OK, problem found. The helm code is this:
(defun helm-org-goto-marker (marker) (switch-to-buffer (marker-buffer marker)) (goto-char (marker-position marker)) (org-show-context) (re-search-backward "^\\*+ " nil t) (org-show-entry)) So, org-show-entry is called and it results in ellipsis instead of children. > org-show-entry > Show the body directly following this heading. > Show the heading too, if it is currently invisible. Fair to say, it does what it says. I will go w/ custom a function which uses org-cycle for now, though the helm implementation would probably be better off having some general org-show function which works like org-show-context (i.e. refers to some list of customizable defaults + takes an argument) but for headings. Is there such function? 27.03.2019, 16:29, "Nicolas Goaziou" <m...@nicolasgoaziou.fr>: > Hello, > > pinkanon pinkanon <pinkanon.pinka...@yandex.ru> writes: > >> Hi! I use helm, which can list all headings using >> helm-org-in-buffer-headings, >> and lets you choose one to jump to. >> When I jump to a heading, the tree is expanded. >> >> My problem is as follows. >> >> When I expand this tree on "(2) Downers" (when it is collapsed): >> >> * (1) Uppers >> ** (2) Downers >> screamers >> *** laughers >> ...and also a quart of tequila >> ** (3) a quart of rum, >> a case of Budweiser, >> ** (4) a pint of raw ether and two dozen amyls. >> >> I get: >> >> * (1) Uppers >> ** (2) Downers >> screamers >> ... >> ** (3) a quart of rum,... >> ** (4) a pint of raw ether and two dozen amyls.... >> >> Instead of: >> >> * (1) Uppers >> ** (2) Downers >> screamers >> *** laughers... >> ** (3) a quart of rum,... >> ** (4) a pint of raw ether and two dozen amyls. >> >> I was advised to modify 'org-show-context-detail, as helm uses >> org-show-context function. >> But none of the possible values fixed the behaviour for me. >> I used 'canonical for the example above: >> (setf (alist-get 'default org-show-context-detail) 'canonical) > > You may want to debug `org-show-set-visibility' first. If it gives no > clue, you may want to report it to Helm. > > Regards, > > -- > Nicolas Goaziou
