Question about WWW::Mechanize::Firefox

[ssj262]

I am a student in biology and have recently started learning perl programming. I have been trying to write a code for my project wherein i need to go to a website search for sequences related to mouse strains and save the data on resulting page in a text file.Earlier i was trying to use Mechanize but then after reading I realized it is not good with javascripts. So i am using WWW::Mechanize::Firefox.

I have added the MozRepl and Firebug addon to firefox. The code seems to run but there is no output at the terminal. I am running it on Ubuntu. I have written the code below. Kindly have a look.

#/usr/bin/perl -w

use WWW::Mechanize::Firefox;
  my $moz = WWW::Mechanize::Firefox->new();
  $moz->get($url);

$url='http://www.informatics.jax.org/javawi2/servlet/WIFetch?page=snpQ
+F';
$form_name1='queryForm';

$strain='availableStrains';
@strvalues=("15","38048","71","25292");
$variation='polymorphismType';
$varvalue='1878510';
$func_class='functionClass';
$funcvalue='1878486';
$output='format';
$outformat='tab';

$moz->get( $url );

$moz->form_name( $form_name1 );
$moz->select($strain,\@strvalues);
sleep 1;

$moz->click({ xpath => '//*[@value=">>"]' });

$moz->select($variation,$varvalue);

$moz->select($func_class,$funcvalue);

$moz->set_fields(
        geneSymname => 'chrnb2'
    );

$moz->select($output,$outformat);

sleep 1;

$moz->click_button( value => 'Search' );

($url2) = $mech -> uri;
$content=get($url2);
print $content,"\n";

exit;

I have a strong feeling that there is some problem with following part.

$moz->click({ xpath => '//*[@value=">>"]' });

The page source for the above button is:

<INPUT VALUE="&gt;&gt;" onClick="setReference(queryForm.availableStrains, queryForm.referenceStrain);" TYPE="button">

This is my first time on perlmonks and pardon me for such a long question. I am looking forward to replies from fellow monks.

[Anonymous Monk]

I have been trying to write a code for my project wherein i need to go to a website search for sequences related to mouse strains and save the data on resulting page in a text file.Earlier i was trying to use Mechanize but then after reading I realized it is not good with javascripts. So i am using WWW::Mechanize::Firefox.

Well, there is more than one INPUT TYPE=Button with a value of ">>"

A better idea would be to match against onClick attribute

or complain to whomever programmed that page to add ID's or name's to each button, they have a contact form

or use some bioperl module to access that page, because data appears available for download and http://www.bioperl.org/wiki/Gbrowse says it uses Gbrowse, surely there must be an interface for scraping Gbrowse already :)

[ssj262]

Thank you for your reply. I checked and first part of the code is working. The problem is with the code below.

$moz->select($variation,$varvalue);

$moz->select($func_class,$funcvalue);

$moz->set_fields(
        geneSymname => 'chrnb2'
    );

$moz->select($output,$outformat);

$moz->submit( $form_name1 );

For the first part I can see strains being added to the box. But for multiple select, I don't see any action happening on the webpage.

[Anonymous Monk]

What does that mean?