Juhana Sadeharju wrote:

> I placed "use_proxy = off" to .wgetrc (which file I did not have earlier)
> and to "~/wget/etc/wgetrc" (which file I had), and tried
>   wget --proxy=off http://www.maqamworld.com
> and it still does not work.
>
> Could there be some system wgetrc files somewhere? I have compiled
> wget on my own to my home directory, and certainly wish that my own
> installation does not use files of some other installation.
>
> Why did you think the ":80" comes from proxy? I have always thought
> it comes from the target site, not from our site. Did you try the
> given command yourself and it worked? Please try now if you did not.
>
> If wget puts the ":80" , then how do I instruct wget to not do that
> no matter what is told somewhere? What part of the source code I should
> edit if that is only what helps?
>
> Though, you should fix this to the wget source because something is
> not working now. I wonder why this "not working" is set as a default
> behaviour to wget...

In the communications world where two computers are talking to one another,
there is no such thing as http://www.maqamworld.com or
http://www.maqamworld.com:80. Those are simply convenient (and readable)
notations for the human beings that use the computers. Run wget with the -d
option and you will see how the computers break all that down:

DEBUG output created by Wget 1.9-beta1 on linux-gnu.

--08:23:18--  http://www.maqamworld.com/
           => `index.html'
Resolving www.maqamworld.com... 66.48.76.90
Caching www.maqamworld.com => 66.48.76.90
Connecting to www.maqamworld.com[66.48.76.90]:80... connected.
Created socket 3.
Releasing 0x81164d8 (new refcount 1).
---request begin---
GET / HTTP/1.0
User-Agent: Wget/1.9-beta1
Host: www.maqamworld.com
Accept: */*
Connection: Keep-Alive

wget does a domain name look up on "www.maqamworld.com" and finds that it
resides at 66.48.76.90. It then opens a socket to that IP address on port
80. (Port 80 is the default port for the HTTP protocol specified in the
first part of the URL -- it's what is used by almost all web sites and
browsers).

Now that the connection is made between your computer and the server, wget
sends a GET request (part of the HTTP protocol) to the server. Included in
that request is the name of the site being retrieved "Host:
www.maqamworld.com", but the port number is never sent by wget to the
server.

By the way, when I ran this, wget created an index.html file that looks
reasonable to me. It is 23,335 bytes long and is identical to what I get if
I do a View Source in my browser and save the text file.

Run the following command and send the output to the list if you continue to
have problems: wget http://www.maqamworld.com -d

Tony

Reply via email to