emacs: lisp/gnus/rfc2047.el comparison

comparison lisp/gnus/rfc2047.el @ 89971:cce1c0ee76ee

Revision: miles@gnu.org--gnu-2004/emacs--unicode--0--patch-36 Merge from emacs--cvs-trunk--0, emacs--gnus--5.10, gnus--rel--5.10 Patches applied: * miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523 Merge from emacs--gnus--5.10, gnus--rel--5.10 * miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-524 - miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-534 Update from CVS * miles@gnu.org--gnu-2004/emacs--gnus--5.10--base-0 tag of miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-464 * miles@gnu.org--gnu-2004/emacs--gnus--5.10--patch-1 Import from CVS branch gnus-5_10-branch * miles@gnu.org--gnu-2004/emacs--gnus--5.10--patch-2 Merge from lorentey@elte.hu--2004/emacs--multi-tty--0, emacs--cvs-trunk--0 * miles@gnu.org--gnu-2004/emacs--gnus--5.10--patch-3 Merge from gnus--rel--5.10 * miles@gnu.org--gnu-2004/emacs--gnus--5.10--patch-4 Merge from gnus--rel--5.10 * miles@gnu.org--gnu-2004/gnus--rel--5.10--patch-18 Update from CVS * miles@gnu.org--gnu-2004/gnus--rel--5.10--patch-19 Remove autoconf-generated files from archive * miles@gnu.org--gnu-2004/gnus--rel--5.10--patch-20 Update from CVS

author	Miles Bader <miles@gnu.org>
date	Thu, 09 Sep 2004 09:36:36 +0000
parents	4c90ffeb71c5 55fd4f77387a
children	e23928ac5a97

comparison

equal deleted inserted replaced

-:a849e5779b8c
+:cce1c0ee76ee
 ;;; rfc2047.el --- functions for encoding and decoding rfc2047 messages
-;; Copyright (C) 1998,1999,2000,02,03,2004  Free Software Foundation, Inc.
+;; Copyright (C) 1998, 1999, 2000, 2002, 2003 Free Software Foundation, Inc.
 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
 ;;	MORIOKA Tomohiko <morioka@jaist.ac.jp>
 ;; This file is part of GNU Emacs.
 ;;; Code:
 (eval-when-compile
 (require 'cl)
-(defvar message-posting-charset))
+(defvar message-posting-charset)
+(unless (fboundp 'with-syntax-table)	; not in Emacs 20
+(defmacro with-syntax-table (table &rest body)
+"Evaluate BODY with syntax table of current buffer set to TABLE.
+The syntax table of the current buffer is saved, BODY is evaluated, and the
+saved table is restored, even in case of an abnormal exit.
+Value is what BODY returns."
+(let ((old-table (make-symbol "table"))
+	    (old-buffer (make-symbol "buffer")))
+	`(let ((,old-table (syntax-table))
+	       (,old-buffer (current-buffer)))
+	   (unwind-protect
+	       (progn
+		 (set-syntax-table ,table)
+		 ,@body)
+	     (save-current-buffer
+	       (set-buffer ,old-buffer)
+	       (set-syntax-table ,old-table))))))))
 (require 'qp)
 (require 'mm-util)
 ;; Fixme: Avoid this (used for mail-parse-charset) mm dependence on gnus.
 (require 'mail-prsvr)
 (require 'base64)
 (autoload 'mm-body-7-or-8 "mm-bodies")
+(eval-and-compile
+;; Avoid gnus-util for mm- code.
+(defalias 'rfc2047-point-at-bol
+(if (fboundp 'point-at-bol)
+	'point-at-bol
+'line-beginning-position))
+(defalias 'rfc2047-point-at-eol
+(if (fboundp 'point-at-eol)
+	'point-at-eol
+'line-end-position)))
 (defvar rfc2047-header-encoding-alist
-'(("Newsgroups\\|Followup-To" . nil)
+'(("Newsgroups" . nil)
+("Followup-To" . nil)
 ("Message-ID" . nil)
-("\\(Resent-\\)?\\(From\\|Cc\\|To\\|Bcc\\|Reply-To\\|Sender\\)" .
