-;;; rfc2047.el --- Functions for encoding and decoding rfc2047 messages
-;; Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003 Free Software Foundation, Inc.
+;;; rfc2047.el --- functions for encoding and decoding rfc2047 messages
+
+;; Copyright (C) 1998, 1999, 2000, 2002, 2003, 2004,
+;; 2005, 2006 Free Software Foundation, Inc.
;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
;; MORIOKA Tomohiko <morioka@jaist.ac.jp>
;; You should have received a copy of the GNU General Public License
;; along with GNU Emacs; see the file COPYING. If not, write to the
-;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
-;; Boston, MA 02111-1307, USA.
+;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+;; Boston, MA 02110-1301, USA.
;;; Commentary:
(require 'qp)
(require 'mm-util)
(require 'ietf-drums)
+;; Fixme: Avoid this (used for mail-parse-charset) mm dependence on gnus.
(require 'mail-prsvr)
(require 'base64)
-;; Fixme: Avoid this (for gnus-point-at-...) mm dependence on gnus.
-(require 'gnus-util)
(autoload 'mm-body-7-or-8 "mm-bodies")
(defvar rfc2047-header-encoding-alist
'(("Newsgroups" . nil)
("Followup-To" . nil)
("Message-ID" . nil)
- ("\\(Resent-\\)?\\(From\\|Cc\\|To\\|Bcc\\|Reply-To\\|Sender\\)" .
- "-A-Za-z0-9!*+/=_")
+ ("\\(Resent-\\)?\\(From\\|Cc\\|To\\|Bcc\\|\\(In-\\)?Reply-To\\|Sender\
+\\|Mail-Followup-To\\|Mail-Copies-To\\|Approved\\)" . address-mime)
(t . mime))
"*Header/encoding method alist.
The list is traversed sequentially. The keys can either be
1) nil, in which case no encoding is done;
2) `mime', in which case the header will be encoded according to RFC2047;
-3) a charset, in which case it will be encoded as that charset;
-4) `default', in which case the field will be encoded as the rest
- of the article.
-5) a string, like `mime', expect for using it as word-chars.")
+3) `address-mime', like `mime', but takes account of the rules for address
+ fields (where quoted strings and comments must be treated separately);
+4) a charset, in which case it will be encoded as that charset;
+5) `default', in which case the field will be encoded as the rest
+ of the article.")
(defvar rfc2047-charset-encoding-alist
'((us-ascii . nil)
(cn-gb-2312 . B)
(euc-kr . B)
(iso-2022-jp-2 . B)
- (iso-2022-int-1 . B))
+ (iso-2022-int-1 . B)
+ (viscii . Q))
"Alist of MIME charsets to RFC2047 encodings.
Valid encodings are nil, `Q' and `B'. These indicate binary (no) encoding,
quoted-printable and base64 respectively.")
-(defvar rfc2047-encoding-function-alist
- '((Q . rfc2047-q-encode-region)
- (B . rfc2047-b-encode-region)
- (nil . ignore))
+(defvar rfc2047-encode-function-alist
+ '((Q . rfc2047-q-encode-string)
+ (B . rfc2047-b-encode-string)
+ (nil . identity))
"Alist of RFC2047 encodings to encoding functions.")
-(defvar rfc2047-q-encoding-alist
- '(("\\(Resent-\\)?\\(From\\|Cc\\|To\\|Bcc\\|Reply-To\\|Sender\\):"
- . "-A-Za-z0-9!*+/" )
- ;; = (\075), _ (\137), ? (\077) are used in the encoded word.
- ;; Avoid using 8bit characters.
- ;; Equivalent to "^\000-\007\011\013\015-\037\200-\377=_?"
- ("." . "\010\012\014\040-\074\076\100-\136\140-\177"))
- "Alist of header regexps and valid Q characters.")
+(defvar rfc2047-encode-encoded-words t
+ "Whether encoded words should be encoded again.")
;;;
;;; Functions for encoding RFC2047 messages
;;;
+(defun rfc2047-qp-or-base64 ()
+ "Return the type with which to encode the buffer.
+This is either `base64' or `quoted-printable'."
