X-Git-Url: http://cgit.sxemacs.org/?p=gnus;a=blobdiff_plain;f=lisp%2Frfc2047.el;h=b2520b26af97c3dc3a619df4ac89f00c418e7030;hp=34683c3a64b7e4f47ef888f4515bde60dfd5f8b5;hb=997404c721a1de533aa9f82d4d5bbc5447bfc23d;hpb=a0c6e9c55f781bf7dacc6c1ae7f405441fcf1071 diff --git a/lisp/rfc2047.el b/lisp/rfc2047.el index 34683c3a6..b2520b26a 100644 --- a/lisp/rfc2047.el +++ b/lisp/rfc2047.el @@ -1,26 +1,23 @@ ;;; rfc2047.el --- functions for encoding and decoding rfc2047 messages -;; Copyright (C) 1998, 1999, 2000, 2002, 2003, 2004 -;; Free Software Foundation, Inc. +;; Copyright (C) 1998-2015 Free Software Foundation, Inc. ;; Author: Lars Magne Ingebrigtsen ;; MORIOKA Tomohiko ;; This file is part of GNU Emacs. -;; GNU Emacs is free software; you can redistribute it and/or modify +;; GNU Emacs is free software: you can redistribute it and/or modify ;; it under the terms of the GNU General Public License as published by -;; the Free Software Foundation; either version 2, or (at your option) -;; any later version. +;; the Free Software Foundation, either version 3 of the License, or +;; (at your option) any later version. ;; GNU Emacs is distributed in the hope that it will be useful, ;; but WITHOUT ANY WARRANTY; without even the implied warranty of -;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ;; GNU General Public License for more details. ;; 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. +;; along with GNU Emacs. If not, see . ;;; Commentary: @@ -30,21 +27,21 @@ ;;; Code: (eval-when-compile - (require 'cl) - (defvar message-posting-charset)) + (require 'cl)) +(defvar message-posting-charset) -(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) +(require 'rfc2045) ;; rfc2045-encode-string (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\ + ("\\(Resent-\\)?\\(From\\|Cc\\|To\\|Bcc\\|\\(In-\\)?Reply-To\\|Sender\ \\|Mail-Followup-To\\|Mail-Copies-To\\|Approved\\)" . address-mime) (t . mime)) "*Header/encoding method alist. @@ -77,6 +74,8 @@ The values can be: (iso-2022-jp . B) (iso-2022-kr . B) (gb2312 . B) + (gbk . B) + (gb18030 . B) (big5 . B) (cn-big5 . B) (cn-gb . B) @@ -98,6 +97,40 @@ quoted-printable and base64 respectively.") (defvar rfc2047-encode-encoded-words t "Whether encoded words should be encoded again.") +(defvar rfc2047-allow-irregular-q-encoded-words t + "*Whether to decode irregular Q-encoded words.") + +(eval-and-compile ;; Necessary to hard code them in `rfc2047-decode-region'. + (defconst rfc2047-encoded-word-regexp + "=\\?\\([^][\000-\040()<>@,\;:*\\\"/?.=]+\\)\\(?:\\*[^?]+\\)?\\?\ +\\(B\\?[+/0-9A-Za-z]*=*\ +\\|Q\\?[ ->@-~]*\ +\\)\\?=" + "Regexp that matches encoded word." + ;; The patterns for the B encoding and the Q encoding, i.e. the ones + ;; beginning with "B" and "Q" respectively, are restricted into only + ;; the characters that those encodings may generally use. + ) + (defconst rfc2047-encoded-word-regexp-loose + "=\\?\\([^][\000-\040()<>@,\;:*\\\"/?.=]+\\)\\(?:\\*[^?]+\\)?\\?\ +\\(B\\?[+/0-9A-Za-z]*=*\ +\\|Q\\?\\(?:\\?+[ -<>@-~]\\)?\\(?:[ ->@-~]+\\?+[ -<>@-~]\\)*[ ->@-~]*\\?*\ +\\)\\?=" + "Regexp that matches encoded word allowing loose Q encoding." + ;; The pattern for the Q encoding, i.e. the one beginning with "Q", + ;; is similar to: + ;; "Q\\?\\(\\?+[^\n=?]\\)?\\([^\n?]+\\?+[^\n=?]\\)*[^\n?]*\\?