code.delx.au - gnu-emacs/blob - lisp/gnus/rfc2231.el

   1 ;;; rfc2231.el --- Functions for decoding rfc2231 headers
   2
   3 ;; Copyright (C) 1998, 1999, 2000 Free Software Foundation, Inc.
   4
   5 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
   6 ;; Maintainer: bugs@gnus.org
   7 ;; This file is part of GNU Emacs.
   8
   9 ;; GNU Emacs is free software; you can redistribute it and/or modify
  10 ;; it under the terms of the GNU General Public License as published by
  11 ;; the Free Software Foundation; either version 2, or (at your option)
  12 ;; any later version.
  13
  14 ;; GNU Emacs is distributed in the hope that it will be useful,
  15 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
  16 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  17 ;; GNU General Public License for more details.
  18
  19 ;; You should have received a copy of the GNU General Public License
  20 ;; along with GNU Emacs; see the file COPYING.  If not, write to the
  21 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  22 ;; Boston, MA 02111-1307, USA.
  23
  24 ;;; Commentary:
  25
  26 ;;; Code:
  27
  28 (eval-when-compile (require 'cl))
  29 (require 'ietf-drums)
  30
  31 (defun rfc2231-get-value (ct attribute)
  32   "Return the value of ATTRIBUTE from CT."
  33   (cdr (assq attribute (cdr ct))))
  34
  35 (defun rfc2231-parse-string (string)
  36   "Parse STRING and return a list.
  37 The list will be on the form
  38  `(name (attribute . value) (attribute . value)...)"
  39   (with-temp-buffer
  40     (let ((ttoken (ietf-drums-token-to-list ietf-drums-text-token))
  41           (stoken (ietf-drums-token-to-list ietf-drums-tspecials))
  42           (ntoken (ietf-drums-token-to-list "0-9"))
  43           (prev-value "")
  44           display-name mailbox c display-string parameters
  45           attribute value type subtype number encoded
  46           prev-attribute)
  47       (ietf-drums-init (mail-header-remove-whitespace
  48                         (mail-header-remove-comments string)))
  49       (let ((table (copy-syntax-table ietf-drums-syntax-table)))
  50         (modify-syntax-entry ?\' "w" table)
  51         ;; The following isn't valid, but one should be liberal
  52         ;; in what one receives.
  53         (modify-syntax-entry ?\: "w" table)
  54         (set-syntax-table table))
  55       (setq c (char-after))
  56       (when (and (memq c ttoken)
  57                  (not (memq c stoken)))
  58         (setq type (downcase (buffer-substring
  59                               (point) (progn (forward-sexp 1) (point)))))
  60         ;; Do the params
  61         (while (not (eobp))
  62           (setq c (char-after))
  63           (unless (eq c ?\;)
  64             (error "Invalid header: %s" string))
  65           (forward-char 1)
  66           ;; If c in nil, then this is an invalid header, but
  67           ;; since elm generates invalid headers on this form,
  68           ;; we allow it.
  69           (when (setq c (char-after))
  70             (if (and (memq c ttoken)
  71                      (not (memq c stoken)))
  72                 (setq attribute
  73                       (intern
  74                        (downcase
  75                         (buffer-substring
  76                          (point) (progn (forward-sexp 1) (point))))))
  77               (error "Invalid header: %s" string))
  78             (setq c (char-after))
  79             (setq encoded nil)
  80             (when (eq c ?*)
  81               (forward-char 1)
  82               (setq c (char-after))
  83               (when (memq c ntoken)
  84                 (setq number
  85                       (string-to-number
  86                        (buffer-substring
  87                         (point) (progn (forward-sexp 1) (point)))))
  88                 (setq c (char-after))
  89                 (when (eq c ?*)
  90                   (setq encoded t)
  91                   (forward-char 1)
  92                   (setq c (char-after)))))
  93             ;; See if we have any previous continuations.