+("\\(Resent-\\)?\\(From\\|Cc\\|To\\|Bcc\\|Reply-To\\|Sender\
-address-mime)
+\\|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
 header regexps or t.
 (cn-big5 . B)
 (cn-gb . B)
 (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))
 "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.")
 ;;;
 ;;; Functions for encoding RFC2047 messages
 ;;;
 (narrow-to-region
 (point)
 (progn
 (forward-line 1)
 (if (re-search-forward "^[^ \n\t]" nil t)
-	 (progn
+	 (rfc2047-point-at-bol)
-	   (beginning-of-line)
-	   (point))
 (point-max))))
 (goto-char (point-min)))
+(defun rfc2047-field-value ()
+"Return the value of the field at point."
+(save-excursion
+(save-restriction
+(rfc2047-narrow-to-field)
+(re-search-forward ":[ \t\n]*" nil t)
+(buffer-substring (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
 	    (re-search-forward "^[^:]+: *" nil t)
 	    (cond
 	     ((eq method 'address-mime)
 	      (rfc2047-encode-region (point) (point-max)))
 	     ((eq method 'mime)
-	      (let ((rfc2047-encoding-type method))
+	      (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) (point-max)
 					   mail-parse-charset)))
+	     ;; We get this when CC'ing messsages to newsgroups with
+	     ;; 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)
+	      (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) (point-max) method)))
 "Return non-nil if any characters in current buffer need encoding in headers.
 The buffer may be narrowed."
 (require 'message)			; for message-posting-charset
 (let ((charsets
 	 (mm-find-mime-charset-region (point-min) (point-max))))
-(and charsets (not (equal charsets (list message-posting-charset))))))
+(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
-(let ((table (make-char-table 'syntax-table '(2))))
+;; (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)
 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)
 (if (eq 'mime rfc2047-encoding-type)
-	;; Simple case -- treat as single word.
+	;; Simple case.  Treat as single word after any initial ASCII
+	;; part and before any tailing ASCII part.  The leading ASCII
+	;; is relevant for instance in Subject headers with `Re:' for
+	;; interoperability with non-MIME clients, and we might as
+	;; well avoid the tail too.
 	(progn
 	  (goto-char (point-min))
 	  ;; Does it need encoding?
-	  (skip-chars-forward "\000-\177" e)
+	  (skip-chars-forward "\000-\177")
 	  (unless (eobp)
-	    (rfc2047-encode b e)))
+	    (skip-chars-backward "^ \n") ; beginning of space-delimited word
+	    (rfc2047-encode (point) (progn
+				      (goto-char e)
+				      (skip-chars-backward "\000-\177")
+				      (skip-chars-forward "^ \n")
+				      ;; end of space-delimited word
+				      (point)))))
 ;; `address-mime' case -- take care of quoted words, comments.
 (with-syntax-table rfc2047-syntax-table
-	(let ((start (point))		; start of current token
+	(let ((start)			; start of current token
 	      end			; end of current token
 	      ;; Whether there's an encoded word before the current
 	      ;; token, either immediately or separated by space.
 	      last-encoded)
 	  (goto-char (point-min))
-	  (condition-case nil	      ; in case of unbalanced quotes
+	  (condition-case nil		; in case of unbalanced quotes
 	      ;; Look for rfc2822-style: sequences of atoms, quoted
 	      ;; strings, specials, whitespace.  (Specials mustn't be
 	      ;; encoded.)
 	      (while (not (eobp))
 		(setq start (point))
 		      (insert ? )
 		      (setq start (point)
 			    end (1+ end)))
 		    (rfc2047-encode start end)
 		    (setq last-encoded t)))))
-	    (error (error "Invalid data for rfc2047 encoding: %s"
+	    (error
-			  (buffer-substring b e)))))))
+	     (error "Invalid data for rfc2047 encoding: %s"
+		    (buffer-substring b e)))))))
 (rfc2047-fold-region b (point))))
 (defun rfc2047-encode-string (string)
 "Encode words in STRING.
 By default, the string is treated as containing addresses (see
-`rfc2047-special-chars')."
+`rfc2047-encoding-type')."
 (with-temp-buffer
 (insert string)
 (rfc2047-encode-region (point-min) (point-max))
 (buffer-string)))
 (defun rfc2047-encode (b e)
 "Encode the word(s) in the region B to E.
 By default, the region is treated as containing addresses (see
-`rfc2047-special-chars')."
+`rfc2047-encoding-type')."
 (let* ((mime-charset (mm-find-mime-charset-region b e))
 	 (cs (if (> (length mime-charset) 1)
 		 ;; Fixme: Instead of this, try to break region into
 		 ;; parts that can be encoded separately.
 		 (error "Can't rfc2047-encode `%s'"
 			(buffer-substring b e))
 	       (setq mime-charset (car mime-charset))
 	       (mm-charset-to-coding-system mime-charset)))
 	 ;; Fixme: Better, calculate the number of non-ASCII
 	 ;; characters, at least for 8-bit charsets.
-	 (encoding (if (assq mime-charset
+	 (encoding (or (cdr (assq mime-charset
-			     rfc2047-charset-encoding-alist)
-		       (cdr (assq mime-charset
 				  rfc2047-charset-encoding-alist))
-		     'B))
+		       ;; For the charsets that don't have a preferred
+		       ;; encoding, choose the one that's shorter.
+		       (save-restriction
+			 (narrow-to-region b e)
+			 (if (eq (mm-qp-or-base64) 'base64)
+			     'B
+			   'Q))))
 	 (start (concat
 		 "=?" (downcase (symbol-name mime-charset)) "?"
 		 (downcase (symbol-name encoding)) "?"))
+	 (factor (case mime-charset
+		   ((iso-8859-5 iso-8859-7 iso-8859-8 koi8-r) 1)
+		   ((big5 gb2312 euc-kr) 2)
+		   (utf-8 4)
+		   (t 8)))
+	 (pre (- b (save-restriction
+		     (widen)
+		     (rfc2047-point-at-bol))))
+	 ;; encoded-words must not be longer than 75 characters,
+	 ;; including charset, encoding etc.  This leaves us with
+	 ;; 75 - (length start) - 2 - 2 characters.  The last 2 is for
+	 ;; possible base64 padding.  In the worst case (iso-2022-*)
+	 ;; each character expands to 8 bytes which is expanded by a
+	 ;; factor of 4/3 by base64 encoding.
+	 (length (floor (- 75 (length start) 4) (* factor (/ 4.0 3.0))))
+	 ;; Limit line length to 76 characters.
+	 (length1 (max 1 (floor (- 76 (length start) 4 pre)
+				(* factor (/ 4.0 3.0)))))
 	 (first t))
 (if mime-charset
 	(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))))
+	      (if first
+		  (progn
+		    (goto-char (min (point-max) (+ length1 (point))))
+		    (setq first nil))
+		(goto-char (min (point-max) (+ length (point)))))
 	      (unless (eobp)
-		(insert ?\n))))
+		(insert ?\n)))
+	    (setq first t))
 	  (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))
 	    (insert start)
 	    (end-of-line)
 	    (insert "?=")
 	    (forward-line 1))))))
+(defun rfc2047-fold-field ()
+"Fold the current header field."
+(save-excursion
+(save-restriction
+(rfc2047-narrow-to-field)
+(rfc2047-fold-region (point-min) (point-max)))))
 (defun rfc2047-fold-region (b e)
 "Fold long lines in region B to E."
 (save-restriction
 (narrow-to-region b e)
 (goto-char (point-min))
 (let ((break nil)
 	  (qword-break nil)
 	  (first t)
 	  (bol (save-restriction
 		 (widen)
-		 (mm-point-at-bol))))
+		 (rfc2047-point-at-bol))))
 (while (not (eobp))
-	(when (and (or break qword-break) (> (- (point) bol) 76))
+	(when (and (or break qword-break)
+		   (> (- (point) bol) 76))
 	  (goto-char (or break qword-break))
 	  (setq break nil
 		qword-break nil)
 	  (if (looking-at "[ \t]")
 	      (insert ?\n)
 	    (insert "\n "))
 	  (setq bol (1- (point)))
 	  ;; Don't break before the first non-LWSP characters.
