lisp/textmodes/tildify.el

   1 ;;; tildify.el --- adding hard spaces into texts
   2
   3 ;; Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
   4 ;;   2005, 2006, 2007, 2008 Free Software Foundation, Inc.
   5
   6 ;; Author:     Milan Zamazal <pdm@zamazal.org>
   7 ;; Version:    4.5
   8 ;; Keywords:   text, TeX, SGML, wp
   9
  10 ;; This file is part of GNU Emacs.
  11
  12 ;; GNU Emacs is free software: you can redistribute it and/or modify
  13 ;; it under the terms of the GNU General Public License as published by
  14 ;; the Free Software Foundation, either version 3 of the License, or
  15 ;; (at your option) any later version.
  16
  17 ;; GNU Emacs is distributed in the hope that it will be useful,
  18 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
  19 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  20 ;; GNU General Public License for more details.
  21
  22 ;; You should have received a copy of the GNU General Public License
  23 ;; along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.
  24
  25 ;;; Commentary:
  26
  27 ;; This package can be typically used for adding forgotten tildes in TeX
  28 ;; sources or adding `&nbsp;' sequences in SGML (e.g. HTML) texts.
  29 ;;
  30 ;; For example, the Czech orthography requires avoiding one letter
  31 ;; prepositions at line endings.  So they should be connected with the
  32 ;; following words by a tilde.  Some users forget to do this all the
  33 ;; time.  The purpose of this program is to check the text and suggest
  34 ;; adding of missing tildes on some places.  It works in a similar
  35 ;; manner to `query-replace-regexp'.
  36 ;;
  37 ;; The functionality of this program is actually performing query
  38 ;; replace on certain regions, but for historical reasons explained
  39 ;; above it is called `tildify'.
  40 ;;
  41 ;; The default variable settings are suited for Czech, so do not try to
  42 ;; understand them if you are not familiar with Czech grammar and spelling.
  43 ;;
  44 ;; The algorithm was inspired by Petr Ol¹ák's program `vlna'.  Abilities of
  45 ;; `tildify.el' are a little limited; if you have improvement suggestions, let
  46 ;; me know.
  47
  48 ;;; Code:
  49
  50
  51 ;;; *** User configuration variables ***
  52
  53
  54 (defgroup tildify nil
  55   "Adding missing hard spaces or other text fragments into texts."
  56   :version "21.1"
  57   :group 'wp)
  58
  59 (defcustom tildify-pattern-alist
  60   '((t "\\([,:;(][ \t]*[a]\\|\\<[AIKOSUVZikosuvz]\\)\\([ \t]+\\|[ \t]*\n[ \t]*\\)\\(\\w\\|[([{\\]\\|<[a-zA-Z]\\)" 2))
  61   "Alist specifying where to insert hard spaces.
  62
  63 Each alist item is of the form (MAJOR-MODE REGEXP NUMBER) or
  64 \(MAJOR-MODE . SYMBOL).
  65
  66 MAJOR-MODE defines major mode, for which the item applies.  It can be either:
  67 - a symbol equal to the major mode of the buffer to be fixed
  68 - t for default item, this applies to all major modes not defined in another
  69   alist item
  70
  71 REGEXP is a regular expression matching the part of a text, where a hard space
  72 is missing.  The regexp is always case sensitive, regardless of the current
  73 `case-fold-search' setting.
  74
  75 NUMBER defines the number of the REGEXP subexpression which should be replaced
  76 by the hard space character.
  77
  78 The form (MAJOR-MODE . SYMBOL) defines alias item for MAJOR-MODE.  For this
  79 mode, the item for the mode SYMBOL is looked up in the alist instead."
  80   :group 'tildify
  81   :type '(repeat (choice (list symbol regexp integer) (cons symbol symbol))))
  82
  83 (defcustom tildify-string-alist
  84   '((latex-mode . "~")
  85     (tex-mode . latex-mode)
  86     (plain-tex-mode . latex-mode)
  87     (sgml-mode . "&nbsp;")
  88     (xml-mode . sgml-mode)
  89     (html-mode . sgml-mode)
  90     (t . " "))
  91   "Alist specifying what is a hard space in the current major mode.
  92
  93 Each alist item is of the form (MAJOR-MODE . STRING) or
  94 \(MAJOR-MODE . SYMBOL).
  95
  96 MAJOR-MODE defines major mode, for which the item applies.  It can be either:
  97 - a symbol equal to the major mode of the buffer to be fixed
  98 - t for default item, this applies to all major modes not defined in another
  99   alist item
 100
 101 STRING defines the hard space, which is inserted at places defined by
 102 `tildify-pattern-alist'.  For example it can be \"~\" for TeX or \"&nbsp;\"
 103 for SGML.
 104
 105 The form (MAJOR-MODE . SYMBOL) defines alias item for MAJOR-MODE.  For this
 106 mode, the item for the mode SYMBOL is looked up in the alist instead."
