Subject: | I love WWW::Mechanize::Firefox |
Date: | Sun, 20 Nov 2016 17:28:03 -0500 |
To: | bug-WWW-Mechanize-FireFox [...] rt.cpan.org |
From: | Jeff I Stalzer <jeffstalzer [...] comcast.net> |
Max,
I have used WWW::Mechanize::Firefox for years and really love it. It is
an outstanding piece of code.
I use it to scrape my on-line banking every month. Every once in a while
the bank decides to change their web site (sometimes just the internals)
which causes me to re-code parts of my scraping solution. Recently they
threw me a curve. The site displays check record "header summaries" that
have to be expanded w/ a button to get detailed info downloaded from the
site. I was using:
@links = $mech3->selector('button');
to get a list of links and then one by one I did:
$mech3->highlight_node($links[$ww]) ;
$mech3->click( { dom=>$links[$ww], synchronize=>0 } );
These last two Mechanize::Firefox commands no longer work on the bank
web page (although I do get a list of buttons) and I have had to
manually "click the button" and then scrape the page.
The bank changed the "button" to:
<button class="expand-btn collapsibleControl"
aria-controls="paymentDetail_20161114085508080608"
data-billid="20161114085508080608"
data-activitypaymenttypecode="0"><span class="expand-status
visuallyhidden" id="hide-status" aria-hidden="true">Hide payment details
for TruGreen Processing Center<span class="sub-detail">TruGreen</span>
Any thoughts on a solution or work around?
I'm using WWW-Mechanize-Firefox-0.78 on Win 7 w/ FireFox 50.0
Thanks, Jeff Stalzer