Dear Nick,

Apologies, you're absolutely right: the answer was already given in the
list and I must have missed it somehow. (setq-default cache-long-scans nil)
fixes the problem.

Thank you very much!

Regards,

FC


On Tue, Jul 22, 2014 at 1:37 AM, Nick Dokos <ndokos@gmail.com> wrote:

> Fletcher Charest <fletcher.charest@gmail.com> writes:
>
>
> > I am confronted to this bug in every version of Org (shipped with Emacs,
> from maint, and from master). Here is how I can reproduce it every time:
> >
> > 1. emacs -Q (in my case GNU Emacs 24.3.50.1, with Org-mode version
> 8.2.3a)
> >
> > 2. Evaluate this code:
> >
> > (setq org-capture-templates
> >       '(("t" "test" entry (file+olp "~/org2.org" "Todo" "Refile") "*
> TODO %?" :prepend t)))
> >
> > 3. Create the "~/org2.org" file like this:
> >
> > * Todo
> > ** Refile
> > * Other headline 1
> > * Other headline 2
> >
> > 4. M-x org-capture, then do a capture with t.
> >
> > 5. The cycling behaviour gets really weird in org2.org, from the
> captured task until the end of the file. It is just like if whole area was
> considered as a single line (this becomes apparent by turning hl-line-mode
> on). This error
> > appears while trying to cycle with TAB in the area: "byte-code: Invalid
> search bound (wrong side of point)".
> >
> > 6. Close the file and reopen it: everything is fine.
> >
> > Is anyone else confronted to this? I don't know if this is Emacs or
> Org-related.
> >
>
> Searching for "invalid search bound" in the list archives will produce
> lots of results. The last time that this appeared it was a bug in emacs
> (a recently introduced caching mechanism) that was fixed by Eli
> Zaretskii (emacs bug 16265).
>
> I cannot reproduce the problem with my setup:
>
> GNU Emacs 24.4.50.2 (x86_64-unknown-linux-gnu, GTK+ Version 3.10.9) of
> 2014-07-07
> Org-mode version 8.2.7 (release_8.2.7 @ /home/nick/elisp/org-mode/lisp/)
>
> so my guess is that your version of emacs does not contain Eli's fix.
>
> Nick
>
>
>
>
>