[வெள்ளி ஜனவரி 24, 2025] Robert Pluim wrote: >>>>>> On Fri, 24 Jan 2025 10:37:37 +0530, Visuwesh <visuwe...@gmail.com> said: > > Visuwesh> * lisp/org-compat.el: Add workaround proposed by Robert Pluim > Visuwesh> <rpl...@gmail.com> to make yank-media work under GNOME reliably > for > Visuwesh> Emacs 29. > > Did I propose it? I donʼt remember 😀
You did, indeed. See https://yhetil.org/87wmlb8djy....@gmail.com. > Visuwesh> Link: https://list.orgmode.org/orgmode/87ed7kttoa....@k-7.ch > Visuwesh> Reported-by: Sébastien Gendre <s...@k-7.ch> > Visuwesh> --- > Visuwesh> lisp/org-compat.el | 15 +++++++++++++++ > Visuwesh> 1 file changed, 15 insertions(+) > > Visuwesh> diff --git a/lisp/org-compat.el b/lisp/org-compat.el > Visuwesh> index 91192629a..3ad330628 100644 > Visuwesh> --- a/lisp/org-compat.el > Visuwesh> +++ b/lisp/org-compat.el > Visuwesh> @@ -1992,6 +1992,21 @@ (make-obsolete-variable > 'org-speed-commands-user > Visuwesh> "configure `org-speed-commands' > instead." "9.5") > Visuwesh> (provide 'org-compat) > > Visuwesh> +;;;; yank-media > Visuwesh> +;; Emacs 29's pgtk port has a bug where it might fail to > return the > Visuwesh> +;; right TARGET. Install a workaround for Emacs <=29 since > the fix > Visuwesh> +;; went to Emacs 30. See bug#72254. > Visuwesh> +;; Org bug report link: > https://list.orgmode.org/orgmode/87ed7kttoa....@k-7.ch > Visuwesh> +;; This should be removed once we drop Emacs 29 support. > Visuwesh> +(when (<= emacs-major-version 29) > > I thought Ihor wanted this to check for a pgtk build? Personally I > think the check for the window system in the cl-defmethod is enough, > but Iʼm not Ihor. I think so too but let's hear from Ihor first. > One way to check for pgtk is checking the result of > (window-system). Or checking whether `pgtk-backend-display-class' is > `fboundp'. I don't think we should use (window-system) since the return value is FRAME dependent: if Someone™ opens a terminal frame first and loads Org while it is focused, they won't get this fix even if their build is a pgtk build. C: [ I think this will definitely break for daemon processes. ] We could rely on (fboundp 'pgtk-backend-display-class) but I somehow don't like it as compared to just leaving this extra check out.
