0

I've noticed something strange: the keyword-search (C-c a s) is much faster than the search of property-search (C-c a m, with prop="blah").

Example:
A property-search C-c a m: Match: PROP1="value1" takes about 20 seconds.
The keyword-search C-c a s: Phrase or [+-]Word/{Regexp} ...: value1 takes 3 seconds! And the property "PROP1" occurs only one time in my agenda.
(I have about 30 agenda files.)
And of course the value of the variable org-use-property-inheritance is nil.

I thought the property-search must be much faster, because org must only search the text of the PROPERTIES block of each entry and not the whole body text of each entry. Or am I wrong?

Also strange: I've tested a normal tag-search (C-c a m): Match: value1 (with the tag: value1). It was also very fast (about 3 seconds). I thought this must be slower because the value of the variable org-agenda-use-tag-inheritance isn't nil (it is: (todo search agenda)).

Any ideas to understand the problem, or to solve it?

(I have the latest version of org-mode: 9.4 and of emacs: 27.1)

1
  • 2
    You can profile the code. I have to say 20s is way too long, but I'm not surprised that the property search is slower: I believe it uses org-element (the Org mode parser) on the agenda files (although you should verify this: I have not checked properly). – NickD Mar 7 at 17:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.