In article <[EMAIL PROTECTED]>, Leo <[EMAIL PROTECTED]> writes:
> * Eli Zaretskii (2006-12-17 06:30 +0200) said: > ^^^^^^^^^^^^^ >>> From: Leo <[EMAIL PROTECTED]> >>> Date: Sun, 17 Dec 2006 03:15:59 +0000 >>> >>> But a file with eight-bit characters can have a correct diff >>> output. What makes ediff fail where diff succeeds? > > > > I'm not sure what you mean, but my crystal ball says that you are > > looking at the output of Diff in a terminal that supports UTF-8 > > encoded characters. If that's the case, then you will only see > > correct output from Diff with UTF-8 encoded files; other encodings > > will show gibberish. > > > > By contrast, Emacs does not support a single encoding, it supports > > many different ones. It needs to know the right encoding to display > > the characters as readable. > I mean in emacs running diff-buffer-with-file or vc-diff. The diff > output displays correctly. That's perhaps because Emacs reads the output of process while decoding by a detected coding system. That method works for your test case, but fails in a case that two files contain non-ascii characters in different encoding (e.g. UTF-8 vs GBK). --- Kenichi Handa [EMAIL PROTECTED] _______________________________________________ emacs-pretest-bug mailing list [email protected] http://lists.gnu.org/mailman/listinfo/emacs-pretest-bug
