Jean-Marc Lasgouttes wrote: >> JMarc, these two seems to have somthing in comon with you ;) : 6768 (at >> point machinery), 6930 (undo broken) > > Concerning 6930, the introduction of AtPoint lead to removing a recordUndo > call for INSET_MODIFY. Since the code is common to all AtPoint entries (and > toggle had no recordUndo entry), this has to be reintroduced somehow. > > The first (easy) try was to call recordUndo whenever the lfun is not marked > readonly. However I do not like lfun relying on such behaviour from the > dispatcher. > > Therefore I decided to add the recordUndo in every place they were needed > (aka reached through stdcontext.inc). > > However, there are other places where INSET_MODIFY is used without any undo > call. Presumably this is handled somewhere in Gui, but I do not know where. > > Should I apply the patch right like it is, or is it better to add a > recordUndo statement in each and every INSET_MODIFY function? (or to revert > to the trivial first solution?)
i didn't get why is the first solution bad. it looks less prone-to-be-forgotten for some corner case or in future additions.(?) pavel
