Lars Gullik Bjønnes wrote:
Abdelrazak Younes <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
| > Abdelrazak Younes <[EMAIL PROTECTED]> writes:
| > | Hello,
| > | | As the title says...
| > | | If I don't get any objection, I'll do the merge tomorrow.
| > Your branch does not look to be up to date wrt trunk.
| > (Or perhaps it is your working copy)
|
| The patch dates from yesterday before some changes from Michael G.
| were committed. I'll of course update the patch before merging.
|
| > You also seems to have some mis-merge from trunk to branch.
|
| If they are not related to the above please tell me what they are.
Changes to lyx.man, some changes of variables names where trunk and
branch differ. timeout_ vs. timeout something.
I'll check that thanks.
Also the change to BufferView::Pimpl::update will mean that
updatemetrics is run a lot less than it used to. (Because of the
redraw_needed check)
I have checked by running in debug mode step by step that the behavior
is correct. But we can still change that afterward.
| > And I am not sure that your change to a synchronous calling
| > frontend<->kernel is good. However it can be put back later.
|
| In this case this is a change to frontend<->frontend because the code
| I moved really belonged to the frontend and not to the kernel.
Not really...
??? I am talking about the view_state_changed signal which belonged to
LyXView and was emitted by BufferView on request of LyXView. So for me
it is _really_ avoiding an unneeded frontend<->frontend communication step.
but I agree this the specifig in this patch is not bad,
Be a bit positive man, a change is good or is bad. Saying "is not bad"
won't help me know your real opinion.
it is perhaps more the redraw_needed thing I am wary about.
As I said in a previous mail, I am sure this is _the_ correct solution
but I need it now to advance on my cleanup. When/if we can find a way to
avoid this check (by returning a bool in BufferView::update) we will
remove that. The cleanup is not done yet.
Abdel.
