Re: Re: form posting question

[drake50]

It doesn't work.

I don't know what to put for the form_number and I don't know what to use for the set_fields.

I don't know how to make any headway on this because it isn't defined as a normal form.

mech-dump --forms http://www.bluffsdogs.com/recent_results.asp
<input> outside <form> at /usr/lib/perl5/site_perl/5.8.0/WWW/Mechanize.pm line 1335
<input> outside <form> at /usr/lib/perl5/site_perl/5.8.0/WWW/Mechanize.pm line 1335

[Corion]

In such cases, ugly as it is, it helps to look at the HTML on the page:

...
<td>
  <input name="radiobutton" type="radio" 
    onMouseUp="MM_goToURL('parent','agree.asp');
    return document.MM_returnValue" value="radiobutton">
   I agree
</td>
<td>
  <input name="radiobutton" type="radio"
    onMouseUp="MM_goToURL('parent','recent_resultsdisagree.asp');
    return document.MM_returnValue" value="radiobutton">
    I disagree
</td>

So, the site uses Javascript, instead of an HTML form to store the result. WWW::Mechanize does not do JavaScript, so you'll have to help it out. Most likely, the MM_goToUrl('parent', 'agree.asp') call moves you to the /agree.asp page. So you could simply try to directly use:

$agent->get('http://www.bluffsdogs.com/agree.asp');

or try that line after you've retrieved the user agreement page. WWW::Mechanize tries to make stuff easy, but some times, you still have to look at the HTML for the cause of the problems...

[drake50]

Using the url agree.asp worked. Thanks!

Just out of curiosity how would you normally deal with javascript?

[bobn]

Send the "webmaster" a nasty email?

[drake50]

The following code worked...

#!/usr/bin/perl -w
# this code worked
use strict;
use WWW::Mechanize;

my $url = 'http://www.bluffsdogs.com/agree.asp';

my $robot = new WWW::Mechanize;
$robot->get($url);
#$robot->form_number('');
#$robot->form_name('');
#$robot->set_fields('radiobutton' => 'true');
#$robot->click();

# Get the reply to my question
my $html = $robot->content();

print "$html";

...BUT...

Now I want to be able to parse the results for links. In another program I used something like:

    my $response = $browser->get($baseurl,@ns_headers);
    die "$baseurl error: ", $response->status_line
      unless $response->is_success;
    die "Weird content type at $baseurl -- ", $response->content_type
      unless $response->content_type eq 'text/html';
    my $html = $response->content;

    # save all files that match pattern given - ignore case
    while( $html =~ m/<a.{1,40}href=(.*?)>/ig ) {
      my $matchline = $1;
      #print "matchline = $matchline\n";
      if($matchline =~  m/\"(.*?)\"/ig ) {
            $matchline=$1;
            $sawcount++;
      }  # else keep what you have
      $savestr= URI->new_abs( $matchline, $response->base );
      #print "savestr = $savestr\n";
      if ($savestr =~ /$pattern/i ) {$savecount++;&storeurl($savestr,$
+trackname,$year)};
    }
    print "I saw $sawcount URLs and thought $savecount where of intere
+st.\n";

Is it possible use things like $savestr= URI->new_abs( $matchline, $response->base ); with Mechanize???

[Corion]

Does the $agent->links method, as found in the WWW::Mechanize documentation not work for you?

my @links = grep {$_->text /$pattern/i} $agent->links();

for (@links) {
  storeurl( $_->url, $trackname, $year );
};

I suggest that you read the WWW::Mechanize documentation, as it contains many helpfull hints about what is possible with WWW::Mechanize.