a bug in encoding.c in Screen 4.0.2

Hi,

First of all, let me express my appreciation of the great software GNU
Screen.  I can't do any work without it.

I've found a bug in the Screen 4.0.2 (latest ?): Run screen in SJIS
mode, and use eucJP in the window, then a extra 0x8E will be inserted
before every character.

I described the detail of the bug in my blog:

  http://blog.gcd.org/archives/50663205.html

Unfortunately it was written in Japanese, and I'm sorry for your
inconvenience.  I hope some of developers of Screen can read Japanese.

The patch to fix the bug follows:

--- encoding.c.org	Mon Sep  8 23:25:23 2003
+++ encoding.c	Mon Aug 28 18:11:57 2006
@@ -1151,6 +1151,7 @@
 	      *statep = c;
 	      return -1;
 	    }
+	  if (!(c & 0x80)) return c;
 	  return c | (KANA << 16);
 	}
       t = c;

If it was a known bug, and already fixed, please ignore this mail.
I shall be very happy if this mail is a little help.


#9291.
http://www.gcd.org/sengoku/		Hiroaki Sengoku <[email protected]>