*" + ;; <--------1-------><----------2,3----------><--4--><-5-> + ;; They mean: + ;; 1. After "Q?", allow "?"s that follow a character other than "=". + ;; 2. Allow "=" after "Q?"; it isn't regarded as the terminator. + ;; 3. In the middle of an encoded word, allow "?"s that follow a + ;; character other than "=". + ;; 4. Allow any characters other than "?" in the middle of an + ;; encoded word. + ;; 5. At the end, allow "?"s. + )) + ;;; ;;; Functions for encoding RFC2047 messages ;;; @@ -141,6 +174,55 @@ This is either `base64' or `quoted-printable'." (re-search-forward ":[ \t\n]*" nil t) (buffer-substring-no-properties (point) (point-max))))) +(defun rfc2047-quote-special-characters-in-quoted-strings (&optional + encodable-regexp) + "Quote special characters with `\\'s in quoted strings. +Quoting will not be done in a quoted string if it contains characters +matching ENCODABLE-REGEXP or it is within parentheses." + (goto-char (point-min)) + (let ((tspecials (concat "[" ietf-drums-tspecials "]")) + (start (point)) + beg end) + (with-syntax-table (standard-syntax-table) + (while (not (eobp)) + (if (ignore-errors + (forward-list 1) + (eq (char-before) ?\))) + (forward-list -1) + (goto-char (point-max))) + (save-restriction + (narrow-to-region start (point)) + (goto-char start) + (while (search-forward "\"" nil t) + (setq beg (match-beginning 0)) + (unless (eq (char-before beg) ?\\) + (goto-char beg) + (setq beg (1+ beg)) + (condition-case nil + (progn + (forward-sexp) + (setq end (1- (point))) + (goto-char beg) + (if (and encodable-regexp + (re-search-forward encodable-regexp end t)) + (goto-char (1+ end)) + (save-restriction + (narrow-to-region beg end) + (while (re-search-forward tspecials nil 'move) + (if (eq (char-before) ?\\) + (if (looking-at tspecials) ;; Already quoted. + (forward-char) + (insert "\\")) + (goto-char (match-beginning 0)) + (insert "\\") + (forward-char)))) + (forward-char))) + (error + (goto-char beg))))) + (goto-char (point-max))) + (forward-list 1) + (setq start (point)))))) + (defvar rfc2047-encoding-type 'address-mime "The type of encoding done by `rfc2047-encode-region'. This should be dynamically bound around calls to @@ -153,86 +235,99 @@ Should be called narrowed to the head of the message." (interactive "*") (save-excursion (goto-char (point-min)) - (let (alist elem method) + (let (alist elem method charsets) (while (not (eobp)) (save-restriction (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. - (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))) - ;; We found something that may perhaps be encoded. - (setq method nil - alist rfc2047-header-encoding-alist) + (setq method nil + alist rfc2047-header-encoding-alist + charsets (mm-find-mime-charset-region (point-min) (point-max))) + ;; M$ Outlook boycotts decoding of a header if it consists + ;; of two or more encoded words and those charsets differ; + ;; it seems to decode all words in a header from a charset + ;; found first in the header. So, we unify the charsets into + ;; a single one used for encoding the whole text in a header. + (let ((mm-coding-system-priorities + (if (= (length charsets) 1) + (cons (mm-charset-to-coding-system (car charsets)) + mm-coding-system-priorities) + mm-coding-system-priorities))) (while (setq elem (pop alist)) (when (or (and (stringp (car elem)) (looking-at (car elem))) (eq (car elem) t)) (setq alist nil method (cdr elem)))) - (goto-char (point-min)) - (re-search-forward "^[^:]+: *" nil t) - (cond - ((eq method 'address-mime) - (rfc2047-encode-region (point) (point-max))) - ((eq method 'mime) - (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 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))) - ;; Hm. - (t))) - (goto-char (point-max))))))) + (if (not (rfc2047-encodable-p)) + (prog2 + (when (eq method 'address-mime) + (rfc2047-quote-special-characters-in-quoted-strings)) + (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 + (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. + (re-search-forward "^[^:]+: *" nil t) + (cond + ((eq method 'address-mime) + (rfc2047-encode-region (point) (point-max))) + ((eq method 'mime) + (let ((rfc2047-encoding-type 'mime)) + (rfc2047-encode-region (point) (point-max)))) + ((eq method 'default) + (if (and (featurep 'mule) + (if (boundp 'enable-multibyte-characters) + (default-value 'enable-multibyte-characters)) + mail-parse-charset) + (mm-encode-coding-region (point) (point-max) + mail-parse-charset))) + ;; We get this when CC'ing messages 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 behavior is wrong, tell me. I have + ;; left the old code commented out below. + ;; -- Per Abrahamsen 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 (or (and (featurep 'mule) + (if (boundp 'enable-multibyte-characters) + (default-value 'enable-multibyte-characters))) + (featurep 'file-coding)) + (mm-encode-coding-region (point) (point-max) method))) + ;; Hm. + (t))) + (goto-char (point-max)))))))) ;; Fixme: This, and the require below may not be the Right Thing, but ;; should be safe just before release. -- fx 2001-02-08 -(eval-when-compile (defvar message-posting-charset)) (defun rfc2047-encodable-p () "Return non-nil if any characters in current buffer need encoding in headers. @@ -243,7 +338,7 @@ The buffer may be narrowed." (goto-char (point-min)) (or (and rfc2047-encode-encoded-words (prog1 - (search-forward "=?" nil t) + (re-search-forward rfc2047-encoded-word-regexp nil t) (goto-char (point-min)))) (and charsets (not (equal charsets (list (car message-posting-charset)))))))) @@ -257,21 +352,17 @@ The buffer may be narrowed." (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. + ;; The following is done to work for setting all elements of the table; + ;; 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)))) + (if (featurep 'xemacs) + (put-char-table t (get-char-table ?a (standard-syntax-table)) table) + (set-char-table-range table t (aref (standard-syntax-table) ?a))) (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) @@ -282,20 +373,21 @@ The buffer may be narrowed." (modify-syntax-entry ?@ "." table) table)) -(defun rfc2047-encode-region (b e) +(defun rfc2047-encode-region (b e &optional dont-fold) "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]")) + "[^\000-\177]+\\|=\\?" + "[^\000-\177]+")) start ; start of current token - end ; end of current token + end begin csyntax ;; Whether there's an encoded word before the current token, ;; either immediately or separated by space. - last-encoded) + 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 @@ -316,9 +408,10 @@ Dynamically bind `rfc2047-encoding-type' to change that." (rfc2047-encode start (point)) (goto-char end)))) ;; `address-mime' case -- take care of quoted words, comments. + (rfc2047-quote-special-characters-in-quoted-strings encodable-regexp) (with-syntax-table rfc2047-syntax-table (goto-char (point-min)) - (condition-case nil ; in case of unbalanced quotes + (condition-case err ; in case of unbalanced quotes ;; Look for rfc2822-style: sequences of atoms, quoted ;; strings, specials, whitespace. (Specials mustn't be ;; encoded.) @@ -329,137 +422,195 @@ Dynamically bind `rfc2047-encoding-type' to change that." (cond ((not (char-after))) ; eob ;; else token start - ((eq ?\" (char-syntax (char-after))) + ((eq ?\" (setq csyntax (char-syntax (char-after)))) ;; Quoted word. (forward-sexp) (setq end (point)) ;; Does it need encoding? (goto-char start) - (skip-chars-forward "\000-\177" end) - (if (= end (point)) - (setq last-encoded nil) - ;; It needs encoding. Strip the quotes first, - ;; since encoded words can't occur in quotes. - (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 - ((eq ?. (char-syntax (char-after))) + (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-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") - (if (and (eq (char-before) ?