+ (save-excursion
+ (let ((limit (min (point-max) (+ 2000 (point-min))))
+ (n8bit 0))
+ (goto-char (point-min))
+ (skip-chars-forward "\x20-\x7f\r\n\t" limit)
+ (while (< (point) limit)
+ (incf n8bit)
+ (forward-char 1)
+ (skip-chars-forward "\x20-\x7f\r\n\t" limit))
+ (if (or (< (* 6 n8bit) (- limit (point-min)))
+ ;; Don't base64, say, a short line with a single
+ ;; non-ASCII char when splitting parts by charset.
+ (= n8bit 1))
+ 'quoted-printable
+ 'base64))))
+
(defun rfc2047-narrow-to-field ()
"Narrow the buffer to the header on the current line."
(beginning-of-line)
(progn
(forward-line 1)
(if (re-search-forward "^[^ \n\t]" nil t)
- (progn
- (beginning-of-line)
- (point))
+ (point-at-bol)
(point-max))))
(goto-char (point-min)))
(save-restriction
(rfc2047-narrow-to-field)
(re-search-forward ":[ \t\n]*" nil t)
- (buffer-substring (point) (point-max)))))
+ (buffer-substring-no-properties (point) (point-max)))))
+
+(defvar rfc2047-encoding-type 'address-mime
+ "The type of encoding done by `rfc2047-encode-region'.
+This should be dynamically bound around calls to
+`rfc2047-encode-region' to either `mime' or `address-mime'. See
+`rfc2047-header-encoding-alist', for definitions.")
(defun rfc2047-encode-message-header ()
"Encode the message header according to `rfc2047-header-encoding-alist'.
(rfc2047-narrow-to-field)
(if (not (rfc2047-encodable-p))
(prog1
- (if (and (eq (mm-body-7-or-8) '8bit)
- (mm-multibyte-p)
- (mm-coding-system-p
- (car message-posting-charset)))
- ;; 8 bit must be decoded.
- ;; Is message-posting-charset a coding system?
- (mm-encode-coding-region
- (point-min) (point-max)
- (car message-posting-charset))
- nil)
+ (if (and (eq (mm-body-7-or-8) '8bit)
+ (mm-multibyte-p)
+ (mm-coding-system-p
+ (car message-posting-charset)))
+ ;; 8 bit must be decoded.
+ (mm-encode-coding-region
+ (point-min) (point-max)
+ (mm-charset-to-coding-system
+ (car message-posting-charset))))
;; No encoding necessary, but folding is nice
- (rfc2047-fold-region
- (save-excursion
- (goto-char (point-min))
- (skip-chars-forward "^:")
- (when (looking-at ": ")
- (forward-char 2))
- (point))
- (point-max)))
+ (when nil
+ (rfc2047-fold-region
+ (save-excursion
+ (goto-char (point-min))
+ (skip-chars-forward "^:")
+ (when (looking-at ": ")
+ (forward-char 2))
+ (point))
+ (point-max))))
;; We found something that may perhaps be encoded.
(setq method nil
alist rfc2047-header-encoding-alist)
(eq (car elem) t))
(setq alist nil
method (cdr elem))))
+ (re-search-forward "^[^:]+: *" nil t)
(cond
- ((stringp method)
- (rfc2047-encode-region (point-min) (point-max) method))
+ ((eq method 'address-mime)
+ (rfc2047-encode-region (point) (point-max)))
((eq method 'mime)
- (rfc2047-encode-region (point-min) (point-max)))
+ (let ((rfc2047-encoding-type 'mime))
+ (rfc2047-encode-region (point) (point-max))))
((eq method 'default)
(if (and (featurep 'mule)
(if (boundp 'default-enable-multibyte-characters)
default-enable-multibyte-characters)
mail-parse-charset)
- (mm-encode-coding-region (point-min) (point-max)
+ (mm-encode-coding-region (point) (point-max)
mail-parse-charset)))
;; We get this when CC'ing messsages to newsgroups with
- ;; 8-bit names. The group name mail copy just get
+ ;; 8-bit names. The group name mail copy just got
;; unconditionally encoded. Previously, it would ask
;; whether to encode, which was quite confusing for the
;; user. If the new behaviour is wrong, tell me. I have
;; left the old code commented out below.
