manoj96067

Hi,

I want to make a daemon-like application for Linux platform which supervises the website activities. I need to fetch the URLs of all websites that are opened by the user, time of the day when it is opened & the time duration for which this url is opened and by-page activity. A code snippet will be much helpful.
Alsoo, I do not want to monitor the URLs of any specific web-browser like firefox, konqueror etc, and want a generalized solution for this problem.

Any help/ code snippet is really very appreciated.

Regards,
Manoj Kumar

Salem

Even if they're using anonymous proxies?

And why do you think such an involved piece of work could be readily summed up with "a code snippet"?

__________________
If you dance barefoot on the broken glass of undefined behaviour, you've got to expect the occasional cut.

Kennedy

Get Squid and proxy all of your users through it. You _can_ proxy the local users too with squid.

brewbuck

For total browser independence, and to avoid relying on a proxy, it seems the only option is to monitor ALL outgoing TCP connections and see if the first data transmitted by the client on those connections looks like an HTTP request header.

That means you need to look for all initiated connections (SYN-ACK-ACK triples), then place the connection in a bucket to scan for HTTP activity -- once you've decided there is no HTTP activity on that connection, you can purge it from the bucket.

A "code snippet" can't possibly convey what is involved here.

__________________
"Congratulations on your purchase. To begin using your quantum computer, set the power switch to both off and on simultaneously." -- raftpeople@slashdot

MK27

If you haven't yet look at the pcap library. A customized packet sniffer could easily do this.

Programming with pcap

__________________

"A man can't just sit around." -- Larry Walters