\() - ;; Look for the end of parentheses. - (let ((string (buffer-substring (point) - (point-max))) - (default-major-mode 'fundamental-mode)) - ;; Use `standard-syntax-table'. - (with-temp-buffer - (insert "(" string) - (goto-char (point-min)) - (condition-case nil - (progn - (forward-list 1) - (setq end (- (point) 3))) - (error nil))))) - ;; Encode text as an unstructured field. - (let ((rfc2047-encoding-type 'mime)) - (rfc2047-encode-region start (+ (point) end)) - (forward-char)) - ;; Skip one ASCII word, or encode continuous words - ;; in which all those contain non-ASCII characters. - (skip-chars-forward " \t\n") - (setq end nil) - (while (not end) - (when (looking-at "[\000-\177]+") - (setq end (match-end 0)) - (if (re-search-forward "[ \t\n]\\|\\Sw" end t) - (goto-char (match-beginning 0)) - (goto-char end) - (setq end nil))) - (unless end - (setq end t) - (when (looking-at "[^\000-\177]+") - (goto-char (match-end 0)) - (while (and (looking-at "[ \t\n]+\\([^ \t\n]+\\)") - (setq end (match-end 0)) - (string-match "[^\000-\177]" - (match-string 1))) - (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 "[^\000-\177]") - (setq end nil)) - (t - (goto-char (1- (match-end 0))) - (unless (= (point) (match-beginning 0)) - (insert " ")))) - (goto-char end) - (skip-chars-forward " \t\n") - (if (and (looking-at "[^ \t\n]+") - (string-match "[^\000-\177]" - (match-string 0))) - (setq end nil) - (goto-char end))))))) - (skip-chars-backward " \t\n") - (setq end (point)) - (goto-char start) - (skip-chars-forward "\000-\177" end) - (if (= end (point)) - (setq last-encoded nil) - (rfc2047-encode start end) - (setq last-encoded t)))))) + (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 - (error "Invalid data for rfc2047 encoding: %s" - (buffer-substring b e))))))) - (rfc2047-fold-region b (point)))) + (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]+" " ")))))))) + (unless dont-fold + (rfc2047-fold-region b (point))) + (goto-char (point-max)))) -(defun rfc2047-encode-string (string) +(defun rfc2047-encode-string (string &optional dont-fold) "Encode words in STRING. By default, the string is treated as containing addresses (see `rfc2047-encoding-type')." - (with-temp-buffer + (mm-with-multibyte-buffer (insert string) - (rfc2047-encode-region (point-min) (point-max)) + (rfc2047-encode-region (point-min) (point-max) dont-fold) (buffer-string))) -(defun rfc2047-encode-1 (column string cs encoder start space &optional eword) +;; From RFC 2047: +;; 2. Syntax of encoded-words +;; [...] +;; While there is no limit to the length of a multiple-line header +;; field, each line of a header field that contains one or more +;; 'encoded-word's is limited to 76 characters. +;; +;; In `rfc2047-encode-parameter' it is bound to nil, so don't defconst it. +(defvar rfc2047-encode-max-chars 76 + "Maximum characters of each header line that contain encoded-words. +According to RFC 2047, it is 76. If it is nil, encoded-words +will not be folded. Too small value may cause an error. You +should not change this value.") + +(defun rfc2047-encode-1 (column string cs encoder start crest tail + &optional eword) "Subroutine used by `rfc2047-encode'." (cond ((string-equal string "") (or eword "")) - ((>= column 76) - (when (and eword - (string-match "\n[ \t]+\\'" eword)) - ;; Reomove a superfluous empty line. - (setq eword (substring eword 0 (match-beginning 0)))) - (rfc2047-encode-1 (length space) string cs encoder start " " - (concat eword "\n" space))) + ((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) + ;; Remove 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) + next len) (while (and prev (<= index limit)) (setq next (concat start @@ -469,28 +620,51 @@ By default, the string is treated as containing addresses (see (substring string 0 (1+ index)) cs) (substring string 0 (1+ index)))) - "?