 	  (skip-chars-forward " \t")
-	  (unless (eobp) (forward-char 1)))
+	  (unless (eobp)
+	    (forward-char 1)))
 	(cond
 	 ((eq (char-after) ?\n)
 	  (forward-char 1)
 	  (setq bol (point)
 		break nil
 	 ((not break)
 	  (if (not (looking-at "=\\?[^=]"))
 	      (if (eq (char-after) ?=)
 		  (forward-char 1)
 		(skip-chars-forward "^ \t\n\r="))
-	    (setq qword-break (point))
+	    ;; Don't break at the start of the field.
+	    (unless (= (point) b)
+	      (setq qword-break (point)))
 	    (skip-chars-forward "^ \t\n\r")))
 	 (t
 	  (skip-chars-forward "^ \t\n\r"))))
-(when (and (or break qword-break) (> (- (point) bol) 76))
+(when (and (or break qword-break)
+		 (> (- (point) bol) 76))
 	(goto-char (or break qword-break))
 	(setq break nil
 	      qword-break nil)
 	  (if (looking-at "[ \t]")
 	      (insert ?\n)
 	    (insert "\n "))
 	(setq bol (1- (point)))
 	;; Don't break before the first non-LWSP characters.
 	(skip-chars-forward " \t")
-	(unless (eobp) (forward-char 1))))))
+	(unless (eobp)
+	  (forward-char 1))))))
+(defun rfc2047-unfold-field ()
+"Fold the current line."
+(save-excursion
+(save-restriction
+(rfc2047-narrow-to-field)
+(rfc2047-unfold-region (point-min) (point-max)))))
 (defun rfc2047-unfold-region (b e)
 "Unfold lines in region B to E."
 (save-restriction
 (narrow-to-region b e)
 (goto-char (point-min))
 (let ((bol (save-restriction
 		 (widen)
-		 (mm-point-at-bol)))
+		 (rfc2047-point-at-bol)))
-	  (eol (mm-point-at-eol))
+	  (eol (rfc2047-point-at-eol)))
-	  leading)
 (forward-line 1)
 (while (not (eobp))
 	(if (and (looking-at "[ \t]")
-		 (< (- (mm-point-at-eol) bol) 76))
+		 (< (- (rfc2047-point-at-eol) bol) 76))
 	    (delete-region eol (progn
 				 (goto-char eol)
 				 (skip-chars-forward "\r\n")
 				 (point)))
-	  (setq bol (mm-point-at-bol)))
+	  (setq bol (rfc2047-point-at-bol)))
-	(setq eol (mm-point-at-eol))
+	(setq eol (rfc2047-point-at-eol))
 	(forward-line 1)))))
 (defun rfc2047-b-encode-region (b e)
 "Base64-encode the header contained in region B to E."
 (save-restriction
 (defun rfc2047-q-encode-region (b e)
 "Quoted-printable-encode the header in region B to E."
 (save-excursion
 (save-restriction
 (narrow-to-region (goto-char b) e)
-(let ((alist rfc2047-q-encoding-alist)
+(let ((bol (save-restriction
-	    (bol (save-restriction
 		   (widen)
-		   (mm-point-at-bol))))
+		   (rfc2047-point-at-bol))))
-	(while alist
+	(quoted-printable-encode-region
-	  (when (looking-at (caar alist))
+	 b e nil
-	    (quoted-printable-encode-region b e nil (cdar alist))
+	 ;; = (\075), _ (\137), ? (\077) are used in the encoded word.
-	    (subst-char-in-region (point-min) (point-max) ?  ?_)
+	 ;; Avoid using 8bit characters.
-	    (setq alist nil))
+	 ;; This list excludes `especials' (see the RFC2047 syntax),
-	  (pop alist))
+	 ;; meaning that some characters in non-structured fields will
+	 ;; get encoded when they con't need to be.  The following is
+	 ;; what it used to be.