;; -- Per Abrahamsen <abraham@dina.kvl.dk> Date: 2001-10-07.
+ ;; Modified by Dave Love, with the commented-out code changed
+ ;; in accordance with changes elsewhere.
((null method)
- (when (delq 'ascii
- (mm-find-charset-region (point-min) (point-max)))
- (rfc2047-encode-region (point-min) (point-max))))
-;;; ((null method)
-;;; (and (delq 'ascii
-;;; (mm-find-charset-region (point-min)
-;;; (point-max)))
-;;; (if (or (message-options-get
-;;; 'rfc2047-encode-message-header-encode-any)
-;;; (message-options-set
-;;; 'rfc2047-encode-message-header-encode-any
-;;; (y-or-n-p
-;;; "Some texts are not encoded. Encode anyway?")))
-;;; (rfc2047-encode-region (point-min) (point-max))
-;;; (error "Cannot send unencoded text"))))
+ (rfc2047-encode-region (point) (point-max)))
+;;; ((null method)
+;;; (if (or (message-options-get
+;;; 'rfc2047-encode-message-header-encode-any)
+;;; (message-options-set
+;;; 'rfc2047-encode-message-header-encode-any
+;;; (y-or-n-p
+;;; "Some texts are not encoded. Encode anyway?")))
+;;; (rfc2047-encode-region (point-min) (point-max))
+;;; (error "Cannot send unencoded text")))
((mm-coding-system-p method)
(if (and (featurep 'mule)
(if (boundp 'default-enable-multibyte-characters)
default-enable-multibyte-characters))
- (mm-encode-coding-region (point-min) (point-max) method)))
+ (mm-encode-coding-region (point) (point-max) method)))
;; Hm.
(t)))
(goto-char (point-max)))))))
The buffer may be narrowed."
(require 'message) ; for message-posting-charset
(let ((charsets
- (mapcar
- 'mm-mime-charset
- (mm-find-charset-region (point-min) (point-max))))
- (cs (list 'us-ascii (car message-posting-charset)))
- found)
- (while charsets
- (unless (memq (pop charsets) cs)
- (setq found t)))
- found))
-
-(defun rfc2047-dissect-region (b e &optional word-chars)
- "Dissect the region between B and E into words."
- (unless word-chars
- ;; Anything except most CTLs, WSP
- (setq word-chars "\010\012\014\041-\177"))
- (let (mail-parse-mule-charset
- words point current
- result word)
- (save-restriction
- (narrow-to-region b e)
- (goto-char (point-min))
- (skip-chars-forward "\000-\177")
- (while (not (eobp))
- (setq point (point))
- (skip-chars-backward word-chars b)
- (unless (eq b (point))
- (push (cons (buffer-substring b (point)) nil) words))
- (setq b (point))
- (goto-char point)
- (setq current (mm-charset-after))
- (forward-char 1)
- (skip-chars-forward word-chars)
- (while (and (not (eobp))
- (eq (mm-charset-after) current))
- (forward-char 1)
- (skip-chars-forward word-chars))
- (unless (eq b (point))
- (push (cons (buffer-substring b (point)) current) words))
- (setq b (point))
- (skip-chars-forward "\000-\177"))
- (unless (eq b (point))
- (push (cons (buffer-substring b (point)) nil) words)))
- ;; merge adjacent words
- (setq word (pop words))
- (while word
- (if (and (cdr word)
- (caar words)
- (not (cdar words))
- (not (string-match "[^ \t]" (caar words))))
- (if (eq (cdr (nth 1 words)) (cdr word))
- (progn
- (setq word (cons (concat
- (car (nth 1 words)) (caar words)
- (car word))
- (cdr word)))
- (pop words)
- (pop words))
- (push (cons (concat (caar words) (car word)) (cdr word))
- result)
- (pop words)
- (setq word (pop words)))
- (push word result)
- (setq word (pop words))))
- result))
-
-(defun rfc2047-encode-region (b e &optional word-chars)
- "Encode all encodable words in region B to E."