=")) - (if (<= (+ column (length next)) 76) - (setq prev next - index (1+ index)) - (setq next prev - prev nil))) - (setq eword (concat eword next)) + "?=") + 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) - eword + (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. + ;; Remove a superfluous empty line. (setq eword (substring eword 0 (match-beginning 0)))) - (rfc2047-encode-1 (length space) (substring string index) - cs encoder start " " - (concat eword "\n" space))))))) + (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. -By default, the region is treated as containing addresses (see -`rfc2047-encoding-type')." +Point moves to the end of the region." (let ((mime-charset (or (mm-find-mime-charset-region b e) (list 'us-ascii))) - cs encoding space eword) + cs encoding tail crest eword) + ;; Use utf-8 as a last resort if determining charset of text fails. + (if (memq nil mime-charset) + (setq mime-charset (list 'utf-8))) (cond ((> (length mime-charset) 1) (error "Can't rfc2047-encode `%s'" (buffer-substring-no-properties b e))) @@ -511,9 +685,19 @@ By default, the region is treated as containing addresses (see '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")) - (setq space (buffer-substring-no-properties (point) b))) + ;; `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 @@ -525,13 +709,21 @@ By default, the region is treated as containing addresses (see 'identity) (concat "=?" (downcase (symbol-name mime-charset)) "?" (upcase (symbol-name encoding)) "?") - (or space " "))) + (or crest " ") + tail)) (delete-region (if (eq (aref eword 0) ?\n) - (point) + (if (bolp) + ;; The line was folded before encoding. + (1- (point)) + (point)) (goto-char b)) - e) + (+ e (length tail))) + ;; `eword' contains `crest' and `tail'. (insert eword) - (unless (or (eolp) + (set-marker b nil) + (set-marker e nil) + (unless (or (/= 0 (length tail)) + (eobp) (looking-at "[ \t\n)]")) (insert " ")))) (t @@ -602,9 +794,10 @@ By default, the region is treated as containing addresses (see (goto-char (or break qword-break)) (setq break nil qword-break nil) - (if (looking-at "[ \t]") - (insert ?\n) - (insert "\n ")) + (if (or (> 0 (skip-chars-backward " \t")) + (looking-at "[ \t]")) + (insert ?\n) + (insert "\n ")) (setq bol (1- (point))) ;; Don't break before the first non-LWSP characters. (skip-chars-forward " \t") @@ -643,6 +836,8 @@ By default, the region is treated as containing addresses (see "Base64-encode the header contained in STRING." (base64-encode-string string t)) +(autoload 'quoted-printable-encode-region "qp") + (defun rfc2047-q-encode-string (string) "Quoted-printable-encode the header in STRING." (mm-with-unibyte-buffer @@ -661,14 +856,125 @@ By default, the region is treated as containing addresses (see (subst-char-in-region (point-min) (point-max) ? ?_) (buffer-string))) +(defun rfc2047-encode-parameter (param value) + "Return and PARAM=VALUE string encoded in the RFC2047-like style. +This is a substitution for the `rfc2231-encode-string' function, that +is the standard but many mailers don't support it." + (let ((rfc2047-encoding-type 'mime) + (rfc2047-encode-max-chars nil)) + (rfc2045-encode-string param (rfc2047-encode-string value t)))) + ;;; ;;; Functions for decoding RFC2047 messages ;;; -(eval-and-compile - (defconst rfc2047-encoded-word-regexp - "=\\?\\([^][\000-\040()<>@,\;:*\\\"/?.=]+\\)\\(?:\\*[^?]+\\)?\ -\\?\\(B\\|Q\\)\\?\\([!->@-~ ]*\\)\\?