rgheck wrote:
Thoroughly? Hard to know. Perhaps we should wait a bit and see if we get
any problems in trunk.
It should be fine in trunk but I am not sure it is 100% safe in branch.
There are a number of calls to start/stopBlinkingCursor(), some of those
are important because we don't want to do something else while
processing a mouse or keyboard event for example. I reckon those things
are cleaner in trunk so there should be no problem but, as I said I am
not sure about branch. The safest solution would be the stop the general
timer you introduced when entering stopBlinkingCursor() and restarting
it when leaving startBlinkingCursor(). Maybe we should do that in trunk
too but I'll let you decide.
I'll patch it into my local 1.5.4svn tree, which
is what I used for actual work, and see if I run into anything bad.
Others should do so, too, please. I guess the main thing for someone to
test here is that forked calls still work right. I don't use any
graphics and such so I'm not sure how to test it.
This is a bit different in branch. The attached patch is what we need
here. I don't know if people will think this is OK: I'm checking
QApplication::cursorFlashTime() in frontends::WorkArea. In principle, I
guess that's a no-no, since we're not in frontends/qt4, but it doesn't
really matter, since there isn't any other frontend and, in trunk,
everything is changed, anyway.
Yep, does not matter IMO.
Abdel.