- (let ((words (rfc2047-dissect-region b e word-chars)) word)
- (save-restriction
- (narrow-to-region b e)
- (delete-region (point-min) (point-max))
- (while (setq word (pop words))
- (if (not (cdr word))
- (insert (car word))
- (rfc2047-fold-region (gnus-point-at-bol) (point))
- (goto-char (point-max))
- (if (> (- (point) (save-restriction
- (widen)
- (gnus-point-at-bol))) 76)
- (insert "\n "))
- ;; Insert blank between encoded words
- (if (eq (char-before) ?=) (insert " "))
- (rfc2047-encode (point)
- (progn (insert (car word)) (point))
- (cdr word))))
- (rfc2047-fold-region (point-min) (point-max)))))
+ (mm-find-mime-charset-region (point-min) (point-max))))
+ (goto-char (point-min))
+ (or (and rfc2047-encode-encoded-words
+ (prog1
+ (search-forward "=?" nil t)
+ (goto-char (point-min))))
+ (and charsets
+ (not (equal charsets (list (car message-posting-charset))))))))
+
+;; Use this syntax table when parsing into regions that may need
+;; encoding. Double quotes are string delimiters, backslash is
+;; character quoting, and all other RFC 2822 special characters are
+;; treated as punctuation so we can use forward-sexp/forward-word to
+;; skip to the end of regions appropriately. Nb. ietf-drums does
+;; things differently.
+(defconst rfc2047-syntax-table
+ ;; (make-char-table 'syntax-table '(2)) only works in Emacs.
+ (let ((table (make-syntax-table)))
+ ;; The following is done to work for setting all elements of the table
+ ;; in Emacs 21 and 22 and XEmacs; it appears to be the cleanest way.
+ ;; Play safe and don't assume the form of the word syntax entry --
+ ;; copy it from ?a.
+ (if (fboundp 'set-char-table-range) ; Emacs
+ (funcall (intern "set-char-table-range")
+ table t (aref (standard-syntax-table) ?a))
+ (if (fboundp 'put-char-table)
+ (if (fboundp 'get-char-table) ; warning avoidance
+ (put-char-table t (get-char-table ?a (standard-syntax-table))
+ table))))
+ (modify-syntax-entry ?\\ "\\" table)
+ (modify-syntax-entry ?\" "\"" table)
+ (modify-syntax-entry ?\( "(" table)
+ (modify-syntax-entry ?\) ")" table)
+ (modify-syntax-entry ?\< "." table)
+ (modify-syntax-entry ?\> "." table)
+ (modify-syntax-entry ?\[ "." table)
+ (modify-syntax-entry ?\] "." table)
+ (modify-syntax-entry ?: "." table)
+ (modify-syntax-entry ?\; "." table)
+ (modify-syntax-entry ?, "." table)
+ (modify-syntax-entry ?@ "." table)
+ table))
+
+(defun rfc2047-encode-region (b e)
+ "Encode words in region B to E that need encoding.
+By default, the region is treated as containing RFC2822 addresses.
+Dynamically bind `rfc2047-encoding-type' to change that."
+ (save-restriction
+ (narrow-to-region b e)
+ (let ((encodable-regexp (if rfc2047-encode-encoded-words
+ "[^\000-\177]+\\|=\\?"
+ "[^\000-\177]+"))
+ start ; start of current token
+ end begin csyntax
+ ;; Whether there's an encoded word before the current token,
+ ;; either immediately or separated by space.
+ last-encoded
+ (orig-text (buffer-substring-no-properties b e)))
+ (if (eq 'mime rfc2047-encoding-type)
+ ;; Simple case. Continuous words in which all those contain
+ ;; non-ASCII characters are encoded collectively. Encoding
+ ;; ASCII words, including `Re:' used in Subject headers, is
+ ;; avoided for interoperability with non-MIME clients and
+ ;; for making it easy to find keywords.
+ (progn
+ (goto-char (point-min))
+ (while (progn (skip-chars-forward " \t\n")
+ (not (eobp)))
+ (setq start (point))
+ (while (and (looking-at "[ \t\n]*\\([^ \t\n]+\\)")
+ (progn
+ (setq end (match-end 0))
+ (re-search-forward encodable-regexp end t)))
+ (goto-char end))
+ (if (> (point) start)
+ (rfc2047-encode start (point))
+ (goto-char end))))
+ ;; `address-mime' case -- take care of quoted words, comments.