+;;;  	 ;; Equivalent to "^\000-\007\011\013\015-\037\200-\377=_?"
+;;;  	 "\010\012\014\040-\074\076\100-\136\140-\177")
+	 "-\b\n\f !#-'*+0-9A-Z\\^`-~\d")
+	(subst-char-in-region (point-min) (point-max) ?  ?_)
 	;; The size of QP encapsulation is about 20, so set limit to
 	;; 56=76-20.
 	(unless (< (- (point-max) (point-min)) 56)
 	  ;; Don't break if it could fit in one line.
 	  ;; Let rfc2047-encode-region break it later.
 	  (goto-char (1+ (point-min)))
 	  (while (and (not (bobp)) (not (eobp)))
 	    (goto-char (min (point-max) (+ 56 bol)))
 	    (search-backward "=" (- (point) 2) t)
 	    (unless (or (bobp) (eobp))
-	      (insert "\n")
+	      (insert ?\n)
 	      (setq bol (point)))))))))
 ;;;
 ;;; Functions for decoding RFC2047 messages
 ;;;
-(defvar rfc2047-encoded-word-regexp
+(eval-and-compile
-"=\\?\\([^][\000-\040()<>@,\;:\\\"/?.=]+\\)\\?\\(B\\|Q\\)\\?\\([!->@-~ +]+\\)\\?=")
+(defconst rfc2047-encoded-word-regexp
+"=\\?\\([^][\000-\040()<>@,\;:\\\"/?.=]+\\)\\?\\(B\\|Q\\)\
+\\?\\([!->@-~ +]*\\)\\?="))
+;; Fixme: This should decode in place, not cons intermediate strings.
+;; Also check whether it needs to worry about delimiting fields like
+;; encoding.
+;; In fact it's reported that (invalid) encoding of mailboxes in
+;; addr-specs is in use, so delimiting fields might help.  Probably
+;; not decoding a word which isn't properly delimited is good enough
+;; and worthwhile (is it more correct or not?), e.g. something like
+;; `=?iso-8859-1?q?foo?=@'.
 (defun rfc2047-decode-region (start end)
 "Decode MIME-encoded words in region between START and END."
 (interactive "r")
 (let ((case-fold-search t)
 (save-restriction
 	(narrow-to-region start end)
 	(goto-char (point-min))
 	;; Remove whitespace between encoded words.
 	(while (re-search-forward
-		(concat "\\(" rfc2047-encoded-word-regexp "\\)"
+		(eval-when-compile
-			"\\(\n?[ \t]\\)+"
+		  (concat "\\(" rfc2047-encoded-word-regexp "\\)"
-			"\\(" rfc2047-encoded-word-regexp "\\)")
+			  "\\(\n?[ \t]\\)+"
+			  "\\(" rfc2047-encoded-word-regexp "\\)"))
 		nil t)
 	  (delete-region (goto-char (match-end 1)) (match-beginning 6)))
 	;; Decode the encoded words.
 	(setq b (goto-char (point-min)))
 	(while (re-search-forward rfc2047-encoded-word-regexp nil t)
 	  (setq e (match-beginning 0))
 	  (insert (rfc2047-parse-and-decode
 		   (prog1
 		       (match-string 0)
 		     (delete-region (match-beginning 0) (match-end 0)))))
+	  ;; Remove newlines between decoded words, though such things
+	  ;; essentially must not be there.
+	  (save-restriction
+	    (narrow-to-region e (point))
+	    (goto-char e)
+	    (while (re-search-forward "[\n\r]+" nil t)
+	      (replace-match " "))
+	    (goto-char (point-max)))
 	  (when (and (mm-multibyte-p)
 		     mail-parse-charset
+		     (not (eq mail-parse-charset 'us-ascii))
 		     (not (eq mail-parse-charset 'gnus-decoded)))
 	    (mm-decode-coding-region b e mail-parse-charset))
 	  (setq b (point)))
 	(when (and (mm-multibyte-p)
 		   mail-parse-charset
 		   (not (eq mail-parse-charset 'us-ascii))
 		   (not (eq mail-parse-charset 'gnus-decoded)))
-	  (mm-decode-coding-region b (point-max) mail-parse-charset))
+	  (mm-decode-coding-region b (point-max) mail-parse-charset))))))
-	(rfc2047-unfold-region (point-min) (point-max))))))
 (defun rfc2047-decode-string (string)
 "Decode the quoted-printable-encoded STRING and return the results."
