I've checked the documentation and other than the fact that you've reported libwww5.76 and CPAN shows libwww-perl-5.76, I can find no discrepancy between what you have here and what the documentation says you should have.

LWP::RobotUA has a very similar example to what you have. You may want to increase your delay somewhat: you have 1 second; default is 1 minute; doco suggests 10 minutes is being nice! Regardless of this, the syntax and return value are as expected.

LWP::UserAgent states the get() method returns a HTTP::Response object: exactly what you have here.

The HTTP::Response documentation describes the methods this object can invoke. Have a read to find out about retrieving header/content/status/etc. information.

I can't comment on what would have happened last week: you haven't provided details and I wasn't there.

Last week it returned HTML so that content printed out the page source, i was relying on it doing this so i could parse it for links, but that didnt happen. Thanks for the HTTP::response refferal.I try that.
Thanks again

There's a few things you can check. First, compare your old code with your new script. Either of the following two code fragments would result in the behaviour you're reporting. (Assuming, of course, a content-type of text/html.)

# Chaining methods
$content = $ua->get($uri)->content;
print $content;

# Providing content to 'print()' via method
print $content->content;
[download]

Another possibility, albeit fairly remote, is that your reinstallation of the modules overwrote some customisations in the previous installation. (Never discount even remote possibilities as swngnmonk found in his hell which is debugging and I can atest to from personal experience.)

Lastly, I'd suggest renaming the variable $content (perhaps to $response) to avoid any confusion with the method content() and also to more accurately reflect the data it holds.

PN5

Definitely not directly related but neither yahoo nor google or alexa or ... like being spidered. They have some very serious anti DOS/spider firewalling happening. To spider sites like these is a challenge, although it can of course be done. For a start don't admit to being a robot, pretend to be IE or if you want to pretend to be a robot make yourself googlebot or one of the well known spiders that people welcome.

I presume you are aware of robots.txt and what it does.

thanks Tachyon, however since the whole point of using robotUa was to follow the rules, i could have used mechanize and pretended to be a browser, but they can block you in thier htaccess file for that. Anyway googlebot isnt written in PERl and i dont think they would buy it was google bot anyway, besides some sites let googlebot index less than other bots. Thanks for the post though, if my RobotUA delay feature doesnt stop taking minutes, i might just resort to mechanize and your idea.
Thanks

my $ua = LWP::RobotUA->new( 'myagentname', 'address@a.tld');
   $ua->delay(4);

my $response = $ua->get('http://bla/../');
print $response->content;
[download]

$Id: .signature,v .99 Sun May 12 19:48:45 2002 hsinclai Exp $
Program terminated
{7} abnormal   CONTACT YOUR SUPERVISOR
[download]