 107   :group 'tildify
 108   :type '(repeat (cons symbol (choice string symbol))))
 109
 110 (defcustom tildify-ignored-environments-alist
 111   '((latex-mode
 112      ("\\\\\\\\" . "")          ; do not remove this
 113      ("\\\\begin{verbatim}" . "\\\\end{verbatim}")
 114      ("\\\\verb\\*?\\(.\\)" . (1))
 115      ("\\$\\$" . "\\$\\$")
 116      ("\\$" . "\\$")
 117      ("\\\\(" . "\\\\)")
 118      ("\\\\[[]" . "\\\\[]]")
 119      ("\\\\begin{math}" . "\\\\end{math}")
 120      ("\\\\begin{displaymath}" . "\\\\end{displaymath}")
 121      ("\\\\begin{equation}" . "\\\\end{equation}")
 122      ("\\\\begin{eqnarray\\*?}" . "\\\\end{eqnarray\\*?}")
 123      ("\\\\[a-zA-Z]+\\( +\\|{}\\)[a-zA-Z]*" . "")
 124      ("%" . "$"))
 125     (plain-tex-mode . latex-mode)
 126     (html-mode
 127      ("<pre[^>]*>" . "</pre>")
 128      ("<dfn>" . "</dfn>")
 129      ("<code>" . "</code>")
 130      ("<samp>" . "</samp>")
 131      ("<kbd>" . "</kbd>")
 132      ("<var>" . "</var>")
 133      ("<PRE[^>]*>" . "</PRE>")
 134      ("<DFN>" . "</DFN>")
 135      ("<CODE>" . "</CODE>")
 136      ("<SAMP>" . "</SAMP>")
 137      ("<KBD>" . "</KBD>")
 138      ("<VAR>" . "</VAR>")
 139      ("<! *--" . "-- *>")
 140      ("<" . ">"))
 141     (sgml-mode . html-mode)
 142     (t nil))
 143   "Alist specifying ignored structured text environments.
 144 Parts of text defined in this alist are skipped without performing hard space
 145 insertion on them.  These setting allow skipping text parts like verbatim or
 146 math environments in TeX or preformatted text in SGML.
 147
 148 Each list element is of the form
 149   (MAJOR-MODE (BEG-REGEX . END-REGEX) (BEG-REGEX . END-REGEX) ... )
 150
 151 MAJOR-MODE defines major mode, for which the item applies.  It can be either:
 152 - a symbol equal to the major mode of the buffer to be fixed
 153 - t for default item, this applies to all major modes not defined in another
 154   alist item
 155
 156 BEG-REGEX is a regexp matching beginning of a text part to be skipped.
 157 END-REGEX defines end of the corresponding text part and can be either:
 158 - a regexp matching the end of the skipped text part
 159 - a list of regexps and numbers, which will compose the ending regexp by
 160   concatenating themselves, while replacing the numbers with corresponding
 161   subexpressions of BEG-REGEX (this is used to solve cases like
 162   \\\\verb<character> in TeX)."
 163   :group 'tildify
 164   :type '(repeat (cons symbol (choice symbol (repeat sexp)))))
 165
 166
 167 ;;; *** Internal variables ***
 168
 169 (defvar tildify-count nil
 170   "Counter for replacements.")
 171
 172
 173 ;;; *** Interactive functions ***
 174
 175 ;;;###autoload
 176 (defun tildify-region (beg end)
 177   "Add hard spaces in the region between BEG and END.
 178 See variables `tildify-pattern-alist', `tildify-string-alist', and
 179 `tildify-ignored-environments-alist' for information about configuration
 180 parameters.
 181 This function performs no refilling of the changed text."
 182   (interactive "*r")
 183   (setq tildify-count 0)
 184   (let (a
 185         z
 186         (marker-end (copy-marker end))
 187         end-env
 188         finish
 189         (ask t)
 190         (case-fold-search nil)
 191         (regexp (tildify-build-regexp)) ; beginnings of environments
 192         aux)
 193     (if regexp
 194         ;; Yes, ignored environments exist for the current major mode,
 195         ;; tildify just texts outside them
 196         (save-excursion
 197           (save-restriction
 198             (widen)
 199             (goto-char (point-min))
 200             (while (not finish)
 201               (setq a (point))
 202               (setq end-env (tildify-find-env regexp))
 203               (setq z (copy-marker (if end-env (1- (point)) (point-max))))
 204               (if (>= (marker-position z) beg)
 205                   (progn
 206                     (or (>= a beg) (setq a beg))
 207                     (or (<= (marker-position z) (marker-position marker-end))
 208                         (setq z marker-end))
 209                     (setq aux (tildify-tildify a (marker-position z) ask))
 210                     (if (eq aux 'force)
 211                         (setq ask nil)
 212                       (if (eq aux nil)
 213                           (setq finish t)))))
 214               (if (>= (marker-position z) (marker-position marker-end))
 215                   (setq finish t))
 216               (or (>= (point) (marker-position z))
 217                   (goto-char (marker-position z)))
 218               (if (not finish)
 219                   (if (re-search-forward end-env nil t)
 220                       (if (> (point) (marker-position marker-end))
 221                           (setq finish t))
 222                     (message
 223                      "End of environment not found: %s" end-env)
 224                     (setq finish t))))))
 225       ;; No ignored environments, tildify directly
 226       (tildify-tildify beg end ask)))
 227   (message "%d spaces replaced." tildify-count))
 228
 229 ;;;###autoload
 230 (defun tildify-buffer ()
 231   "Add hard spaces in the current buffer.
