On Wed, Jun 29, 2022 at 07:11:10PM -0700, Samuel Wales wrote: > a few things taht are probably all completely obvious or investigated > or irrelevant just in case. just brainstorm.
I appreciate that! That's really what I'm asking for is ideas. I don't mind writing a bit of code, but I'm not sure where to start. > do you have everything relevant in the same subtrees? i.e. not > wanting granular, can you search upward for a dominating entry kind of > like git searching upward for the .git dir or so? property drawer > could control what's a dominating entry. you probably thoguht of this > or of having whatever categories as tags or categories in entries > though. in any case that would clock. you could even have clocking > clock into that no matter wher eyou are via some timer in principle. > just a brainstorm. you said dynamic som perhaps there is no > dominating entry for each category though. The core issue is I want to report on total time as an aggregate, not on time per task. Clock reporting gives me precise time accounting for a specific task. I want to fill whole hours with whatever tasks happened at those times chronologically. > org-time-stamp-rounding-minutes and org-clock-rounding-minutes . i > presume you don't find them relevant. I saw that. Interesting they don't appear to change the result, they modify the input when you record the time by rounding. > reminder: inactive ts in the clocked notation is usualy treated > separately by org [i.e. not hte same type of ts] from bare ia. I had considered perhaps converting inactive timestamps into clocking records. Unfortunately I think the core issue remains in aggregating by task, not by time. ------------------------------------------------------------------ Russell Adams rlad...@adamsinfoserv.com https://www.adamsinfoserv.com/