text on that line, before dragging up or down). - yes, mostly it behaves basically like you describe - full CPU usage but not "stuck". It seems like the time that it takes to return to normal after I click somewhere else is very variable, so I guess maybe it was taking extra long yesterday because I was using very large files (with line wrapping enabled as well), and I was letting it run for a long time before clicking somewhere else.
Here using mousepad w/ gtk2, just a single ctrl + click causes it keep consuming 20~25% of CPU. However using mousepad w/ gtk3 this can't be reproduced. Maybe a bug in gtksourceview2?
The column selection code had to be disabled/compiled-out in GTK+ 3 because it doesn't expose some stuff that it did in GTK+ 2 which that code needs to actually work (I forget the specifics, something like the code accessed a sealed member relating to input methods).
I'm marking "WONTFIX" but I actually don't mind if we fix this bug, just first somebody has to make it work with the changes in GTK+ 3.