Mechanizing around Javascript

PerlSufi has asked for the wisdom of the Perl Monks concerning the following question:

Hello monks, I need to get to a page that has a report I need to download. I have been struggling with this for a couple days and I need some help. Here is my code:
UPDATE: When I run

$mech->get('https://www.uspspostalone.com/fullservice/downloadAcsRepor
+tSearch.html');
print $mech->status();
[download]

I get a 200. Does that mean I can successfully retrieve my file located on that page? (with mechanize filling out the form stuff etc)

#! /usr/bin/perl

#needs javascript navigation capability around line 34 -date 4/17/13
use WWW::Mechanize;
use LWP::UserAgent;


#This block of code follows from the CAPS homepage to the Account Inqu
+iries link
my $mech = WWW::Mechanize->new();
my $url = "https://gateway.usps.com/bcg/login.htm";
    $mech->get($url);
if ($mech->success()){
print "Successful Connection\n";
} else {
    print "Not a successful connection\n"; }



#This block of code is intended to fill in the required forms
$mech->get($url);
my $usr = "asdf"; #marked out for post
my $pw = "asdffgg";
$mech->form_number(1);
$mech->field( "login_name", $usr);
$mech->form_number(2);
$mech->field("user_password", $pw);
$mech->click();




#follow link to dashboard
$mech->follow_link( url=> 'https://prodp1.usps.com/adminweb/view.htm?r
+equestPage=P1DASHBOARD');
$mech->follow_link( url=> 'https://www.uspspostalone.com/CIMweb/owner.
+do?group_name=Mailer&landing=Dashboard');


print $mech->content();


#my $ua = LWP::UserAgent->new();
#my $response = $ua->get('https://prodp1.usps.com/adminweb/view.htm?re
+questPage=P1DASHBOARD');
#my $response2 = $ua->get('https://www.uspspostalone.com/CIMweb/owner.
+do?group_name=Mailer&landing=Dashboard');
#my $response3 = $ua->get('https://www.uspspostalone.com/CIMweb/owner.
+do?group_name=Mailer&landing=MailingReports');
#my $response4 = $ua->get('https://www.uspspostalone.com/fullservice/d
+ashboard.html?reset=true');
#my $response5 = $ua->get ('https://www.uspspostalone.com/fullservice/
+downloadAcsReportSearch.html');

#the last link above is the page I need to end up at. I commented this
+ out for now
[download]

Inspecting the page's java script has:

<script language="javascript"> 
function expendEvent(index)
{
    if (document.getElementById("table"+ index).style.visible == 'hidd
+en')
    {
        document.getElementById("table"+ index).style.visible = 'visib
+le';
        document.getElementById("image"+ index).src="images/hide_detai
+ls.gif";
            }
    else
    {document.getElementById("table"+ index).style.visible = 'hidden';
    document.getElementById("image"+ index).src="images/show_details.g
+if";
    }

    

    if (document.getElementById("table"+ index).style.display == 'none
+')
    {document.getElementById("table"+ index).style.display = 'inline';
    
    }
    else
    {    document.getElementById("table"+ index).style.display = 'none
+';}

}
</script>
[download]

I get "link not found.." error when I run the script. Do I need to extract the tables that have the links?

Comment on Mechanizing around Javascript Select or Download Code

Replies are listed 'Best First'.
Re: Mechanizing around Javascript by runrig (Abbot) on Apr 25, 2013 at 22:26 UTC
`if ($mech->success()){` [download] The default is autocheck is ON, which means you don't have to check for success, it will die if unsuccessful. `$mech->follow_link( url=> 'https://...` [download] Why follow_link? Just get() it.	[reply] [d/l] [select]
Re: Mechanizing around Javascript by Gangabass (Vicar) on Apr 26, 2013 at 12:26 UTC
There is no form #2 on your login page. I recommend you to rewrite your code like this: `$mech->get($url); my $usr = "asdf"; #marked out for post my $pw = "asdffgg"; $mech->submit_form ( with_fields => { login_name => $usr, user_password => $pw, }, );` [download] Also there is no need to use `follow_link` with absolute URL just use `get`	[reply] [d/l] [select]
Re^2: Mechanizing around Javascript by PerlSufi (Friar) on Apr 26, 2013 at 14:21 UTC
Thanks. I have used the get() to receive my page, but I am still having problems getting te file by clicking a button in a form named "searchForm". It doesn't recognize the form id	[reply]
Re^3: Mechanizing around Javascript by Gangabass (Vicar) on Apr 26, 2013 at 14:56 UTC
Please show HTML code you have in your browser for this page (usually you can get it by pressing `Ctrl+U`).	[reply] [d/l]
Re^4: Mechanizing around Javascript by PerlSufi (Friar) on Apr 26, 2013 at 15:30 UTC
Re^5: Mechanizing around Javascript by Gangabass (Vicar) on Apr 27, 2013 at 02:13 UTC


Think about Loose Coupling
	PerlMonks