Richard Stallman <[EMAIL PROTECTED]> writes: > The code that gives this advice handles only the simple cases. > As you have pointed out, to even try to handle the hard cases > is very hard.
True, but without any hooks, it is not only hard -- it is impossible. > We certainly don't want to do that now, and it might not be worth > doing ever. For that specific purpose (the bad advise), it may be sufficient for the change-window-buffer-functions hook to simply note that the previous window is already visible in another window, and as such it could just avoid giving the false advise (i.e. not try to give the correct advise). Alternatively, it could make a temporary binding for `q' and a [restore] button in the *Help* buffer which restores the previous window and point of that window (making suitable checks in case the old buffer no longer exists etc.) And give the advice that "click [restore] button to restore window". > I also see potential problems in > putting hooks on every state-changing activity in Emacs, such as > displaying a buffer in a window. IMO it is rather obscure that we have a hook (several in fact) which are run _after_ displaying a buffer in a window -- but no hook which is run before the change. -- Kim F. Storm <[EMAIL PROTECTED]> http://www.cua.dk _______________________________________________ emacs-pretest-bug mailing list [email protected] http://lists.gnu.org/mailman/listinfo/emacs-pretest-bug