  94             (when (and prev-attribute
  95                        (not (eq prev-attribute attribute)))
  96               (push (cons prev-attribute prev-value) parameters)
  97               (setq prev-attribute nil
  98                     prev-value ""))
  99             (unless (eq c ?=)
 100               (error "Invalid header: %s" string))
 101             (forward-char 1)
 102             (setq c (char-after))
 103             (cond
 104              ((eq c ?\")
 105               (setq value
 106                     (buffer-substring (1+ (point))
 107                                       (progn (forward-sexp 1) (1- (point))))))
 108              ((and (memq c ttoken)
 109                    (not (memq c stoken)))
 110               (setq value (buffer-substring
 111                            (point) (progn (forward-sexp 1) (point)))))
 112              (t
 113               (error "Invalid header: %s" string)))
 114             (when encoded
 115               (setq value (rfc2231-decode-encoded-string value)))
 116             (if number
 117                 (setq prev-attribute attribute
 118                       prev-value (concat prev-value value))
 119               (push (cons attribute value) parameters))))
 120
 121         ;; Take care of any final continuations.
 122         (when prev-attribute
 123           (push (cons prev-attribute prev-value) parameters))
 124
 125         (when type
 126           `(,type ,@(nreverse parameters)))))))
 127
 128 (defun rfc2231-decode-encoded-string (string)
 129   "Decode an RFC2231-encoded string.
 130 These look like \"us-ascii'en-us'This%20is%20%2A%2A%2Afun%2A%2A%2A\"."
 131   (with-temp-buffer
 132     (let ((elems (split-string string "'")))
 133       ;; The encoded string may contain zero to two single-quote
 134       ;; marks.  This should give us the encoded word stripped
 135       ;; of any preceding values.
 136       (insert (car (last elems)))
 137       (goto-char (point-min))
 138       (while (search-forward "%" nil t)
 139         (insert
 140          (prog1
 141              (string-to-number (buffer-substring (point) (+ (point) 2)) 16)
 142            (delete-region (1- (point)) (+ (point) 2)))))
 143       ;; Encode using the charset, if any.
 144       (when (and (< (length elems) 1)
 145                  (not (equal (intern (car elems)) 'us-ascii)))
 146         (mm-decode-coding-region (point-min) (point-max)
 147                                  (intern (car elems))))
 148       (buffer-string))))
 149
 150 (defun rfc2231-encode-string (param value)
 151   "Return and PARAM=VALUE string encoded according to RFC2231."
 152   (let ((control (ietf-drums-token-to-list ietf-drums-no-ws-ctl-token))
 153         (tspecial (ietf-drums-token-to-list ietf-drums-tspecials))
 154         (special (ietf-drums-token-to-list "*'%\n\t"))
 155         (ascii (ietf-drums-token-to-list ietf-drums-text-token))
 156         (num -1)
 157         spacep encodep charsetp charset broken)
 158     (with-temp-buffer
 159       (insert value)
 160       (goto-char (point-min))
 161       (while (not (eobp))
 162         (cond
 163          ((or (memq (following-char) control)
 164               (memq (following-char) tspecial)
 165               (memq (following-char) special))
 166           (setq encodep t))
 167          ((eq (following-char) ? )
 168           (setq spacep t))
 169          ((not (memq (following-char) ascii))
 170           (setq charsetp t)))
 171         (forward-char 1))
 172       (when charsetp
 173         (setq charset (mm-encode-body)))
 174       (cond
 175        ((or encodep charsetp)
 176         (goto-char (point-min))
 177         (while (not (eobp))
 178           (when (> (current-column) 60)
 179             (insert "\n")
 180             (setq broken t))
 181           (if (or (not (memq (following-char) ascii))
 182                   (memq (following-char) control)
 183                   (memq (following-char) tspecial)
 184                   (memq (following-char) special)
 185                   (eq (following-char) ? ))
 186               (progn
 187                 (insert "%" (format "%02x" (following-char)))
 188                 (delete-char 1))
 189             (forward-char 1)))
 190         (goto-char (point-min))
 191         (insert (or charset "ascii") "''")
 192         (goto-char (point-min))
 193         (if (not broken)
 194             (insert param "*=")
 195           (while (not (eobp))
 196             (insert param "*" (format "%d" (incf num)) "*=")
 197             (forward-line 1))))
 198        (spacep
 199         (goto-char (point-min))
 200         (insert param "=\"")
 201         (goto-char (point-max))
 202         (insert "\""))
 203        (t
 204         (goto-char (point-min))
 205         (insert param "=")))
 206       (buffer-string))))
 207
 208 (provide 'rfc2231)
 209
 210 ;;; rfc2231.el ends here