 (let ((m (mm-multibyte-p)))
-(with-temp-buffer
+(if (string-match "=\\?" string)
-(when m
+	(with-temp-buffer
-	(mm-enable-multibyte))
+	  ;; Fixme: This logic is wrong, but seems to be required by
-(insert string)
+	  ;; Gnus summary buffer generation.  The value of `m' depends
-(inline
+	  ;; on the current buffer, not global multibyteness or that
-	(rfc2047-decode-region (point-min) (point-max)))
+	  ;; of the string.  Also the string returned should always be
-(buffer-string))))
+	  ;; multibyte in a multibyte session, i.e. the buffer should
+	  ;; be multibyte before `buffer-string' is called.
+	  (when m
+	    (mm-enable-multibyte))
+	  (insert string)
+	  (inline
+	    (rfc2047-decode-region (point-min) (point-max)))
+	  (buffer-string))
+;; Fixme: As above, `m' here is inappropriate.
+(if (and m
+	       mail-parse-charset
+	       (not (eq mail-parse-charset 'us-ascii))
+	       (not (eq mail-parse-charset 'gnus-decoded)))
+	  (mm-decode-coding-string string mail-parse-charset)
+	(mm-string-as-multibyte string)))))
 (defun rfc2047-parse-and-decode (word)
 "Decode WORD and return it if it is an encoded word.
-Return WORD if not."
+Return WORD if it is not not an encoded word or if the charset isn't
+decodable."
 (if (not (string-match rfc2047-encoded-word-regexp word))
 word
 (or
 (condition-case nil
 	 (rfc2047-decode
 	  (match-string 1 word)
 	  (upcase (match-string 2 word))
 	  (match-string 3 word))
 (error word))
-word)))
+word)))				; un-decodable
+(defun rfc2047-pad-base64 (string)
+"Pad STRING to quartets."
+;; Be more liberal to accept buggy base64 strings. If
+;; base64-decode-string accepts buggy strings, this function could
+;; be aliased to identity.
+(case (mod (length string) 4)
+(0 string)
+(1 string) ;; Error, don't pad it.
+(2 (concat string "=="))
+(3 (concat string "="))))
 (defun rfc2047-decode (charset encoding string)
 "Decode STRING from the given MIME CHARSET in the given ENCODING.
 Valid ENCODINGs are \"B\" and \"Q\".
 If your Emacs implementation can't decode CHARSET, return nil."
 	(setq cs (mm-charset-to-coding-system mail-parse-charset)))
 (when cs
 (when (and (eq cs 'ascii)
 		 mail-parse-charset)
 	(setq cs mail-parse-charset))
-;; Ensure unibyte result in Emacs 20.
+(mm-decode-coding-string
-(let (default-enable-multibyte-characters)
+(cond
-	(with-temp-buffer
+	((equal "B" encoding)
-	  (mm-decode-coding-string
+	 (base64-decode-string
-	   (cond
+	  (rfc2047-pad-base64 string)))
-	    ((equal "B" encoding)
+	((equal "Q" encoding)
-	     (base64-decode-string string))
+	 (quoted-printable-decode-string
-	    ((equal "Q" encoding)
+	  (mm-replace-chars-in-string string ?_ ? )))
-	     (quoted-printable-decode-string
+	(t (error "Invalid encoding: %s" encoding)))
-	      (mm-replace-chars-in-string string ?_ ? )))
+cs))))
-	    (t (error "Invalid encoding: %s" encoding)))
-	   cs))))))
 (provide 'rfc2047)
 ;;; arch-tag: a07fe3d4-22b5-4c4a-bd89-b1f82d5d36f6
 ;;; rfc2047.el ends here

Mercurial > emacs

comparison lisp/gnus/rfc2047.el @ 89971:cce1c0ee76ee