page screenshot using www::mechanize::firefox

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

I'm creating a script to take screenshot an element of web page. It can take full picture of long page using content_as_png but using element_as_png, only the content visible is captured, the scrollable part is not. Here is the script -

use WWW::Mechanize::Firefox;
use File::Spec;

my $mech = WWW::Mechanize::Firefox->new(
  bufsize => 10_000_000, # PNGs might become huge
  autoclose => false
);
$mech->get('http://layout.jquery-dev.net/demos/simple.html');

  my $png = $mech->content_as_png();
  my @links = $mech->by_id('center');
  my $png2 = $mech->element_as_png(@links);
  open my $fh, '>', 'page.png'
      or die "Couldn't save to 'page.png': $!";
  binmode $fh;
  print {$fh} $png;
  close $fh;
[download]

Is there an existing way to do it or will I have to dig deeper in code.

Comment on page screenshot using www::mechanize::firefox Download Code

Replies are listed 'Best First'.
Re: page screenshot using www::mechanize::firefox by Corion (Patriarch) on Mar 02, 2012 at 15:54 UTC
There is a bug in WWW::Mechanize::Firefox 0.59: Line 3327 is `if( isNaN( target_rect.scalex * target_rect.scaley )) {` [download] ... and should be `if( isNaN( target_rect.scalex * target_rect.scaley ) \|\| target +_rect.scalex * target_rect.scaley == 0) {` [download] Version 0.60 will fix this - it should be on CPAN within the next few hours. Update: But that bug does not "fix" your desired behaviour. There is no way to take a "complete screenshot" of that page with Javascript enabled, because that page resizes only a part of itself. If that page had an "outer" scrollbar, taking a screenshot would take a screenshot of the whole page, not only of the visible part. As it is, you will have to use the technique already mentioned by Anonymous Monk, take a screenshot of an element, and/or scroll a bit.	[reply] [d/l] [select]
Re^2: page screenshot using www::mechanize::firefox by satyashani (Initiate) on Mar 02, 2012 at 17:27 UTC
Thanks to the great monks for the quick wisdom.	[reply]
Re: page screenshot using www::mechanize::firefox by moritz (Cardinal) on Mar 02, 2012 at 15:05 UTC
If you're not sold on WWW::Mechanize::Firefox, you can use phantomjs for "headless" screenshots. here's an example script. Note that this is based on a different rending engine (Webkit, whereas firefox uses Gecko). Perl 6 - second systems done right	[reply]
Re^2: page screenshot using www::mechanize::firefox by satyashani (Initiate) on Mar 02, 2012 at 18:42 UTC
This one does the job perfectly. Now I don't need to start firefox everytime. Thanks.	[reply]
Re: page screenshot using www::mechanize::firefox by Anonymous Monk on Mar 02, 2012 at 14:51 UTC
Is there an existing way to do it or will I have to dig deeper in code. You'll have to write some javascript, see Re: Perl Mechanize issues - how to make a script running faster with less overhead, take screenshot, scroll down, take screenshot, scroll down ...	[reply]
Re^2: page screenshot using www::mechanize::firefox by satyashani (Initiate) on Mar 02, 2012 at 15:19 UTC
Good idea, working on it until I strike the right solution.	[reply]