From: Trey Campbell <drtrey@...>
Date: 2004-02-23T00:11:19+09:00
Subject: Re: Puzzling...

Ruby Tuesday wrote:

>I've tried both using \n\n and \r\n\r\n on Perl and Ruby as well. Both
>resulted the same.
>
>When I tried it with Mozilla, the "Content-type: text/plain" works. Do you
>guys think it is the IE things? Thanks
>
>  
>
I doubt that it's an IE vs. Mozilla. This is just basic HTTP stuff. The 
RFC specifies that each message header end with a CRLF, with a "line" 
containing only a CRLF separating the headers from the message body. So 
your Content-type header, ending with two (or more) such sequences 
should work; the first to terminate the header the second to separate 
the header and body, and any subsequent ones are treated as part of the 
body.

print "Content-type: text/html\r\n\r\n"
print "Your body goes here..."

Should and, at least in my experience, does work.

The RFC also admonishes applications to be tolerant and accept slightly 
incorrectly formed documents:

"The line terminator for message-header fields is the sequence CRLF. 
However, we recommend that applications, when parsing such headers, 
recognize a single LF as a line terminator and ignore the leading CR."

(from http://www.w3.org/Protocols/rfc2616/rfc2616-sec19.html#sec19.3)

The following works for me:

#!C:/ruby/bin/ruby.exe
print "Content-type: text/html\n\n"
print  "1st line "
print  "2nd line "
print  "3rd line "

I can replace the first line with one ending in \r\n\r\n (what I had 
originally because it's correct, not just tolerated by well-behaved 
applications). I can also replace with \n\n\n (or \r\n\r\n\r\n) which 
generates an extra blank line in the document.

BTW, I'm running on WinXP SP1 with Apache 2.0.47 and Ruby 1.8.0. I 
tested with IE 6.0 and FireFox 0.8. The directory that contains the 
scripts is set up as follows in httpd.conf:

<Directory "C:/Apache2/htdocs/ruby">
    Options Indexes FollowSymLinks ExecCGI
    Order allow,deny
    Allow from all
</Directory>

Bottom line, what you posted *should* work. It seems like something else 
must be going on (that is, the problem's not in your Ruby script itself).

Cheers,
Trey