=")) +(defvar rfc2047-quote-decoded-words-containing-tspecials nil + "If non-nil, quote decoded words containing special characters.") + +(defvar rfc2047-allow-incomplete-encoded-text t + "*Non-nil means allow incomplete encoded-text in successive encoded-words. +Dividing of encoded-text in the place other than character boundaries +violates RFC2047 section 5, while we have a capability to decode it. +If it is non-nil, the decoder will decode B- or Q-encoding in each +encoded-word, concatenate them, and decode it by charset. Otherwise, +the decoder will fully decode each encoded-word before concatenating +them.") + +(defun rfc2047-strip-backslashes-in-quoted-strings () + "Strip backslashes in quoted strings. `\\\"' remains." + (goto-char (point-min)) + (let (beg) + (with-syntax-table (standard-syntax-table) + (while (search-forward "\"" nil t) + (unless (eq (char-before) ?\\) + (setq beg (match-end 0)) + (goto-char (match-beginning 0)) + (condition-case nil + (progn + (forward-sexp) + (save-restriction + (narrow-to-region beg (1- (point))) + (goto-char beg) + (while (search-forward "\\" nil 'move) + (unless (memq (char-after) '(?\")) + (delete-char -1)) + (forward-char))) + (forward-char)) + (error + (goto-char beg)))))))) + +(defun rfc2047-charset-to-coding-system (charset &optional allow-override) + "Return coding-system corresponding to MIME CHARSET. +If your Emacs implementation can't decode CHARSET, return nil. + +If allow-override is given, use `mm-charset-override-alist' to +map undesired charset names to their replacement. This should +only be used for decoding, not for encoding." + (when (stringp charset) + (setq charset (intern (downcase charset)))) + (when (or (not charset) + (eq 'gnus-all mail-parse-ignored-charsets) + (memq 'gnus-all mail-parse-ignored-charsets) + (memq charset mail-parse-ignored-charsets)) + (setq charset mail-parse-charset)) + (let ((cs (mm-charset-to-coding-system charset nil allow-override))) + (cond ((eq cs 'ascii) + (setq cs (or (mm-charset-to-coding-system mail-parse-charset) + 'raw-text))) + ((mm-coding-system-p cs)) + ((and charset + (listp mail-parse-ignored-charsets) + (memq 'gnus-unknown mail-parse-ignored-charsets)) + (setq cs (mm-charset-to-coding-system mail-parse-charset)))) + (if (eq cs 'ascii) + 'raw-text + cs))) + +(autoload 'quoted-printable-decode-string "qp") + +(defun rfc2047-decode-encoded-words (words) + "Decode successive encoded-words in WORDS and return a decoded string. +Each element of WORDS looks like (CHARSET ENCODING ENCODED-TEXT +ENCODED-WORD)." + (let (word charset cs encoding text rest) + (while words + (setq word (pop words)) + (if (and (setq cs (rfc2047-charset-to-coding-system + (setq charset (car word)) t)) + (condition-case code + (cond ((char-equal ?B (nth 1 word)) + (setq text (base64-decode-string + (rfc2047-pad-base64 (nth 2 word))))) + ((char-equal ?Q (nth 1 word)) + (setq text (quoted-printable-decode-string + (mm-subst-char-in-string + ?_ ? (nth 2 word) t))))) + (error + (message "%s" (error-message-string code)) + nil))) + (if (and rfc2047-allow-incomplete-encoded-text + (eq cs (caar rest))) + ;; Concatenate text of which the charset is the same. + (setcdr (car rest) (concat (cdar rest) text)) + (push (cons cs text) rest)) + ;; Don't decode encoded-word. + (push (cons nil (nth 3 word)) rest))) + (while rest + (setq words (concat + (or (and (setq cs (caar rest)) + (condition-case code + (mm-decode-coding-string (cdar rest) cs) + (error + (message "%s" (error-message-string code)) + nil))) + (concat (when (cdr rest) " ") + (cdar rest) + (when (and words + (not (eq (string-to-char words) ? ))) + " "))) + words) + rest (cdr rest))) + words)) ;; Fixme: This should decode in place, not cons intermediate strings. ;; Also check whether it needs to worry about delimiting fields like @@ -680,38 +986,102 @@ By default, the region is treated as containing addresses (see ;; 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." +(defun rfc2047-decode-region (start end &optional address-mime) + "Decode MIME-encoded words in region between START and END. +If ADDRESS-MIME is non-nil, strip backslashes which precede characters +other than `\"' and `\\' in quoted strings." (interactive "r") (let ((case-fold-search t) - b e) + (eword-regexp + (if rfc2047-allow-irregular-q-encoded-words + (eval-when-compile + (concat "[\n\t ]*\\(" rfc2047-encoded-word-regexp-loose "\\)")) + (eval-when-compile + (concat "[\n\t ]*\\(" rfc2047-encoded-word-regexp "\\)")))) + b e match words) (save-excursion (save-restriction (narrow-to-region start end) - (goto-char (point-min)) - ;; Remove whitespace between encoded words. - (while (re-search-forward - (eval-when-compile - (concat "\\(" 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. + (when address-mime + (rfc2047-strip-backslashes-in-quoted-strings)) + (goto-char (setq b start)) + ;; Look for the encoded-words. + (while (setq match (re-search-forward eword-regexp nil t)) + (setq e (match-beginning 1) + end (match-end 0) + words nil) + (while match + (push (list (match-string 2) ;; charset + (char-after (match-beginning 3)) ;; encoding + (substring (match-string 3) 2) ;; encoded-text + (match-string 1)) ;; encoded-word + words) + ;; Look for the subsequent encoded-words. + (when (setq match (looking-at eword-regexp)) + (goto-char (setq end (match-end 0))))) + ;; Replace the encoded-words with the decoded one. + (delete-region e end) + (insert (rfc2047-decode-encoded-words (nreverse words))) (save-restriction (narrow-to-region e (point)) (goto-char e) + ;; Remove newlines between decoded words, though such + ;; things essentially must not be there. (while (re-search-forward "[\n\r]+" nil t) (replace-match " ")) + (setq end (point-max)) + ;; Quote decoded words if there are special characters + ;; which might violate RFC2822. + (when (and rfc2047-quote-decoded-words-containing-tspecials + (let ((regexp (car (rassq + 'address-mime + rfc2047-header-encoding-alist)))) + (when regexp + (save-restriction + (widen) + (and + ;; Don't quote words if already quoted. + (not (and (eq (char-before e) ?\") + (eq (char-after end) ?\"))) + (progn + (beginning-of-line) + (while (and (memq (char-after) '(? ?\t)) + (zerop (forward-line -1)))) + (looking-at regexp))))))) + (let (quoted) + (goto-char e) + (skip-chars-forward " \t") + (setq start (point)) + (setq quoted (eq (char-after) ?\")) + (goto-char (point-max)) + (skip-chars-backward " \t" start) + (if (setq quoted (and quoted + (> (point) (1+ start)) + (eq (char-before) ?\"))) + (progn + (backward-char) + (setq start (1+ start) + end (point-marker))) + (setq end (point-marker))) + (goto-char start) + (while (search-forward "\"" end t) + (when (prog2 + (backward-char) + (zerop (% (skip-chars-backward "\\\\") 2)) + (goto-char (match-beginning 0))) + (insert "\\")) + (forward-char)) + (when (and (not quoted) + (progn + (goto-char start) + (re-search-forward + (concat "[" ietf-drums-tspecials "]") + end t))) + (goto-char start) + (insert "\"") + (goto-char end) + (insert "\"")) + (set-marker end nil))) (goto-char (point-max))) (when (and (mm-multibyte-p) mail-parse-charset @@ -725,25 +1095,41 @@ By default, the region is treated as containing addresses (see (not (eq mail-parse-charset 'gnus-decoded))) (mm-decode-coding-region b (point-max) mail-parse-charset)))))) -(defun rfc2047-decode-string (string) - "Decode the quoted-printable-encoded STRING and return the results." - (let ((m (mm-multibyte-p))) +(defun rfc2047-decode-address-region (start end) + "Decode MIME-encoded words in