 232 See variables `tildify-pattern-alist', `tildify-string-alist', and
 233 `tildify-ignored-environments-alist' for information about configuration
 234 parameters.
 235 This function performs no refilling of the changed text."
 236   (interactive  "*")
 237   (tildify-region (point-min) (point-max)))
 238
 239
 240 ;;; *** Auxiliary functions ***
 241
 242 (defun tildify-build-regexp ()
 243   "Build start of environment regexp."
 244   (let ((alist (tildify-mode-alist tildify-ignored-environments-alist))
 245         regexp)
 246     (when alist
 247       (setq regexp (caar alist))
 248       (setq alist (cdr alist))
 249       (while alist
 250         (setq regexp (concat regexp "\\|" (caar alist)))
 251         (setq alist (cdr alist)))
 252       regexp)))
 253
 254 (defun tildify-mode-alist (mode-alist &optional mode)
 255   "Return alist item for the MODE-ALIST in the current major MODE."
 256   (if (null mode)
 257       (setq mode major-mode))
 258   (let ((alist (cdr (or (assoc mode mode-alist)
 259                         (assoc t mode-alist)))))
 260     (if (and alist
 261              (symbolp alist))
 262         (tildify-mode-alist mode-alist alist)
 263       alist)))
 264
 265 (defun tildify-find-env (regexp)
 266   "Find environment using REGEXP.
 267 Return regexp for the end of the environment or nil if no environment was
 268 found."
 269   ;; Find environment
 270   (if (re-search-forward regexp nil t)
 271       ;; Build end-env regexp
 272       (let ((match (match-string 0))
 273             (alist (tildify-mode-alist tildify-ignored-environments-alist))
 274             expression)
 275         (save-match-data
 276           (while (not (eq (string-match (caar alist) match) 0))
 277             (setq alist (cdr alist))))
 278         (if (stringp (setq expression (cdar alist)))
 279             expression
 280           (let ((result "")
 281                 aux)
 282             (while expression
 283               (setq result (concat result
 284                                    (if (stringp (setq aux (car expression)))
 285                                        expression
 286                                      (regexp-quote (match-string aux)))))
 287               (setq expression (cdr expression)))
 288             result)))
 289     ;; Return nil if not found
 290     nil))
 291
 292 (defun tildify-tildify (beg end ask)
 293   "Add tilde characters in the region between BEG and END.
 294 This function does not do any further checking except of for comments and
 295 macros.
 296
 297 If ASK is nil, perform replace without asking user for confirmation.
 298
 299 Returns one of symbols: t (all right), nil (quit), force (replace without
 300 further questions)."
 301   (save-excursion
 302     (goto-char beg)
 303     (let* ((alist (tildify-mode-alist tildify-pattern-alist))
 304            (regexp (car alist))
 305            (match-number (cadr alist))
 306            (tilde (tildify-mode-alist tildify-string-alist))
 307            (end-marker (copy-marker end))
 308            answer
 309            bad-answer
 310            replace
 311            quit
 312            (message-log-max nil))
 313       (while (and (not quit)
 314                   (re-search-forward regexp (marker-position end-marker) t))
 315         (when (or (not ask)
 316                   (progn
 317                     (goto-char (match-beginning match-number))
 318                     (setq bad-answer t)
 319                     (while bad-answer
 320                       (setq bad-answer nil)
 321                       (message "Replace? (yn!q) ")
 322                       (setq answer (read-event)))
 323                     (cond
 324                      ((or (eq answer ?y) (eq answer ? ) (eq answer 'space))
 325                       (setq replace t))
 326                      ((eq answer ?n)
 327                       (setq replace nil))
 328                      ((eq answer ?!)
 329                       (setq replace t
 330                             ask nil))
 331                      ((eq answer ?q)
 332                       (setq replace nil
 333                             quit t))
 334                      (t
 335                       (message "Press y, n, !, or q.")
 336                       (setq bad-answer t)))
 337                     replace))
 338           (replace-match tilde t t nil match-number)
 339           (setq tildify-count (1+ tildify-count))))
 340       ;; Return value
 341       (cond
 342        (quit nil)
 343        ((not ask) 'force)
 344        (t t)))))
 345
 346
 347 ;;; *** Announce ***
 348
 349 (provide 'tildify)
 350
 351
 352 ;; Local variables:
 353 ;; coding: iso-latin-2
 354 ;; End:
 355
 356 ;; arch-tag: fc9b05a6-7355-4639-8170-dcf57853ba22
 357 ;;; tildify.el ends here