I've made some tests (with telnet and ethereal) and found out that if I change the
HTTPSocket class>>httpGetDocument: args: accept: request: to not use
the port in the GET requeast (Host: www.btrl.ro instead of Host: www.btrl.ro:80)

sock sendCommand: 'GET ' , page , ' HTTP/1.0' , CrLf
					, (mimeType
							ifNil: ['']
							ifNotNil: ['ACCEPT: ' , mimeType , CrLf]) , 'ACCEPT:  
			text/html' , CrLf , HTTPBlabEmail , requestString , self
userAgentString , CrLf , 'Host: ' , serverName , CrLf.

it works ok. This is strange because I do not have this problem from my
work network, only from home. From my work network I get the content using both 
approaches (with / without port number). Could it be that at work I go through a proxy that is also 
a HTTP cache? and maybe it changes my HTTP requests in some way before passing them by ...


Thanks

Matthew S. Hamrick wrote:

> Hi Dan,
>     I don't have the bandwidth to track this down completely... But I 
> might be able to give you a few pointers. From the unix command line, 
> I did a:
>
> telnet www.btrl.ro 80
>
> and when I was connected, issued the following:
>
> GET /ro/istoric.php?vid=2 HTTP/1.0
>
> Instead of getting a response, the server closed the connection. My 
> guess is that this server really wants to see certain headers and if 
> it doesn't see them, it drops the connection as a security measure. 
> You might want to try the retreiveContentsAccept: or 
> retreiveContentsArgs: accept: methods and experiment with headers. If 
> you really want to get into it, you could use ethereal to find out 
> what headers your web browser is providing the server and simply add 
> those to your request.
>
> Sorry I don't have a lot of time to devote to this... but I thought I 
> would respond as it's similar to an issue I had earlier. The paragraph 
> above is how I solved my problem.
>
> /etc
> Matt H.
>
> On Dec 14, 2004, at 10:50 AM, Dan Corneanu wrote:
>