1 ;;; rfc2231.el --- Functions for decoding rfc2231 headers
3 ;; Copyright (C) 1998, 1999, 2000, 2002, 2003, 2004,
4 ;; 2005 Free Software Foundation, Inc.
6 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
7 ;; This file is part of GNU Emacs.
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)
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.
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., 51 Franklin Street, Fifth Floor,
22 ;; Boston, MA 02110-1301, USA.
28 (eval-when-compile (require 'cl
))
31 (autoload 'mm-encode-body
"mm-bodies")
32 (autoload 'mail-header-remove-whitespace
"mail-parse")
33 (autoload 'mail-header-remove-comments
"mail-parse")
35 (defun rfc2231-get-value (ct attribute
)
36 "Return the value of ATTRIBUTE from CT."
37 (cdr (assq attribute
(cdr ct
))))
39 (defun rfc2231-parse-qp-string (string)
40 "Parse QP-encoded string using `rfc2231-parse-string'.
41 N.B. This is in violation with RFC2047, but it seem to be in common use."
42 (rfc2231-parse-string (rfc2047-decode-string string
)))
44 (defun rfc2231-parse-string (string)
45 "Parse STRING and return a list.
46 The list will be on the form
47 `(name (attribute . value) (attribute . value)...)"
49 (let ((ttoken (ietf-drums-token-to-list ietf-drums-text-token
))
50 (stoken (ietf-drums-token-to-list ietf-drums-tspecials
))
51 (ntoken (ietf-drums-token-to-list "0-9"))
53 display-name mailbox c display-string parameters
54 attribute value type subtype number encoded
55 prev-attribute prev-encoded
)
56 (ietf-drums-init (mail-header-remove-whitespace
57 (mail-header-remove-comments string
)))
58 (let ((table (copy-syntax-table ietf-drums-syntax-table
)))
59 (modify-syntax-entry ?
\' "w" table
)
60 (modify-syntax-entry ?
* " " table
)
61 (modify-syntax-entry ?\
; " " table)
62 (modify-syntax-entry ?
= " " table
)
63 ;; The following isn't valid, but one should be liberal
64 ;; in what one receives.
65 (modify-syntax-entry ?\
: "w" table
)
66 (set-syntax-table table
))
68 (when (and (memq c ttoken
)
69 (not (memq c stoken
)))
70 (setq type
(downcase (buffer-substring
71 (point) (progn (forward-sexp 1) (point)))))
76 (error "Invalid header: %s" string
))
78 ;; If c in nil, then this is an invalid header, but
79 ;; since elm generates invalid headers on this form,
81 (when (setq c
(char-after))
82 (if (and (memq c ttoken
)
83 (not (memq c stoken
)))
88 (point) (progn (forward-sexp 1) (point))))))
89 (error "Invalid header: %s" string
))
94 (if (not (memq c ntoken
))
100 (point) (progn (forward-sexp 1) (point)))))
101 (setq c
(char-after))
105 (setq c
(char-after)))))
106 ;; See if we have any previous continuations.
107 (when (and prev-attribute
108 (not (eq prev-attribute attribute
)))
109 (push (cons prev-attribute
111 (rfc2231-decode-encoded-string prev-value
)
114 (setq prev-attribute nil
118 (error "Invalid header: %s" string
))
120 (setq c
(char-after))
124 (buffer-substring (1+ (point))
125 (progn (forward-sexp 1) (1- (point))))))
126 ((and (or (memq c ttoken
)
127 (> c ?
\177)) ;; EXTENSION: Support non-ascii chars.
128 (not (memq c stoken
)))
129 (setq value
(buffer-substring
130 (point) (progn (forward-sexp) (point)))))
132 (error "Invalid header: %s" string
)))
134 (setq prev-attribute attribute
135 prev-value
(concat prev-value value
)
136 prev-encoded encoded
)
137 (push (cons attribute
139 (rfc2231-decode-encoded-string value
)
143 ;; Take care of any final continuations.
145 (push (cons prev-attribute
147 (rfc2231-decode-encoded-string prev-value
)
152 `(,type
,@(nreverse parameters
)))))))
154 (defun rfc2231-decode-encoded-string (string)
155 "Decode an RFC2231-encoded string.
156 These look like \"us-ascii'en-us'This%20is%20%2A%2A%2Afun%2A%2A%2A\"."
158 (let ((elems (split-string string
"'")))
159 ;; The encoded string may contain zero to two single-quote
160 ;; marks. This should give us the encoded word stripped
161 ;; of any preceding values.
162 (insert (car (last elems
)))
163 (goto-char (point-min))
164 (while (search-forward "%" nil t
)
167 (string-to-number (buffer-substring (point) (+ (point) 2)) 16)
168 (delete-region (1- (point)) (+ (point) 2)))))
169 ;; Encode using the charset, if any.
170 (when (and (mm-multibyte-p)
172 (not (equal (intern (downcase (car elems
))) 'us-ascii
)))
173 (mm-decode-coding-region (point-min) (point-max)
174 (intern (downcase (car elems
)))))
177 (defun rfc2231-encode-string (param value
)
178 "Return and PARAM=VALUE string encoded according to RFC2231."
179 (let ((control (ietf-drums-token-to-list ietf-drums-no-ws-ctl-token
))
180 (tspecial (ietf-drums-token-to-list ietf-drums-tspecials
))
181 (special (ietf-drums-token-to-list "*'%\n\t"))
182 (ascii (ietf-drums-token-to-list ietf-drums-text-token
))
184 spacep encodep charsetp charset broken
)
187 (goto-char (point-min))
190 ((or (memq (following-char) control
)
191 (memq (following-char) tspecial
)
192 (memq (following-char) special
))
194 ((eq (following-char) ?
)
196 ((not (memq (following-char) ascii
))
200 (setq charset
(mm-encode-body)))
202 ((or encodep charsetp
)
203 (goto-char (point-min))
205 (when (> (current-column) 60)
208 (if (or (not (memq (following-char) ascii
))
209 (memq (following-char) control
)
210 (memq (following-char) tspecial
)
211 (memq (following-char) special
)
212 (eq (following-char) ?
))
214 (insert "%" (format "%02x" (following-char)))
217 (goto-char (point-min))
218 (insert (symbol-name (or charset
'us-ascii
)) "''")
219 (goto-char (point-min))
223 (insert (if (>= num
0) " " "\n ")
224 param
"*" (format "%d" (incf num
)) "*=")
227 (goto-char (point-min))
229 (goto-char (point-max))
232 (goto-char (point-min))
238 ;;; arch-tag: c3ab751d-d108-406a-b301-68882ad8cd63
239 ;;; rfc2231.el ends here