Created
March 15, 2013 11:35
-
-
Save unycone/5169273 to your computer and use it in GitHub Desktop.
To retrieve raw video ulrs from youtube.
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
require 'uri' | |
require 'net/http' | |
require 'cgi' | |
class ScrapeYoutube | |
@@youtube_url = 'http://www.youtube.com' | |
@@youtube_info_path = '/get_video_info?video_id=' | |
@@key = "url_encoded_fmt_stream_map" | |
def self.showWebmUrl(video_id) | |
map = getVideoInfo(video_id) | |
map[@@key].each do |m| | |
if (m["type"].start_with?("video/webm")) then | |
puts m["url"] | |
end | |
end | |
end | |
def self.showInfo(video_id) | |
printBody(getVideoInfo(video_id)) | |
end | |
def self.getVideoInfo(video_id) | |
url = URI.parse(@@youtube_url+@@youtube_info_path+video_id) | |
req = Net::HTTP::Get.new(url.to_s) | |
res = Net::HTTP.start(url.host, url.port) {|http| | |
http.request(req) | |
} | |
map = toMap(res.body) | |
map[@@key] = parseFmt(map[@@key]) | |
return map | |
end | |
def self.printBody(map, indent="") | |
map.each_pair do |k, v| | |
if (v.kind_of?(Hash)) then | |
puts indent + k + " =>" | |
printBody(v, " "+indent) | |
elsif (v.kind_of?(Array)) then | |
puts indent + k + " =>" | |
v.each do |e| | |
printBody(e, " "+indent) | |
puts("") | |
end | |
else | |
puts indent + k + " => " + v | |
end | |
end | |
end | |
def self.toMap(str) | |
map = Hash::new | |
str.split('&').each do |e| | |
ary = e.split('=') | |
map[ary[0]] = URI.decode(ary[1] || "") | |
end | |
return map | |
end | |
def self.parseFmt(str) | |
dst = [] | |
map = Hash::new | |
str.split(',').each do |e| | |
dst << toMap(e) | |
end | |
return dst | |
end | |
end | |
ScrapeYoutube::showWebmUrl("sPasebVMIW4") | |
ScrapeYoutube::showInfo("sPasebVMIW4") |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment