emacs-orgmode
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Accepted] [Orgmode] Bug: possible bug in latex export [7.01trans (relea


From: Carsten Dominik
Subject: [Accepted] [Orgmode] Bug: possible bug in latex export [7.01trans (release_6.36.735.g15ca.dirty)]
Date: Sun, 15 Aug 2010 08:36:19 +0200 (CEST)

Patch 175 (http://patchwork.newartisans.com/patch/175/) is now "Accepted".

Maintainer comment: No comment

This relates to the following submission:

http://mid.gmane.org/%3C22496.1280243082%40maps%3E

Here is the original message containing the patch:

> Content-Type: text/plain; charset="utf-8"
> MIME-Version: 1.0
> Content-Transfer-Encoding: 7bit
> Subject: [Orgmode] Bug: possible bug in latex export [7.01trans
>       (release_6.36.735.g15ca.dirty)]
> Date: Tue, 27 Jul 2010 20:04:42 -0000
> From: Stephen Eglen <address@hidden>
> X-Patchwork-Id: 175
> Message-Id: <address@hidden>
> To: David Maus <address@hidden>
> Cc: address@hidden, Stephen Eglen <address@hidden>
> 
> > FYI: This is a know limitation of the LaTeX exporter[1] and the
> > current state of this issue (dealing with skipped levels) is that
> > patches for the problem are welcome.  I agree that even if
> > skipped-level-headlines are not allowed, they shouldn't be silently
> > droped neither.
> > 
> > Best,
> >   -- David
> > 
> > [1] http://thread.gmane.org/gmane.emacs.orgmode/26413
> 
> I think I have a patch for this, see below.  I've tested it briefly to
> see that it works with both oddeven and odd level headings, but I would
> appreciate other people taking a closer look/checking that it doesn't
> break other things.  I think I found the problem, in that if you have a
> structure like
> 
> * A 
> 
> *** B
> 
> (i.e. a level one and a level 3)
> 
> after it has found the level 1 content, it expects to find a level 2
> subtree, whereas in fact there could be a level 2+ tree.  See the
> comment in the code for the change I made.
> 
> Stephen
> 
> 
> diff --git a/lisp/org-latex.el b/lisp/org-latex.el
> index 24dd8c1..8530723 100644
> --- a/lisp/org-latex.el
> +++ b/lisp/org-latex.el
> @@ -937,13 +937,28 @@ Return a list reflecting the document structure."
>  (defun org-export-latex-parse-subcontent (level odd)
>    "Extract the subcontent of a section at LEVEL.
>  If ODD Is non-nil, assume subcontent only contains odd sections."
> -  (if (not (org-re-search-forward-unprotected
> -         (concat "^\\(\\(?:\\*\\)\\{"
> -                 (number-to-string (+ (if odd 4 2) level))
> -                 "\\}\\) \\(.*\\)$")
> -         nil t))
> -      nil ; subcontent is nil
> -    (org-export-latex-parse-global (+ (if odd 2 1) level) odd)))
> +
> +  (let (nstars new-level)
> +    ;; In the search, we should not assume there will be exactly
> +    ;; LEVEL+1 stars in the next heading, as there may be more than
> +    ;; that number of stars.  hence the regexp should be \\*{N,}
> +    ;; rather than just \\*{N} (i.e. no upper bound, but N is minimum
> +    ;; number of stars to expect.)
> +    ;; We then have to check how many stars were found, rather than
> +    ;; assuming there were exactly N.
> +    (when (org-re-search-forward-unprotected
> +        (concat "^\\(\\(?:\\*\\)\\{"
> +                (number-to-string (+ (if odd 4 2) level))
> +                ",\\}\\) \\(.*\\)$")
> +        nil t)
> +      (setq nstars (1- (- (match-end 1) (match-beginning 1))))
> +      (setq new-level (if odd 
> +                       (/ (+ 3 nstars) 2);; not entirely sure why +3!
> +                     nstars)))
> +    (if nstars
> +     (org-export-latex-parse-global new-level odd)
> +      nil)                           ; subcontent is nil
> +      ))
>  
>  ;;; Rendering functions:
>  (defun org-export-latex-global (content)
> 



reply via email to

[Prev in Thread] Current Thread [Next in Thread]