Looking at the HTML source code doesn't show all the downloaded files, a proxy does

When pages do not download correctly, and you don't know why, the first thing you do is to look at the HTML source code of that page, but not all the downloaded files appear always clearly. If your source includes a javascript that computes the name of the downloaded file, sometimes, you will find it very hard and even impossible to know which file was supposed to be downloaded. Some websites aim to make it cryptic for preventing you to download a part of their stuff from outside of a browser.

Happily, there is a trick to know every single file you download: if you have have access to the log files of a http proxy server, it will possible for you to know exactly which files have been downloaded.

Normally, you won't have access to such a proxy server, but it is very easy to configure your own logging proxy server.
In this article, I will configure the squid proxy server under GNU/Linux but Windows users can adapt the stuff here for their case, since squid is open source software and runs also on MS platforms.

Installing squid

Installing squid under Ubuntu is as easy as opening a terminal (Applications > Accessories > Terminal) and issuing

sudo aptitude install squid

Select all Open in new window

You will have to provide your password.

Configuring squid

Now, the trick is to configure squid for running in user mode, not as root. Create a directory in your home directory. My choice was to create the directory squid inside of a my home directory. Edit a file called squid.conf in that directory by opening a terminal and typing:

mkdir ~/squid
                      cd ~/squid
                      gedit squid.conf

Select all Open in new window

In that file, you have to provide the content below:

acl all src 0.0.0.0/0.0.0.0
                      acl localhost src 127.0.0.1/32
                      http_access allow localhost 
                      http_access deny all
                      http_port 3128
                      cache_log /home/_myname_/squid/cache.log
                      cache_store_log none
                      pid_filename /home/_myname_/squid/pid
                      cache deny all
                      cache_dir null /tmp
                      logformat custom %mt %<st %ru
                      access_log /home/_myname_/squid/access.log custom

Select all Open in new window

Attention: where you see _myname_ in this configuration file, you have to provide the name of your home directory.
This will make squid listen to port 3128 of the local host and write the file name of all the files that pass through the proxy server into the log file /home/_myname_/squid/access.log. We will only log the MIME type, the size and the URL of the file.

Starting and stopping squid

For starting the proxy server, you have to issue squid -f squid.conf -N -d 1 in the directory ~/squid. For stopping, you just interrupt the proxy by hitting control-C in the same terminal where you launched squid. You can also stop it by issuing squid -k shutdown -f ~/squid/squid.conf in any other terminal.
When something goes wrong, the logfile cache.log will give you more information. In some cases, you will need to stop squid by killing it explicitly with its PID. If you don't understand how to do this, just issue:

sudo killall squid

Select all Open in new window

Enable your browser to pass through the proxy server

For logging the activity of your browser, you now need to tell your browser to use your local proxy, by setting 127.0.0.1 as IP address of the proxy and 3128 as the port.

Start surfing

Once the proxy is running and your browser point to it, you will see a lot of lines appearing into the access.log file, with the MIME type of the file, its size and its complete URL. Now you can scan that file to find the information you need. Enjoy.