logo

multimedia-dl

Unnamed repository; edit this file 'description' to name the repository.

openings-moe.pl (1474B)

    

  1. #!/usr/bin/env perl
  2. # Multimedia-DL: Youtube-DL inspired scraper
  3. # Copyright © 2021 Multimedia-DL Authors <https://hacktivis.me/git/multimedia-dl/>
  4. # SPDX-License-Identifier: AGPL-3-only
  5. use strict;
  6. use utf8;
  7. use HTML::TreeBuilder::XPath;
  8. use URI;
  9. use LWP::UserAgent;
  10. require HTTP::Request;
  12. my $ua = LWP::UserAgent->new;
  14. # Picked this one for it's stability
  15. my $webkit_ua = "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/14.0 Safari/605.1.15";
  17. $ua->agent($webkit_ua . "Multimedia-DL/1.0");
  19. if($#ARGV != 0) {
  20. 	print "usage: multimedia-dl <url>\n";
  21. 	exit 1;
  22. }
  24. my $req = HTTP::Request->new(GET => $ARGV[0]);
  26. my $res = $ua->request($req);
  28. sub scrap_html_response {
  29. 	my ($res) = @_;
  31. 	my $tree = HTML::TreeBuilder::XPath->new_from_content($res->content) or die "HTML parsing failed";
  33. 	my $title = $tree->findvalue('//title');
  34. 	if($title) {
  35. 		print STDERR "Title: ", $title, "\n";
  36. 	}
  38. 	foreach($tree->findvalues('//a[@download]/@href')) {
  39. 		my $href = $_;
  40. 		my $href_abs = URI->new_abs($_, $res->base);
  42. 		if ($href =~ /^subtitles\//) {
  43. 			print "--sub-file='", $href_abs, "' ";
  44. 		} else {
  45. 			print "'", $href_abs, "' ";
  46. 		}
  47. 	}
  49. 	print "--title='", $title, "' ";
  51. 	print "\n";
  52. }
  54. if($res->is_success) {
  55. 	my $content_type = $res->header("Content-Type");
  57. 	if(($content_type == "text/html") or ($content_type == "application/xhtml+xml")) {
  58. 		scrap_html_response($res);
  59. 	}
  60. } else {
  61. 	print "Got ", $res->status_line, " instead of 2xx\n";
  62. }