sprint

#!/usr/bin/perl
use strict;

use WWW::Mechanize;
use File::Path;

########################################################################################################################
#                Change only the configuration settings in this section, nothing above or below it.                    #
########################################################################################################################

# Credentials
my $username = "someone";
my $password = "somepassword";

# Enclose value in double quotes, folders with spaces in the name are ok.
my $root_folder = "/Users/john/Documents/Personal/Utilities/Sprint/";

# Numeric account number, change to match yours
my $account  = "874000001";

########################################################################################################################
########################################################################################################################

# Suddenly web robot.
my $mech = WWW::Mechanize->new();
$mech->agent_alias('Mac Safari');

# Base URL for PDF statements.
$mech->get("http://mysprint.sprint.com/mysprint/pages/sl/global/login.jsp");

# Login, blah.
$mech->submit_form(
  form_id => 'frmUserLoginDL',
  fields  => { USER     => $username,
               PASSWORD => $password,
             },
);

# Dumb thing uses a meta refresh...
$mech->follow_link(url_regex => qr/CollectDevicePrint\.do/);

# Now a magic bounce...
my $pm_fp = "version=1&pm_fpua=mozilla/5.0 (macintosh; intel mac os x 10_9_3) applewebkit/537.36 (khtml, like gecko) " .
            "chrome/35.0.1916.153 safari/537.36|5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, " .
            "like Gecko) Chrome/35.0.1916.153 Safari/537.36|MacIntel&pm_fpsc=24|1920|1200|1178&pm_fpsw=&pm_fptz=-6" .
            "&pm_fpln=lang=en-US|syslang=|userlang=&pm_fpjv=1&pm_fpco=1";
foreach my $form ($mech->forms()) {
    map { $_->readonly(0) } $form->inputs();
}
$mech->submit_form(
  form_name => 'LoginForm',
  fields    => { pm_fp => $pm_fp },
);

# Another meta refresh...
$mech->follow_link(url_regex => qr/ReturnToCaller\.do/);

# Another magic form bounce...
$mech->submit_form(
  form_name => 'CallbackForm',
);

# Get the initial bill page.
$mech->get("https://myaccountportal.sprint.com/servlet/ecare?inf_action=login&action=accountBill&sl=111100&selaccount=$account");

# Finally we can get to the billing history page.
$mech->get("https://myaccountportal.sprint.com/servlet/ecare?inf_action=downloadDates&isBillHist=true");
my $page = $mech->content();

# Now we need to get all PDF links. Jackasses didn't put direct links, javascript constructs them onclick. Some of them
# are just "billImage", but others are "billImageFromOlive" ... no idea of the difference.
while ($page =~ /(\/servlet\/ecare\?inf_template=\/servlet\/billImage(?:FromOlive)*\?billDate=)(\d\d)\/(\d\d)\/(\d{4})/g) {
    # Extract the date.
    my $year = $4;
    my $date = "$year-$3-$2";
    my $link = "$1$2/$3/$year";

    # This will create any nested directories necessary. Mostly for the year.
    File::Path::make_path("$root_folder$year");

    # Does the YYYY-MM-DD.pdf file exist?
    unless (-f "$root_folder$year/$date.pdf") {
        # We need a copy of the $mech object.
        my $pdf = $mech->clone();
        $pdf->get($link, ':content_file' => "$root_folder$year/$date.pdf");
        # Let's do a notification...
        #system("/usr/local/bin/terminal-notifier -message \"Sprint document dated $date has been downloaded.\" -title \"Statement Retrieved\" ");

    }
}

# It seems possible to get statements that aren't listed on the history page. Let's see if we can let them grab those
# too. Note: These only seem to go back to about 2007, always seem to use the 1st for the day of month. Runs forever,
# comment out again after you've grabbed them.
# if (1) {
#   for (my $year = 2008; $year--; $year > 2007) {
#     for my $month ("01" .. "12") {
#       #for () {
#         my $date = "$year-$month-01";

#          # This will create any nested directories necessary. Mostly for the year.
#          File::Path::make_path("$root_folder$year");

#         unless (-f "$root_folder$year/$date.pdf") {
#           # Need to clone it.
#           my $pdf = $mech->clone();
#           my $filepath = "$root_folder$year/$date.pdf";
#           my $link = "/servlet/ecare?inf_template=/servlet/billImageFromOlive?billDate=01/$month/$year";
#           $pdf->get($link, ':content_file' => $filepath);
#           # Check that it was successful. Always get a 200 response code, so we'll check mimetype for app/pdf.
#           if ($pdf->ct() ne "application/pdf") { unlink $filepath; print "Nothing for $date\n"; }
#           else { print "Found $date\n"; }
#         }
#       #}
#     }
#   }
# }