miraculously i made a buggy solution for myself only. it might sort wrong if you have [-] in rest of line.
so fwiw. in org-list.el. for anybody who is searching for same thing: ((= dcst ?x) (or (and (stringp (match-string 1)) (replace-regexp-in-string "\\[-\\]" "[ ]" ;; " - [X] whatever") ;; " - [ ] whatever") ;; " - [-] whatever") (match-string 1))) On 7/1/22, Samuel Wales <samolog...@gmail.com> wrote: > i am confused by the custom sorting function for plain > lists. are there examples? [note: still on maint.] > > i want to ignore [-] for sorting by checked. it can be equal to [ ]. > i don't need it to be custom but that seems available. > > a rationale and possible interesting solutions are below, but i'm ok > with anything. current x/X is always flawed for me. > > > rationale: > > suppose you have a long list like > > - [ ] hello > + [ ] hi > + [ ] greetings > ... [long list] > > suppose you mark greetings as [X] with c-c c-c, at least with my > settings, hello will become [-] to indicate "partly X". > > suppose hello is still high priority. but you don't notice it's there > because you use very large fonts and it is not on the same page. it's > in the middle. you keep your list in priority sequence. you have at > the top something like > > - [ ] bonjour > - [ ] some kind of greeting > ... > > and most are spc like that and - is rare. suppose you mark bonjour X > with c-c c-c. now it is in your face and you want to move it down. > so you sort by checked. now it is out of your face. but you didn't > notice that your hello moved down also. > > this isn't particularly a bug; it is just that - is part of sorting > and it is hardcoded to be below SPC [i think]. > > so hello gets moved down a whoooooooooooooole lot. its place in the > list is gone. you aren't even looking at the bottom of hte list > because it is so long. it is as if hello has disappeared. > > and this is because you marked a sub-item as X. and then sorted the > top level by checked. and didn't notice. > > so i'm thinking this is a feature that could cause unexpected results. > [because it did that to me. existence proof.] > > and there's nothing really wrong with existing semantics, but i'd want > - to be ignored in sorting, because of the above. > > i can think of some possible solutions. for example > > - have something like an x X command that makes - eqal to spc or > custom variable for sorting with ability to specify '(? ?x) thus > ignoring ?-. > - have a command that moves all X to a sibling header so you don't > need sorting to get X out of your face > - have the possibility of this all working on sublists too > > > [i kinda want all.... there are more ideas. please do not shoot me > or say i am destroying the spirit and letter of org and the milky way > galaxy. these are just brainstorms. possibilities to possibly > consider, not analyzed to perfection.] > > thanks! > > > On 5/8/17, Kyle Meyer <k...@kyleam.com> wrote: >> Kyle Meyer <k...@kyleam.com> writes: >> >>> Nicolas Goaziou <m...@nicolasgoaziou.fr> writes: >>> >>>> The we may not need `call-interactively' at all. >>>> >>>> WDYT? >>> >>> Yeah, I agree that there's no need for call-interactively here because >>> the interactive forms of org-table-sort-lines, org-sort-list, >>> org-sort-entries are covered by org-sort's. >>> >>> Switched call-interactively to funcall in c1addc825. >> >> Ehh, I should have looked more closely at org-table-sort-lines. Unlike >> org-sort-entries and org-sort-list, it uses called-interactively-p to >> determine whether it should prompt the user. I've put the >> org-call-with-arg back, at least for now. >> >> -- >> Kyle >> >> > > > -- > The Kafka Pandemic > > A blog about science, health, human rights, and misopathy: > https://thekafkapandemic.blogspot.com > -- The Kafka Pandemic A blog about science, health, human rights, and misopathy: https://thekafkapandemic.blogspot.com