Jens Schmidt <[email protected]> writes: > On 2026-05-04 05:09, J.D. Smith wrote: > >> [...] > > Sorry for the late reply, and sorry for not replying to your mail in > more detail. My provider had IMAP connectivity issues, and now I'm > catching up. > > > Some short remarks, will try to test more when I have more time: > > - There is still some debugging output active: > > (message "Setting appearance with inside: %s showing: %s beg: %d end: > %d" ...)
Fixed in latest. > - Maybe it's just me, but for a moment I thought I could even use > `org-inside-toggle-hidden' without `org-inside-mode' being active. > I just misunderstood your instructions. Probably you should > mention that in o-i-t-h's docstring or even user-error out when > o-i-t-h gets called w/o o-i-m being on. Updated doc string. > - Please allow setting `org-inside-appearance' to nil. With o-i-t-h > in place that setting does make sense, IMHO. And if I first switch > on o-i-m and then set o-i-a to nil, things work as I'd like to have > them: No automatic effects at all, but o-i-t-h makes the markers > temporarily visible. A good idea, and easy to support; please check the latest. > - And here is another funny repro: > > make EMACS=/usr/local/emacs-30.2/bin/emacs repro > M-: (setq org-hide-emphasis-markers t) > M-x org-inside-mode RET (yes, in *scratch*!) > C-x C-f ~/tmp/test.org RET > M-x org-inside-mode RET > Move into emphasized text and let the cursor turn "bar" > C-x b *scratch* RET > > => Cursor is bar-shaped in *scratch*. I've improved the buffer change logic to "re-sense" as needed, but also to avoid "double entry" bugs. Apparently sometimes (but not always) the real cursor-sensor function is called when a buffer appears and point is inside sensor-marked text. This will need thorough checking on by v30 and v31, since for the latter we do not use a global `window-buffer-change-functions' entry.