region between START and END. +Backslashes which precede characters other than `\"' and `\\' in quoted +strings are stripped." + (rfc2047-decode-region start end t)) + +(defun rfc2047-decode-string (string &optional address-mime) + "Decode MIME-encoded STRING and return the result. +If ADDRESS-MIME is non-nil, strip backslashes which precede characters +other than `\"' and `\\' in quoted strings." + ;; (let ((m (mm-multibyte-p))) (if (string-match "=\\?" string) (with-temp-buffer - ;; Fixme: This logic is wrong, but seems to be required by - ;; Gnus summary buffer generation. The value of `m' depends - ;; on the current buffer, not global multibyteness or that - ;; of the string. Also the string returned should always be - ;; multibyte in a multibyte session, i.e. the buffer should - ;; be multibyte before `buffer-string' is called. - (when m - (mm-enable-multibyte)) + ;; We used to only call mm-enable-multibyte if `m' is non-nil, + ;; but this can't be the right criterion. Don't just revert this + ;; change if it encounters a bug. Please help me fix it + ;; right instead. --Stef + ;; The string returned should always be multibyte in a multibyte + ;; session, i.e. the buffer should be multibyte before + ;; `buffer-string' is called. + (mm-enable-multibyte) (insert string) (inline - (rfc2047-decode-region (point-min) (point-max))) + (rfc2047-decode-region (point-min) (point-max) address-mime)) (buffer-string)) + (when address-mime + (setq string + (with-temp-buffer + (when (mm-multibyte-string-p string) + (mm-enable-multibyte)) + (insert string) + (rfc2047-strip-backslashes-in-quoted-strings) + (buffer-string)))) ;; Fixme: As above, `m' here is inappropriate. - (if (and m + (if (and ;; m mail-parse-charset (not (eq mail-parse-charset 'us-ascii)) (not (eq mail-parse-charset 'gnus-decoded))) @@ -759,24 +1145,15 @@ By default, the region is treated as containing addresses (see (if (and (fboundp 'detect-coding-string) ;; string is purely ASCII (eq (detect-coding-string string t) 'undecided)) - string - (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 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) - (string-to-char (match-string 2 word)) - (match-string 3 word)) - (error word)) - word))) ; un-decodable + string + (mm-decode-coding-string string mail-parse-charset)) + (mm-string-to-multibyte string)))) ;; ) + +(defun rfc2047-decode-address-string (string) + "Decode MIME-encoded STRING and return the result. +Backslashes which precede characters other than `\"' and `\\' in quoted +strings are stripped." + (rfc2047-decode-string string t)) (defun rfc2047-pad-base64 (string) "Pad STRING to quartets." @@ -793,38 +1170,6 @@ decodable." (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 the characters \"B\" and \"Q\". -If your Emacs implementation can't decode CHARSET, return nil." - (if (stringp charset) - (setq charset (intern (downcase charset)))) - (if (or (not charset) - (eq 'gnus-all mail-parse-ignored-charsets) - (memq 'gnus-all mail-parse-ignored-charsets) - (memq charset mail-parse-ignored-charsets)) - (setq charset mail-parse-charset)) - (let ((cs (mm-charset-to-coding-system charset))) - (if (and (not cs) charset - (listp mail-parse-ignored-charsets) - (memq 'gnus-unknown mail-parse-ignored-charsets)) - (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)) - (mm-decode-coding-string - (cond - ((char-equal ?B encoding) - (base64-decode-string - (rfc2047-pad-base64 string))) - ((char-equal ?Q encoding) - (quoted-printable-decode-string - (mm-subst-char-in-string ?_ ? string t))) - (t (error "Invalid encoding: %c" encoding))) - cs)))) - (provide 'rfc2047) -;;; arch-tag: a07fe3d4-22b5-4c4a-bd89-b1f82d5d36f6 ;;; rfc2047.el ends here