Problem with HttpContentCompressor (unexpected end of file when uncompressing)

[Sylvain Lebresne]

Hi,

It seems that the HttpContentCompressor forget some footer needed by gzip.
At least on linux (tried on an ubuntu and opensuse), lauching the
netty http snoop
example (netty 3.2.1 Final with no modification whatsoever), if I
request the server
(with 'accept-encoding: gzip') and tries to gunzip the response, I get
the right output
but with an 'unexpected end of file' at the end.

Is that a bug or am I doing something wrong ?

Following is the output of a simple wget | zcat (again, against the
example shipping with netty):

$ wget --header="accept-encoding: gzip" http://localhost:8080 -O - -q | zcat
WELCOME TO THE WILD WILD WEB SERVER
===================================
VERSION: HTTP/1.0
HOSTNAME: localhost:8080
REQUEST_URI: /

HEADER: User-Agent = Wget/1.11.4
HEADER: Accept = */*
HEADER: Host = localhost:8080
HEADER: Connection = Keep-Alive
HEADER: accept-encoding = gzip


gzip: stdin: unexpected end of file


--
Sylvain