Ship Mints <shipmints@gmail.com> writes:

> 1.  ( ) text/plain          (*) text/html           
>
> On Tue, Mar 18, 2025 at 11:15 AM Thierry Volpiatto <thievol@posteo.net> wrote:
>
>     Ship Mints <shipmints@gmail.com> writes:
>   
>     > 1.  ( ) text/plain          (*) text/html           
>     >
>     > On Tue, Mar 18, 2025 at 2:55 AM Thierry Volpiatto <thievol@posteo.net> wrote:
>     >
>     >     Sorry for late reply, was busy.
>     >   
>     >     Ship Mints <shipmints@gmail.com> writes:
>     >   
>     >     > On Sun, Mar 16, 2025 at 5:10 PM Ship Mints <shipmints@gmail.com> wrote:
>     >     >
>     >     >     On Sat, Mar 15, 2025 at 10:18 AM Ship Mints <shipmints@gmail.com> wrote:
>     >     >   
>     >     >         On Sat, Mar 15, 2025 at 1:37 AM Thierry Volpiatto <thievol@posteo.net> wrote:
>     >     >       
>     >     >             Ship Mints <shipmints@gmail.com> writes:
>     >     >           
>     >     >             > I have workarounds that work only for the most simplistic cases.  Many
>     >     >             > of our bookmarks themselves contain embedded bookmarks and bookmark
>     >     >             > references (which are individually addressable so can be used
>     >     >             > separately) with window-states we need to restore in tab-bar tabs that
>     >     >             > they represent.
>     >     >           
>     >     >             I don't really understand what your packages are doing or are intended
>     >     >             doing, but FWICS in bufferlo: You are using in some places
>     >     >             (bookmark-jump name #'ignore); why don't you do all this work (restore
>     >     >             window-states in tab) in DISPLAY-FUNCTION instead of using `ignore`?
>     >     >             Your handler would be much simpler by moving the window-state-put and
>     >     >             alike calls in DISPLAY-FUNCTION:
>     >     >           
>     >     >             (bookmark-jump name #'your_function_restoring_window_or_frame_state) 
>     >     >           
>     >     >             Using (bookmark-jump name #'ignore) with all the code that jump to
>     >     >             frame/tab etc... in the handler is just a workaround to fix the previous
>     >     >             buggy behavior of bookmark--jump-via. IMO.
>     >     >           
>     >     >             It would be good to start with a good example or recipe to see if we can
>     >     >             find a good solution.
>     >     >
>     >     >         We need the bookmarks to work from the bookmark menu where no display-function overrides are supported.
>     >     >       
>     >     >         I suggest we add bookmark-record keys that indicate to bookmark-jump to inhibit/or allow messing with window-configurations.  The bufferlo bookmarks (and Adam's if he
>     wants)
>     >     would
>     >     >         contain these hint keys.
>     >     >       
>     >     >         '(bookmark-jump-inhibit-window-actions . t) ; or whatever we come up with
>     >     >       
>     >     >         I can contrive an example, if necessary, but I believe y'all get the point.  Nested bookmarks whose handlers expect their window-configurations not to be messed with up
>     the
>     >     >         chain, will always be broken without additional controls.
>     >     >
>     >     >     The attached patch implements such a scheme that works for us, and is transparent to other bookmark uses.
>     >     >
>     >     > Perhaps we should restore bookmark--jump-via to its previous behavior
>     >     > and better document the "rules of the road" for bookmark handlers. 
>     >     > For simple file- and point-based bookmarks, handlers need to ensure
>     >     > that when they return, the selected window and current buffer are
>     >     > what's intended.  For bookmark handlers that perform other actions,
>     >     > those rules need not apply to leverage the bookmark infrastructure.
>     >   
>     >     What we could do is propose a more flexible solution so that you could
>     >     use whatever you want for bookmark--jump-via; With what you have proposed so
>     >     far, you still have the problem of DISPLAY-FUNCTION which will always
>     >     run (I see there is comments about this problem in your mentionned
>     >     packages), with the patch below you could define a display-function
>     >     entry in your bookmark-record e.g. (display-function . ignore) and then
>     >     add a special method for bookmark--jump-via:
>     >   
>     >     (cl-defmethod bookmark--jump-via (bookmark-name-or-record (_ (eql 'ignore)))
>     >       (do_watever_you_want_here)) ; e.g. run only the handler fn.
>     >   
>     >     NOTE: I used 'ignore as example but you could use whatever you want.
>     >   
>     >     Here the patch:
>     >   
>     >     diff --git a/lisp/bookmark.el b/lisp/bookmark.el