+ (with-syntax-table rfc2047-syntax-table
+ (goto-char (point-min))
+ (condition-case err ; in case of unbalanced quotes
+ ;; Look for rfc2822-style: sequences of atoms, quoted
+ ;; strings, specials, whitespace. (Specials mustn't be
+ ;; encoded.)
+ (while (not (eobp))
+ ;; Skip whitespace.
+ (skip-chars-forward " \t\n")
+ (setq start (point))
+ (cond
+ ((not (char-after))) ; eob
+ ;; else token start
+ ((eq ?\" (setq csyntax (char-syntax (char-after))))
+ ;; Quoted word.
+ (forward-sexp)
+ (setq end (point))
+ ;; Does it need encoding?
+ (goto-char start)
+ (if (re-search-forward encodable-regexp end 'move)
+ ;; It needs encoding. Strip the quotes first,
+ ;; since encoded words can't occur in quotes.
+ (progn
+ (goto-char end)
+ (delete-backward-char 1)
+ (goto-char start)
+ (delete-char 1)
+ (when last-encoded
+ ;; There was a preceding quoted word. We need
+ ;; to include any separating whitespace in this
+ ;; word to avoid it getting lost.
+ (skip-chars-backward " \t")
+ ;; A space is needed between the encoded words.
+ (insert ? )
+ (setq start (point)
+ end (1+ end)))
+ ;; Adjust the end position for the deleted quotes.
+ (rfc2047-encode start (- end 2))
+ (setq last-encoded t)) ; record that it was encoded
+ (setq last-encoded nil)))
+ ((eq ?. csyntax)
+ ;; Skip other delimiters, but record that they've
+ ;; potentially separated quoted words.
+ (forward-char)
+ (setq last-encoded nil))
+ ((eq ?\) csyntax)
+ (error "Unbalanced parentheses"))
+ ((eq ?\( csyntax)
+ ;; Look for the end of parentheses.
+ (forward-list)
+ ;; Encode text as an unstructured field.
+ (let ((rfc2047-encoding-type 'mime))
+ (rfc2047-encode-region (1+ start) (1- (point))))
+ (skip-chars-forward ")"))
+ (t ; normal token/whitespace sequence
+ ;; Find the end.
+ ;; Skip one ASCII word, or encode continuous words
+ ;; in which all those contain non-ASCII characters.
+ (setq end nil)
+ (while (not (or end (eobp)))
+ (when (looking-at "[\000-\177]+")
+ (setq begin (point)
+ end (match-end 0))
+ (when (progn
+ (while (and (or (re-search-forward
+ "[ \t\n]\\|\\Sw" end 'move)
+ (setq end nil))
+ (eq ?\\ (char-syntax (char-before))))
+ ;; Skip backslash-quoted characters.
+ (forward-char))
+ end)
+ (setq end (match-beginning 0))
+ (if rfc2047-encode-encoded-words
+ (progn
+ (goto-char begin)
+ (when (search-forward "=?" end 'move)
+ (goto-char (match-beginning 0))
+ (setq end nil)))
+ (goto-char end))))
+ ;; Where the value nil of `end' means there may be
+ ;; text to have to be encoded following the point.
+ ;; Otherwise, the point reached to the end of ASCII
+ ;; words separated by whitespace or a special char.
+ (unless end
+ (when (looking-at encodable-regexp)
+ (goto-char (setq begin (match-end 0)))
+ (while (and (looking-at "[ \t\n]+\\([^ \t\n]+\\)")
+ (setq end (match-end 0))
+ (progn
+ (while (re-search-forward
+ encodable-regexp end t))
+ (< begin (point)))
+ (goto-char begin)
+ (or (not (re-search-forward "\\Sw" end t))
+ (progn
+ (goto-char (match-beginning 0))
+ nil)))
+ (goto-char end))
+ (when (looking-at "[^ \t\n]+")
+ (setq end (match-end 0))
+ (if (re-search-forward "\\Sw+" end t)
+ ;; There are special characters better
+ ;; to be encoded so that MTAs may parse
+ ;; them safely.
