"Jacob S. Gordon" <[email protected]> writes: > On 2026-05-10 07:02, Ihor Radchenko wrote: >> This assumes that all the instances of filtered/invisible entry are >> really hidden. Do we really need to make such assumption? > > IUUC ‘(not (invisible-p …))’ should be used instead to respect > ‘buffer-invisibility-spec’.
This is right. I initially made that small amendment myself, but then saw that the invisibility check is only done once. > ... Or, are you referring to the fact that I > only check visibility of the closest overlay? The latter. > ... In my testing I couldn’t > come up with a scenario where only some of the clocking overlays were > hidden, There is currently no such scenario. But there is nothing stopping some future agenda filters from applying filters to certain days only. So, I'd prefer something more future-proof that does not rely on undocumented state of affairs. > ... but I can check inside the loop: > > (sort (remq nil (delete-dups > (mapcar (lambda (o) > (when-let* ((_ (eq (overlay-get o 'type) > 'org-agenda-clocking)) > (start (overlay-start o)) > (_ (not (invisible-p start)))) > start)) > (overlays-in (point-min) (point-max))))) > :key (lambda (p) (count-lines p pt t))) Yup, something like that. -- Ihor Radchenko // yantar92, Org mode maintainer, Learn more about Org mode at <https://orgmode.org/>. Support Org development at <https://liberapay.com/org-mode>, or support my work at <https://liberapay.com/yantar92>