>     >     index 99bb26e83cc..e594387f364 100644
>     >     --- a/lisp/bookmark.el
>     >     +++ b/lisp/bookmark.el
>     >     @@ -1259,7 +1259,7 @@ it to the name of the bookmark currently being set, advancing
>     >        "Hook run after `bookmark-jump' jumps to a bookmark.
>     >      Useful for example to unhide text in `outline-mode'.")
>     >   
>     >     -(defun bookmark--jump-via (bookmark-name-or-record display-function)
>     >     +(cl-defgeneric bookmark--jump-via (bookmark-name-or-record display-function)
>     >        "Handle BOOKMARK-NAME-OR-RECORD, then call DISPLAY-FUNCTION.
>     >      DISPLAY-FUNCTION is called with the new buffer as argument.
>     >   
>     >     @@ -1319,8 +1319,12 @@ DISPLAY-FUNC would be `switch-to-buffer-other-window'."
>     >        ;; Don't use `switch-to-buffer' because it would let the
>     >        ;; window-point override the bookmark's point when
>     >        ;; `switch-to-buffer-preserve-window-point' is non-nil.
>     >     -  (bookmark--jump-via bookmark (or display-func 'pop-to-buffer-same-window)))
>     >     +  (bookmark-jump-1 bookmark display-func))
>     >   
>     >     +(defun bookmark-jump-1 (bookmark display-func)
>     >     +  (let ((dfn (or (bookmark-prop-get bookmark 'display-function)
>     >     +                 display-func 'pop-to-buffer-same-window)))
>     >     +    (bookmark--jump-via bookmark dfn)))
>     >   
>     >      ;;;###autoload
>     >      (defun bookmark-jump-other-window (bookmark)
>     >     @@ -2303,7 +2307,7 @@ the related behaviors of `bookmark-save' and `bookmark-bmenu-save'."
>     >              (pop-up-windows t))
>     >          (delete-other-windows)
>     >          (switch-to-buffer (other-buffer) nil t)
>     >     -    (bookmark--jump-via bmrk 'pop-to-buffer)
>     >     +    (bookmark-jump-1 bmrk 'pop-to-buffer)
>     >          (bury-buffer menu)))
>     >
>     >     @@ -2317,7 +2321,7 @@ the related behaviors of `bookmark-save' and `bookmark-bmenu-save'."
>     >        "Select this line's bookmark in other window, leaving bookmark menu visible."
>     >        (interactive nil bookmark-bmenu-mode)
>     >        (let ((bookmark (bookmark-bmenu-bookmark)))
>     >     -    (bookmark--jump-via bookmark 'switch-to-buffer-other-window)))
>     >     +    (bookmark-jump-1 bookmark 'switch-to-buffer-other-window)))
>     >
>     >      (defun bookmark-bmenu-other-frame ()
>     >     @@ -2333,7 +2337,7 @@ The current window remains selected."
>     >        (interactive nil bookmark-bmenu-mode)
>     >        (let ((bookmark (bookmark-bmenu-bookmark))
>     >             (fun (lambda (b) (display-buffer b t))))
>     >     -    (bookmark--jump-via bookmark fun)))
>     >     +    (bookmark-jump-1 bookmark fun)))
>     >   
>     >      (defun bookmark-bmenu-other-window-with-mouse (event)
>     >        "Jump to bookmark at mouse EVENT position in other window.
>     >   
>     >     Also I guess trying to call bookmark-jump-other-window and friends is
>     >     failing with your special bookmarks, with this it would run just as
>     >     bookmark-jump without (possible) errors.
>     >   
>     >     WDYT?
>     >
>     > Thanks for the continuing discussion.
>     >
>     > The concept will work but it feels a bit over-engineered.
>   
>     It is not, it is quite simple.
>   
>     > The approach of ignoring save-window-excursion and display-func via
>     > bookmark-record entries or using properties on the handler seem less
>     > intrusive or a mix, if we feel that's appropriate.
>   
>     I proposed this solution to help you cleaning up your code which is full
>     of workarounds for the current behavior (prior 31).  Of course if you
>     don't want to make an effort to update your code, what you propose is
>     simpler (i.e. you have nothing to change on your side), but generally we
>     (external emacs extensions developers) try to adapt ourselves to Emacs
>     source and not the contrary.
>
> Thanks for the input.
>
> The idea that I "don't want to make an effort" is insulting.

Sorry if you take it like this, it was not the intention.

>   Perhaps a little less coffee.

I don't drink coffee.

>     > Why not just fix the eww bookmark handler to do its own
>     > save-window-excursion, again, rather than make a default bookmark jump
>     > behavior policy change?
>   
>     Because the problem is not just about eww, but more generally on how
>     bookmark handlers work.
>
> Curious to know which other ones are broken?  I read eww and w3m.

It is not only about eww AND w3m.  The point is not if things are broken
or not, it is to provide a good API for all bookmarks (and future kind
of bookmarks).


> What do the Emacs maintainers think about this as a matter of taste,
> easy adoption for other bookmark users, and idiomatic usage?

Now Eli and other maintainers will decide what is the best for emacs.