Re: [AUCTeX] Re: tex2text

[Andreas Roehler]

> > > > > Couldn't find a convenient way to transfer tex-files
> > > > > into plain text.
> > > > http://www.tex.ac.uk/cgi-bin/texfaq2html?label=toascii
>
> > AR> OK, thanks.
>
> > AR> Unfortunatly:
>
> > AR> - dvi2tty
>
> > AR> doesn't handle words splitted by tex formatter,
> > AR> i.e. soft or syllable hyphens are still visible as
>
> > AR> Char: - (45, #o55, #x2d)
>
> DVI doesn't make a different between soft and hard hyphens unless you have
> a special font.

So it seems a common problem.

Here is `dinbrief2text' as I use it. How far it would be
onto a more general solution?

__
Andreas Roehler

;;; Commentary: Delete tex formatting
;;; elements. Preserve address, signature, phone
;;; etc. at a convenient place. Works with code
;;; employing `dinbrief' at the moment. Watching
;;; execution, I should be easy to adapt the script
;;; according to other packages in use.

;; This example code should facilitate a check

;; \documentclass[12pt]{dinbrief}
;; \usepackage{geometry}
;; \geometry{a4paper,bottom=2cm,top=2cm,left=35mm}
;; \usepackage[utf8x]{inputenc}
;; \usepackage[T1]{fontenc}
;; \usepackage{german}
;; \nowindowrules
;; \nowindowtics
;; % \stdaddress{JOE ANYONE}
;; \address{JOE ANYONE\\
;; ANYSTREET. 0 \\
;; 10407 Berlin\\}
;; \signature{JOE ANYONE}
;; % \place{XY-Stadt}
;; \begin{document}
;; \phone{+4930}{60000003}
;; \begin{letter}{An das\\
;; Final Judgement\\
;; 32. Senat\\
;; Postfach 7300\\[\medskipamount]
;; {\bf 10863 Berlin}}
;; \backaddress{JOE ANYONE, ANYSTREET. 0, 11067 Berlin}
;; \yourmail{7 XY 544/93 u.a.}
;; % \sign{123456}
;; \subject{Free Speech in Europe}
;; \opening{Sehr geehrter Herr Vorsitzender Richter K.,}
;;


;; It's not just § 130 German Legal Code, there are a lot of laws
;; made against free speech in Europa nowadays.
;;
;; \begin{quote}
;;
;; As philosopher Ludwig Wittgenstein put it:
;;
;; \glqq What we cannot speak about we must pass over
;; in silence\grqq{} . ... \end{quote}
;;
;; Is this the intended result?

;;
;; \closing{Regards}
;; % \ps{Wir bitten um schnelle Erledigung.}
;; % \cc{Kopie an:}
;; % \encl{Abschrift der Urkunde}
;; \end{letter}
;; \end{document}

;;; Code:

(defun tex2text ()
 "Delete tex-formatters, but preserve address, signature, phone etc."
 (interactive)
 (let ((oldbuf (buffer-name))
   (newbuf (concat (substring (buffer-name) 0 (string-match "\\." 
(buffer-name))) ".txt"))
   (backadress-flag nil))
   (set-buffer (get-buffer-create newbuf))
   (switch-to-buffer (current-buffer))
   (erase-buffer)
   (insert-buffer oldbuf)
   (goto-char (point-min))
   (while (re-search-forward "\\\\dots{}" nil t 1)
     (replace-match "..."))
   (goto-char (point-min))
   (while (re-search-forward "^%+.*\n" nil t 1)
     (replace-match ""))
   (goto-char (point-min))
   ;;     ;; {\bf 10863 Berlin}}
   (while (re-search-forward "^{\\\\[[:graph:]]+ \\([^}]+\\)}" nil t 1)
     (replace-match (match-string 1)))
   (goto-char (point-min))
   ;; [\medskipamount]
   (while (re-search-forward "\\[\\\\[^\]]+]" nil t 1)
     (replace-match ""))
   (goto-char (point-min))
   (goto-char (point-min))
   ;;     ^\\\\begin{letter}
   ;;         (string= (match-string 1) "\\subject")
   (while (re-search-forward "^[ \t]*\\\\begin{letter}\\|\\\\subject" 
nil t 1)
     (replace-match "\n")
     (kill-list-atpt)
     (insert (replace-regexp-in-string "[{}\\]" "" (format "%s" (car 
kill-ring))))
     (insert "\n"))
   (while (re-search-forward "^[ \t]*\\\\\\(end\\){quote}" nil t 1)
     (replace-match "\"")
     (forward-char -1)
     (delete-region (point) (progn (skip-chars-backward " \t\n\f") 
(point))))
   (goto-char (point-min))
   (while (re-search-forward "^[ \t]*\\\\\\(begin\\){quote}" nil t 1)
     (replace-match "\"")
     (delete-region (point) (progn (skip-chars-forward " \t\n\f") 
(point))))
   (goto-char (point-min))
   (while (re-search-forward "\\\\\\(glqq\\)" nil t 1)
     (if (save-match-data (in-string-p))
     (progn
       (replace-match "\'")
       (delete-region (point) (progn (skip-chars-forward " \t\n\f") 
(point)))
       (re-search-forward "\\\\\\(grqq\\){}" nil t 1)
       (replace-match "\'")
       (forward-char -1)
       (delete-region (point) (progn (skip-chars-backward " \t\n\f") 
(point))))
   (replace-match "\"")
   (delete-region (point) (progn (skip-chars-forward " \t\n\f") (point)))
   (re-search-forward "\\\\\\(grqq\\){}" nil t 1)
   (replace-match "\"")
   (forward-char -1)
   (delete-region (point) (progn (skip-chars-backward " \t\n\f") 
(point)))))
   (goto-char (point-min))
   (when (re-search-forward "^\\\\backaddress{\\([^}]+\\)}" nil t 1)
     (setq backadress-flag t)
     (save-excursion
   (goto-char (point-min))
   (insert (replace-regexp-in-string "[{}]" "" (match-string 1)))
   (newline))
     (kill-region (point) (line-beginning-position))
     (newline))
   (goto-char (point-min))
   (when (re-search-forward "^\\\\address" nil t 1)
     (if backadress-flag
     (progn
       (save-excursion
         (kill-region (point) (progn (forward-list) (point))))
       (kill-region (point) (line-beginning-position)))
   (replace-match (match-string 2))))
   (goto-char (point-min))
   (while (re-search-forward "^\\\\[^[{ ]+$" nil t 1)
     (replace-match ""))
   (goto-char (point-min))
   (while
   (re-search-forward 
"\\(^\\\\[^\[{]+\\)[\[{]\\{1\\}\\([^]}\n]+[\]}]\\{1\\}.*\\)$" nil t 1)
     (cond
      ((or (string= (match-string 1) "\\ps"))
   (save-excursion
     (goto-char (point-max))
     (newline 2)
     (insert "PS ")
     (insert (replace-regexp-in-string "[{}]" "" (match-string 2))))
   (kill-region (point) (line-beginning-position)))
      ((or (string= (match-string 1) "\\phone")
       (string= (match-string 1) "\\yourmail")
       (string= (match-string 1) "\\textbf")
       (string= (match-string 1) "\\closing"))
   (replace-match (replace-regexp-in-string "[{}]" "" (match-string 2)))
   (newline))
      ((string= (match-string 1) "\\opening")
   (replace-match (replace-regexp-in-string "[{}]" "" (match-string 2))))
      ((string= (match-string 1) "\\signature")
   (save-excursion
     (goto-char (point-max))
     (insert (replace-regexp-in-string "[{}]" "" (match-string 2))))
   (kill-region (point) (line-beginning-position)))
      (t
   (replace-match "\n"))))
   (goto-char (point-min))
   (while (re-search-forward "[\\]+" nil t 1)
     (replace-match "")))
 (nur-eine-leerzeile))