the buffer's multibyteness instead of decoding them directly in the unibyte
buffer that causes unexpected conversion in Emacs 23 (unicode).
+2006-12-22 Katsumi Yamaoka <yamaoka@jpl.org>
+
+ * nnrss.el (nnrss-fetch): Replace buffer's contents with the decoded
+ one after turning on the buffer's multibyteness instead of decoding
+ them directly in the unibyte buffer that causes unexpected conversion
+ in Emacs 23 (unicode).
+
2006-12-21 Andreas Seltenreich <uwi7@rz.uni-karlsruhe.de>
* message.el (message-generate-hashcash): Fix custom type.
(nnheader-remove-cr-followed-by-lf)
;; Decode text according to the encoding attribute.
(when (setq cs (nnrss-get-encoding))
- (mm-decode-coding-region (point-min) (point-max) cs)
- (mm-enable-multibyte))
+ (insert (prog1
+ (mm-decode-coding-string (buffer-string) cs)
+ (erase-buffer)
+ (mm-enable-multibyte))))
(goto-char (point-min))
;; Because xml-parse-region can't deal with anything that isn't