Angus Leeming wrote:
Abdelrazak Younes wrote:
Angus Leeming wrote:
I haven't looked at the patch but I do have a question: why LyXView?
Good question. There are two reasons why setBuffer and LoadLyXFile are
transfered to LyXView instead of WorkArea:
...
Does it make sense?
Having looked at the code now, it makes perfect sense. The change is entirely
reasonable. IMO you can shove this in.
Once I've checked that qt3 is compiling and working, I'll do that. Well,
unless I hear some objection of course... or if Lars wants to merge the
branch bits by bits.
Here's a scenario for you to think about though. LyX acts on LFUNs. These
behaviours can be triggered by a user pressing a menu button which (as you note)
automatically associates the LFUN with a LyXView. However, they can also be
triggered by an LFUN being received by the LyXServer.
What happens in the (hypothetical for now) case when we have two LyXViews
neither of which have focus and the LyXServer receives the setBuffer LFUN? Which
LyXView do we choose is "active"?
We will need to adjust the protocol. The client will be asked to provide
an ID for the LyXView. I guess the LyXServer will have to give this ID
to the client on connection.
Just something to think about...
Yes... Are there many of these client programs?
Abdel.
