-
-
Save lotrfan/28c4a266468bb7658e95 to your computer and use it in GitHub Desktop.
Download Google Reader archives
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/perl | |
use strict; | |
use warnings; | |
require LWP::UserAgent; | |
require HTTP::Cookies; | |
require HTML::Form; | |
use Term::ReadKey; | |
use URI::Escape; | |
our $FEED = ''; | |
our $FEED_NAME = 'feed'; | |
if (@ARGV) { | |
if ($#ARGV >= 1) { | |
$FEED_NAME = $ARGV[0]; | |
$FEED = $ARGV[1]; | |
} else { | |
$FEED = $ARGV[0]; | |
} | |
} | |
our $NUMBER_OF_ITEMS = 1000; | |
if (! $FEED) { | |
print "Usage: <script-name> <feed-url>\n"; | |
print "or : <script-name> <feed-name> <feed-url>\n"; | |
print "You may have to use either single-quotes for the feed-url,\n"; | |
print "or take care using shell escapes.\n"; | |
exit(1); | |
} | |
our $LOGIN_URL = 'https://accounts.google.com/ServiceLogin'; | |
our $COOKIE_FILE = 'cookies.dat'; | |
our $CACHE_URL = 'http://www.google.com/reader/atom/feed/%feed%?r=n&n=%num%'; | |
our $CONTINUED_URL = $CACHE_URL . '&c=%cont%'; | |
our ($user, $pass); | |
our $CookieJar = HTTP::Cookies->new( | |
file => $COOKIE_FILE, | |
autosave => 1, | |
); | |
my $ua = LWP::UserAgent->new; | |
$ua->timeout(100); | |
$ua->cookie_jar($CookieJar); | |
push @{ $ua->requests_redirectable }, 'POST'; | |
#$ua->show_progress(1); | |
if (! -f $COOKIE_FILE) { | |
login(); | |
} | |
our $f = $FEED; | |
our $url = $CACHE_URL; | |
$url =~ s/%feed%/$f/; | |
$url =~ s/%num%/$NUMBER_OF_ITEMS/; | |
our $pageNum = 1; | |
our $padding = int(log(20000 / $NUMBER_OF_ITEMS)); | |
our $response; | |
while ($url) { | |
my $filename = sprintf "%s-%0" . $padding . "d.xml", $FEED_NAME, $pageNum; | |
printf "Page %d => %s\n", $pageNum, $filename; | |
$response = $ua->get($url); | |
open(my $fh, ">", $filename); | |
binmode($fh, ":utf8"); | |
print $fh $response->decoded_content; | |
close($fh); | |
$url = ''; | |
if ($response->decoded_content =~ /<gr:continuation>(.*?)<\/gr:continuation>/) { | |
#print "Found continuation: $1\n"; | |
my $cont = $1; | |
$url = $CONTINUED_URL; | |
$url =~ s/%feed%/$f/; | |
$url =~ s/%num%/$NUMBER_OF_ITEMS/; | |
$url =~ s/%cont%/$cont/; | |
} | |
$pageNum ++; | |
} | |
sub login { | |
print "Username: "; | |
$user = ReadLine(0); | |
chomp $user; | |
print "Password: "; | |
ReadMode('noecho'); | |
$pass = ReadLine(0); | |
print "\n"; | |
ReadMode(0); | |
chomp $pass; | |
my $response = $ua->get($LOGIN_URL); | |
my $login_form = HTML::Form->parse($response); | |
$login_form->find_input("Email")->value($user); | |
$login_form->find_input("Passwd")->value($pass); | |
my $request = $login_form->click; | |
$ua->request($request); | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment