>>>>> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes:
Juergen> Jean-Marc Lasgouttes wrote: It was a patch against 1.5. Try Juergen> the attached patch for 1.4 >> I probably missed part of the thread. Why do you have to revert >> the logic of the update flag? Juergen> I do not reverse it, I just renamed it. The problem was that Juergen> at the end of LyxFunc::dispatch, the screen was always Juergen> updated in the following way: Juergen> if (update) view()->update(Update::FitCursor | Juergen> Update::Force); else view()->update(Update::FitCursor); Juergen> In the case of saving, however, no update at all is required, Juergen> since that jumps back to the cursor. So I have introduced Juergen> another bool (defaulting to true), which can be set to false Juergen> so that none of these updates is done. I called that "update" Juergen> and renamed the old "update" to "updateforce" since that Juergen> strikes me most appropriate. We already have a way to avoid updates, it is the LyXAction::NoUpdate flag that can be added to a lfun. This is not taken in account at LyXFunc::dispatch level, though. Your patch only fixes a few cases. Try for example to open Tools>Preferences to see what I mean. I think what we should do instead of avoiding this update in some cases is to apply FitCursor only when needed. Martin, are you reading us? JMarc