+ (cond ((= end (point)))
+ ((looking-at (concat "\\sw*\\("
+ encodable-regexp
+ "\\)"))
+ (setq end nil))
+ (t
+ (goto-char (1- (match-end 0)))
+ (unless (= (point) (match-beginning 0))
+ ;; Separate encodable text and
+ ;; delimiter.
+ (insert " "))))
+ (goto-char end)
+ (skip-chars-forward " \t\n")
+ (if (and (looking-at "[^ \t\n]+")
+ (string-match encodable-regexp
+ (match-string 0)))
+ (setq end nil)
+ (goto-char end)))))))
+ (skip-chars-backward " \t\n")
+ (setq end (point))
+ (goto-char start)
+ (if (re-search-forward encodable-regexp end 'move)
+ (progn
+ (unless (memq (char-before start) '(nil ?\t ? ))
+ (if (progn
+ (goto-char start)
+ (skip-chars-backward "^ \t\n")
+ (and (looking-at "\\Sw+")
+ (= (match-end 0) start)))
+ ;; Also encode bogus delimiters.
+ (setq start (point))
+ ;; Separate encodable text and delimiter.
+ (goto-char start)
+ (insert " ")
+ (setq start (1+ start)
+ end (1+ end))))
+ (rfc2047-encode start end)
+ (setq last-encoded t))
+ (setq last-encoded nil)))))
+ (error
+ (if (or debug-on-quit debug-on-error)
+ (signal (car err) (cdr err))
+ (error "Invalid data for rfc2047 encoding: %s"
+ (mm-replace-in-string orig-text "[ \t\n]+" " "))))))))
+ (rfc2047-fold-region b (point))
+ (goto-char (point-max))))
-(defun rfc2047-encode-string (string &optional word-chars)
- "Encode words in STRING."
- (with-temp-buffer
+(defun rfc2047-encode-string (string)
+ "Encode words in STRING.
+By default, the string is treated as containing addresses (see
+`rfc2047-encoding-type')."
+ (mm-with-multibyte-buffer
(insert string)
- (rfc2047-encode-region (point-min) (point-max) word-chars)
+ (rfc2047-encode-region (point-min) (point-max))
(buffer-string)))
-(defun rfc2047-encode (b e charset)
- "Encode the word in the region B to E with CHARSET."
- (let* ((mime-charset (mm-mime-charset charset))
- (cs (mm-charset-to-coding-system mime-charset))
- (encoding (or (cdr (assq mime-charset
+(defvar rfc2047-encode-max-chars 76
+ "Maximum characters of each header line that contain encoded-words.
+If it is nil, encoded-words will not be folded. Too small value may
+cause an error. Don't change this for no particular reason.")
+
+(defun rfc2047-encode-1 (column string cs encoder start crest tail
+ &optional eword)
+ "Subroutine used by `rfc2047-encode'."
+ (cond ((string-equal string "")
+ (or eword ""))
+ ((not rfc2047-encode-max-chars)
+ (concat start
+ (funcall encoder (if cs
+ (mm-encode-coding-string string cs)
+ string))
+ "?="))
+ ((>= column rfc2047-encode-max-chars)
+ (when eword
+ (cond ((string-match "\n[ \t]+\\'" eword)
+ ;; Reomove a superfluous empty line.
+ (setq eword (substring eword 0 (match-beginning 0))))
+ ((string-match "(+\\'" eword)
+ ;; Break the line before the open parenthesis.
+ (setq crest (concat crest (match-string 0 eword))
+ eword (substring eword 0 (match-beginning 0))))))
+ (rfc2047-encode-1 (length crest) string cs encoder start " " tail
+ (concat eword "\n" crest)))
+ (t
+ (let ((index 0)
+ (limit (1- (length string)))
+ (prev "")
+ next len)
+ (while (and prev
+ (<= index limit))
+ (setq next (concat start
+ (funcall encoder
+ (if cs
+ (mm-encode-coding-string
+ (substring string 0 (1+ index))
+ cs)
+ (substring string 0 (1+ index))))
+ "?=")
+ len (+ column (length next)))
+ (if (> len rfc2047-encode-max-chars)
+ (setq next prev
+ prev nil)
+ (if (or (< index limit)
+ (<= (+ len (or (string-match "\n" tail)
+ (length tail)))
+ rfc2047-encode-max-chars))
+ (setq prev next
+ index (1+ index))
+ (if (string-match "\\`)+" tail)
+ ;; Break the line after the close parenthesis.
