Hello,
I've observed, with the following minimal Org configuration,
--8<---------------cut here---------------start------------->8---
(require 'org)
(define-key global-map
(kbd "C-c a") 'org-agenda)
(setq org-agenda-files '("~/ecm.org"))
;; turn on individual word search
(setq org-agenda-search-view-always-boolean t
org-agenda-search-view-search-words-only t)
;; match part of a word
(setq org-agenda-search-view-force-full-words nil)
--8<---------------cut here---------------end--------------->8---
that the search string is:
- WELL interpreted as individual parts for `C-c a s', but
- NOT interpreted as individual parts for `C-c / /' inside the document itself
You can see (on http://screencast.com/t/JkjcirV2DZ) that the following ECM
document is found when searching
`C-c a s histo entre RET'.
--8<---------------cut here---------------start------------->8---
#+TITLE: ECM
#+LANGUAGE: fr
* Opérations
** TODO Ajouter l'historique de l'entreprise
La question "Historique de l'entreprise bon ou pas ?" est inutile dans le
processus...
Intégrer l'*historique PFI de l'entreprise* pour vérifier que ...
** TODO Historique
** TODO Entreprise
--8<---------------cut here---------------end--------------->8---
But, when inside the document,
- `C-c / / histo entre RET' FAILS
- `C-c / / historique entreprise RET' FAILS
- `C-c / / historique de l'entreprise RET' works
- `C-c / / rique de l'entreprise RET' works
That is:
- (setq org-agenda-search-view-always-boolean t)
is only honored for `C-c a s', NOT for `C-c / /'
- (setq org-agenda-search-view-force-full-words nil)
is honored for both `C-c a s' and `C-c / /'
Best regards,
Seb
--
Sebastien Vauban
