Re: [O] [bug] Symbol's function definition is void: org-pop-to-buffer-same-window

[Nick Dokos]

Sebastien Vauban <address@hidden> wrote:

> > I don't think it's a bug in org: if you start with a clean copy of the
> > repo (make clean; make) and have your load-path pointing there, you
> > should not see any problems.
> >
> > IIUC, Michael Bach's problem (a fairly common one, btw) was that he was
> > mixing different installations: he was picking up a recent org-capture
> > that was using org-pop-to-buffer-same-window, but an older org-compat
> > that did not include the definition of that function.
> 
> I don't see why I would suddenly have mixed paths in my .emacs file -- though
> this can always happen, as it is changed almost daily, here and there.
> 
> > You have to make sure that all of the pieces come from the same place,
> > but there is still a gotcha in the sense that an old .elc file will
> > be picked up by emacs in preference to a new .el file (possibly with
> > a warning, but who looks at warnings?)
> >
> > Hence assuming you are using git to keep up to date:
> >
> >     make clean
> 
> I don't use compiled files, so problem is not that.
> 
> >     make
> >     C-h v load-path RET
> 
> load-path's value is
> (...
> "c:/home/sva/src/org-mode/contrib/lisp"
> "c:/home/sva/src/org-mode/lisp"
> ...
> "c:/Program Files/Emacs-24.0/lisp/org"
> ...)
> 
> >     [make sure that your load path points at your git directory *first*]
> 
> Checked.
> 
> >     M-x org-reload RET
> >     [or perhaps safer: restart emacs]
> >
> > And use M-x locate-library to make sure that emacs agrees with you about
> > where things are picked up from.
> 
> Library is file ~/src/org-mode/lisp/org-compat.el
> 
> Library is file ~/src/org-mode/lisp/org.el
> 
> It's not obvious what I would have changed for getting this now... but I'll
> diff my emacs config files, if you say you don't have such a problem.
> 

It is possible that any particular pull from the git repo will get you
a buggy version - so when I said it's not a bug, I should have said:
it's not a bug with the version that I tested at the time. That's
reason #357 for *always* reporting the org version one is using.

With that caveat stated, if the steps above don't resolve the problem,
then it might be a bug in org, but, given the history of problems with
mixed installs[fn:1], it behooves one to make sure that one's install does
*not* have that problem: locate-library is your friend.

And, fwiw, I don't see the problem with Org-mode version 7.7
(release_7.7.617.gb1f2).

Nick

Footnotes:

[fn:1] ... assuming I have not misinterpreted the underlying cause of
       Michael Bach's problem in the thread that Kenny Myers quoted, of
       course.