+ (setq tail (concat (substring tail 0 (match-end 0))
+ "\n "
+ (substring tail (match-end 0)))
+ prev next
+ index (1+ index))
+ (setq next prev
+ prev nil)))))
+ (if (> index limit)
+ (concat eword next tail)
+ (if (= 0 index)
+ (if (and eword
+ (string-match "(+\\'" eword))
+ (setq crest (concat crest (match-string 0 eword))
+ eword (substring eword 0 (match-beginning 0)))
+ (setq eword (concat eword next)))
+ (setq crest " "
+ eword (concat eword next)))
+ (when (string-match "\n[ \t]+\\'" eword)
+ ;; Reomove a superfluous empty line.
+ (setq eword (substring eword 0 (match-beginning 0))))
+ (rfc2047-encode-1 (length crest) (substring string index)
+ cs encoder start " " tail
+ (concat eword "\n" crest)))))))
+
+(defun rfc2047-encode (b e)
+ "Encode the word(s) in the region B to E.
+Point moves to the end of the region."
+ (let ((mime-charset (or (mm-find-mime-charset-region b e) (list 'us-ascii)))
+ cs encoding tail crest eword)
+ (cond ((> (length mime-charset) 1)
+ (error "Can't rfc2047-encode `%s'"
+ (buffer-substring-no-properties b e)))
+ ((= (length mime-charset) 1)
+ (setq mime-charset (car mime-charset)
+ cs (mm-charset-to-coding-system mime-charset))
+ (unless (and (mm-multibyte-p)
+ (mm-coding-system-p cs))
+ (setq cs nil))
+ (save-restriction
+ (narrow-to-region b e)
+ (setq encoding
+ (or (cdr (assq mime-charset
rfc2047-charset-encoding-alist))
- 'B))
- (start (concat
- "=?" (downcase (symbol-name mime-charset)) "?"
- (downcase (symbol-name encoding)) "?"))
- (first t))
- (save-restriction
- (narrow-to-region b e)
- (when (eq encoding 'B)
- ;; break into lines before encoding
- (goto-char (point-min))
- (while (not (eobp))
- (goto-char (min (point-max) (+ 15 (point))))
- (unless (eobp)
- (insert "\n"))))
- (if (and (mm-multibyte-p)
- (mm-coding-system-p cs))
- (mm-encode-coding-region (point-min) (point-max) cs))
- (funcall (cdr (assq encoding rfc2047-encoding-function-alist))
- (point-min) (point-max))
- (goto-char (point-min))
- (while (not (eobp))
- (unless first
- (insert " "))
- (setq first nil)
- (insert start)
- (end-of-line)
- (insert "?=")
- (forward-line 1)))))
+ ;; For the charsets that don't have a preferred
+ ;; encoding, choose the one that's shorter.
+ (if (eq (rfc2047-qp-or-base64) 'base64)
+ 'B
+ 'Q)))
+ (widen)
+ (goto-char e)
+ (skip-chars-forward "^ \t\n")
+ ;; `tail' may contain a close parenthesis.
+ (setq tail (buffer-substring-no-properties e (point)))
+ (goto-char b)
+ (setq b (point-marker)
+ e (set-marker (make-marker) e))
+ (rfc2047-fold-region (point-at-bol) b)
+ (goto-char b)
+ (skip-chars-backward "^ \t\n")
+ (unless (= 0 (skip-chars-backward " \t"))
+ ;; `crest' may contain whitespace and an open parenthesis.
+ (setq crest (buffer-substring-no-properties (point) b)))
+ (setq eword (rfc2047-encode-1
+ (- b (point-at-bol))
+ (mm-replace-in-string
+