X-Git-Url: http://cgit.sxemacs.org/?a=blobdiff_plain;f=lisp%2Fmm-bodies.el;h=5b8874c94d956128d978939374087534f5b31a5a;hb=0651fabaac80cf08698f066dae0af33f29b91a9a;hp=3ac1aa2965307afa9b9118726ce74ff38b367bc1;hpb=27df114db202e8a922cf5e6c59a90be21701765c;p=gnus

diff --git a/lisp/mm-bodies.el b/lisp/mm-bodies.el
index 3ac1aa296..5b8874c94 100644
--- a/lisp/mm-bodies.el
+++ b/lisp/mm-bodies.el
@@ -1,5 +1,5 @@
 ;;; mm-bodies.el --- Functions for decoding MIME things
-;; Copyright (C) 1998,99 Free Software Foundation, Inc.
+;; Copyright (C) 1998, 1999, 2000 Free Software Foundation, Inc.
 
 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
 ;;	MORIOKA Tomohiko <morioka@jaist.ac.jp>
@@ -36,7 +36,20 @@
 
 ;; 8bit treatment gets any char except: 0x32 - 0x7f, CR, LF, TAB, BEL,
 ;; BS, vertical TAB, form feed, and ^_
-(defvar mm-8bit-char-regexp "[^\x20-\x7f\r\n\t\x7\x8\xb\xc\x1f]")
+(defvar mm-7bit-chars "\x20-\x7f\r\n\t\x7\x8\xb\xc\x1f")
+
+(defcustom mm-body-charset-encoding-alist
+  '((iso-2022-jp . 7bit)
+    (iso-2022-jp-2 . 7bit))
+  "Alist of MIME charsets to encodings.
+Valid encodings are `7bit', `8bit', `quoted-printable' and `base64'."
+  :type '(repeat (cons (symbol :tag "charset")
+		       (choice :tag "encoding"
+			       (const 7bit)
+			       (const 8bit)
+			       (const quoted-printable)
+			       (const base64))))
+  :group 'mime)
 
 (defun mm-encode-body ()
   "Encode a body.
@@ -45,20 +58,20 @@ If there is more than one non-ASCII MULE charset, then list of found
 MULE charsets are returned.
 If successful, the MIME charset is returned.
 If no encoding was done, nil is returned."
-  (if (not (featurep 'mule))
+  (if (not (mm-multibyte-p))
       ;; In the non-Mule case, we search for non-ASCII chars and
-      ;; return the value of `mm-default-charset' if any are found.
+      ;; return the value of `mail-parse-charset' if any are found.
       (save-excursion
 	(goto-char (point-min))
 	(if (re-search-forward "[^\x0-\x7f]" nil t)
-	    mm-default-charset
+	    (or mail-parse-charset
+		(mm-read-charset "Charset used in the article: "))
 	  ;; The logic in `mml-generate-mime-1' confirms that it's OK
 	  ;; to return nil here.
 	  nil))
     (save-excursion
       (goto-char (point-min))
-      (let ((charsets
-	     (delq 'ascii (mm-find-charset-region (point-min) (point-max))))
+      (let ((charsets (mm-find-mime-charset-region (point-min) (point-max)))
 	    charset)
 	(cond
 	 ;; No encoding.
@@ -69,38 +82,58 @@ If no encoding was done, nil is returned."
 	  charsets)
 	 ;; We encode.
 	 (t
-	  (let ((mime-charset 
-		 (mm-mime-charset (car charsets) (point-min) (point-max)))
+	  (let ((charset (car charsets))
 		start)
 	    (when (or t
 		      ;; We always decode.
 		      (not (mm-coding-system-equal
-			    mime-charset buffer-file-coding-system)))
+			    charset buffer-file-coding-system)))
 	      (while (not (eobp))
-		(if (eq (char-charset (char-after)) 'ascii)
+		(if (eq (mm-charset-after) 'ascii)
 		    (when start
 		      (save-restriction
 			(narrow-to-region start (point))
-			(mm-encode-coding-region start (point) mime-charset)
+			(mm-encode-coding-region start (point) charset)
 			(goto-char (point-max)))
 		      (setq start nil))
 		  (unless start
 		    (setq start (point))))
 		(forward-char 1))
 	      (when start
-		(mm-encode-coding-region start (point) mime-charset)
+		(mm-encode-coding-region start (point) charset)
 		(setq start nil)))
-	    mime-charset)))))))
+	    charset)))))))
+
+(defun mm-body-encoding (charset &optional encoding)
+  "Do Content-Transfer-Encoding and return the encoding of the current buffer."
+  (let ((bits (mm-body-7-or-8)))
+    (cond
+     ((and (not mm-use-ultra-safe-encoding) (eq bits '7bit))
+      bits)
+     ((and (not mm-use-ultra-safe-encoding)
+	   (or (eq t (cdr message-posting-charset))
+	       (memq charset (cdr message-posting-charset))
+	       (eq charset mail-parse-charset)))
+      bits)
+     (t
+      (let ((encoding (or encoding
+			  (cdr (assq charset mm-body-charset-encoding-alist))
+			  (mm-qp-or-base64))))
+	(when mm-use-ultra-safe-encoding
+	  (setq encoding (mm-safer-encoding encoding)))
+	(mm-encode-content-transfer-encoding encoding "text/plain")
+	encoding)))))
 
