compressed file's buffer-file-coding-system

[SAITO Takuya]

compressed file's buffer-file-coding-system is always 
default-buffer-file-coding-system regardless of the file contents.

To reproduce this bug, prepare gzipped HELLO file and
start emacs -q, then type the following:

M-x auto-compression-mode
C-xC-f HELLO
M-: buffer-file-coding-system
  => iso-2022-7bit-unix (this is the correct value)
C-xC-f HELLO.gz
M-: buffer-file-coding-system
  => japanese-iso-8bit (in my environment)

If I remove "(setq last-coding-system-used coding)" from
decode-coding-inserted-region, buffer-file-coding-system is set to
correct value in above case.



In GNU Emacs 21.3.50.1 (i686-pc-linux-gnu, X toolkit, Xaw3d scroll bars)
 of 2004-07-04 on reed
configured using `configure '--prefix=/usr' '--libexecdir=/usr/lib' 
'--infodir=/usr/lib/emacs/21.3.50/info' '--with-x-toolkit=athena' 
'--without-xim''

Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: ja_JP.eucJP
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: C
  value of $LANG: ja_JP.eucJP
  locale-coding-system: japanese-iso-8bit
  default-enable-multibyte-characters: t