-(defun mm-body-encoding ()
-  "Return the encoding of the current buffer."
+(defun mm-body-7-or-8 ()
+  "Say whether the body is 7bit or 8bit."
   (cond
    ((not (featurep 'mule))
     (if (save-excursion
 	  (goto-char (point-min))
-	  (re-search-forward mm-8bit-char-regexp nil t))
-	'8bit
-      '7bit))
+	  (skip-chars-forward mm-7bit-chars)
+	  (eobp))
+	'7bit
+      '8bit))
    (t
     ;; Mule version
     (if (and (null (delq 'ascii
@@ -110,7 +143,7 @@ If no encoding was done, nil is returned."
 	     ;;!!!Emacs 20.3.  Sometimes.
 	     (save-excursion
 	       (goto-char (point-min))
-	       (skip-chars-forward "\0-\177")
+	       (skip-chars-forward mm-7bit-chars)
 	       (eobp)))
 	'7bit
       '8bit))))
@@ -126,12 +159,29 @@ If no encoding was done, nil is returned."
 	   ((eq encoding 'quoted-printable)
 	    (quoted-printable-decode-region (point-min) (point-max)))
 	   ((eq encoding 'base64)
-	    (base64-decode-region (point-min) (point-max)))
+	    (base64-decode-region
+	     (point-min)
+	     ;; Some mailers insert whitespace
+	     ;; junk at the end which
+	     ;; base64-decode-region dislikes.
+	     ;; Also remove possible junk which could
+	     ;; have been added by mailing list software.
+	     (save-excursion
+	       (goto-char (point-min))
+	       (while (re-search-forward "^[\t ]*\r?\n" nil t)
+		 (delete-region (match-beginning 0) (match-end 0)))
+	       (goto-char (point-max))
+	       (when (re-search-backward "^[A-Za-z0-9+/]+=*[\t ]*$" nil t)
+		 (forward-line)
+		 (delete-region (point) (point-max)))
+	       (point-max))))
 	   ((memq encoding '(7bit 8bit binary))
+	    ;; Do nothing.
 	    )
 	   ((null encoding)
+	    ;; Do nothing.
 	    )
-	   ((eq encoding 'x-uuencode)
+	   ((memq encoding '(x-uuencode x-uue))
 	    (funcall mm-uu-decode-function (point-min) (point-max)))
 	   ((eq encoding 'x-binhex)
 	    (funcall mm-uu-binhex-decode-function (point-min) (point-max)))
@@ -143,7 +193,7 @@ If no encoding was done, nil is returned."
 	 (message "Error while decoding: %s" error)
 	 nil))
     (when (and
-	   (memq encoding '(base64 x-uuencode x-binhex))
+	   (memq encoding '(base64 x-uuencode x-uue x-binhex))
 	   (equal type "text/plain"))
       (goto-char (point-min))
       (while (search-forward "\r\n" nil t)
@@ -152,33 +202,54 @@ If no encoding was done, nil is returned."
 (defun mm-decode-body (charset &optional encoding type)
   "Decode the current article that has been encoded with ENCODING.
 The characters in CHARSET should then be decoded."
-  (setq charset (or charset rfc2047-default-charset))
+  (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))
   (save-excursion
     (when encoding
       (mm-decode-content-transfer-encoding encoding type))
     (when (featurep 'mule)
-      (let (mule-charset)
-	(when (and charset
-		   (setq mule-charset (mm-charset-to-coding-system charset))
-		   ;; buffer-file-coding-system 
-					;Article buffer is nil coding system
-					;in XEmacs
-		   enable-multibyte-characters
+      (let ((mule-charset (mm-charset-to-coding-system charset)))
+	(if (and (not mule-charset)
+		 (listp mail-parse-ignored-charsets)
+		 (memq 'gnus-unknown mail-parse-ignored-charsets))
+	    (setq mule-charset 
+		  (mm-charset-to-coding-system mail-parse-charset)))
+	(when (and charset mule-charset
+		   ;; buffer-file-coding-system
+		   ;;Article buffer is nil coding system
+		   ;;in XEmacs
+		   (mm-multibyte-p)
 		   (or (not (eq mule-charset 'ascii))
-		       (setq mule-charset rfc2047-default-charset)))
+		       (setq mule-charset mail-parse-charset))
+		   (not (eq mule-charset 'gnus-decoded)))
 	  (mm-decode-coding-region (point-min) (point-max) mule-charset))))))
 
 (defun mm-decode-string (string charset)
   "Decode STRING with CHARSET."
-  (setq charset (or charset rfc2047-default-charset))
+  (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))
   (or
    (when (featurep 'mule)
-     (let (mule-charset)
-       (when (and charset
-		  (setq mule-charset (mm-charset-to-coding-system charset))
-		  enable-multibyte-characters
+     (let ((mule-charset (mm-charset-to-coding-system charset)))
+       (if (and (not mule-charset)
+		(listp mail-parse-ignored-charsets)
+		(memq 'gnus-unknown mail-parse-ignored-charsets))
+	   (setq mule-charset 
+		 (mm-charset-to-coding-system mail-parse-charset)))
+       (when (and charset mule-charset
+		  (mm-multibyte-p)
 		  (or (not (eq mule-charset 'ascii))
-		      (setq mule-charset rfc2047-default-charset)))
+		      (setq mule-charset mail-parse-charset)))
 	 (mm-decode-coding-string string mule-charset))))
    string))