使用 ruby 下载 youtube 视频以及字幕

来源：互联网发布：淘宝小号用什么注册编辑：程序博客网时间：2024/04/30 14:20

最近喜欢上了 Ruby,非常神奇，所以一直想用它做些什么！

正好女朋友想学些专业英语，就去youtube 上翻了一下，有很多呀，关键还有字幕，

只不过直接在网页上看不是很方便，字幕离视频太远了，看字幕看不到视频，看视频有看不到字幕；

用了代理，虽然看着不卡，但是视频太模糊了，所以决定自己下载他高清版本的视频在电脑上看。

正好拿 Ruby 再练一次手，上次用Ruby 登陆新浪微博用的也挺好的，哈哈

要做的有三步：

1：下载视频，网上搜的大部分已经不适用了，不过youtube 获取 flv 的格式没有大变

2：下载字幕文件，并转换成 srt 字幕格式

因为刚学 Ruby 很多 Ruby 的高级功能没用，所以基本还是当平常的语言用的，比如每个函数有都要用return 才安心！

呵呵，慢慢学着改吧吧！

里面包含了 cookie.rb 这个事自己原来写新浪微博登录的时候用到的，这里其实也用不到，直接去掉就可以了！

因为众所周知的原因 youtube 在中国是不能直接访问的，所以需要使用socket 代理

类中：

[ruby] view plaincopy

@ProxyAddress='192.168.0.56'
@ProxyPort=7777

是定义代理地址的，如果 @ProxyAddress 为 nil 则不适用代理！

另外 socksify 可能需要使用 gem 安装

直接贴代码了：

[ruby] view plaincopy

require 'net/http'
require 'uri'
require 'cgi'
require 'socksify/http'
require 'json'
require 'rexml/document'
require_relative '../lib/cookie'
class RstTime
def initialize(sec)
@sec = sec.to_f
end
def Time
tsec = @sec.to_i
#puts tsec.to_s
ms = @sec - tsec
ms = (ms * 1000).to_i
sec = tsec % 60
tsec /= 60
minute = tsec % 60
tsec /= 60
hour = tsec
sprintf("%02d:%02d:%02d,%03d",hour,minute,sec,ms)
end
end
class YoutubeVideo
def initialize page_uri
@ProxyAddress='192.168.0.56'
@ProxyPort=7777
@cookies = Cookie.new()
@UserAgent = 'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:14.0) Gecko/20100101 Firefox/14.0.1'
@VideoUri = URI.parse page_uri
@UriPars = CGI.parse(@VideoUri.query)
if @UriPars.has_key?('v')
@VideoID = @UriPars['v'].first
end
end
def IsValid
return @VideoID != nil
end
def DownloadFile(url ,saveto)
uri = URI(url)
writetotal = 0
saveto = saveto.gsub('/','-')
Net::HTTP.SOCKSProxy(@ProxyAddress, @ProxyPort).start(uri.host, uri.port) do |http|
request = Net::HTTP::Get.new uri.request_uri
http.request request do |response|
open saveto, 'w' do |io|
response.read_body do |chunk|
writetotal += chunk.length
io.write chunk
end
end
end
end
return writetotal
#open(saveto, "wb") do |file|
#file.write(res.body)
#end
end
def DownloadVideos()
index = 0;
#puts JSON.pretty_generate(@downloadUrl)
downloadto = @title.first + ".flv"
@downloadUrl.each do |x|
puts "download video:" + downloadto
break if DownloadFile(x,downloadto) > 0
end
end
def DownloadLyrics()
return if @lyricUrl == nil
lyric = @title.first + ".flv.srt"
res = HttpGet(@lyricUrl+'&type=track&lang=en&name&kind=asr&fmt=1')
doc = REXML::Document.new(res.body)
index = 1
lyric = lyric.gsub('/','-')
open lyric, 'w' do |io|
doc.elements.each('transcript/text') do |ele|
start = ele.attributes['start'].to_f
starttime = RstTime.new(start)
endtime = RstTime.new(start + ele.attributes['dur'].to_f)
io.write index.to_s + "\n"
io.write starttime.Time + ' --> ' + endtime.Time + "\n"
io.write ele.text
io.write "\n"
io.write "\n"
index += 1
end
end
end
def ParseVideoUri()
return nil if @VideoID == nil
infoUri = "http://www.youtube.com/get_video_info?video_id=#{@VideoID}&el=detailpage"
res = HttpGet(infoUri)
return nil if res.code == 200
@videoPars = CGI.parse(res.body)
@url_encoded_fmt_stream_map =CGI.parse(@videoPars['url_encoded_fmt_stream_map'].first)
@rvs =CGI.parse(@videoPars['rvs'].first)
#puts @url_encoded_fmt_stream_map
@videoPars.delete('url_encoded_fmt_stream_map')
@videoPars.delete('rvs')
@title = @videoPars['title']
#puts @url_encoded_fmt_stream_map['url'].class
@downloadUrl = @url_encoded_fmt_stream_map['url']
@signs = @url_encoded_fmt_stream_map['sig']
@downloadUrl.each_index do |x|
@downloadUrl[x] = @downloadUrl[x] + '&signature=' + @signs[x]
end
@lyricUrl = @videoPars['ttsurl'].first
end
def HttpGet(url,limit=10,hdrs=nil)
raise ArgumentError, 'too many HTTP redirects' if limit == 0
uri = URI(url)
ck = @cookies.GetDomainCookie(uri.host)
req = Net::HTTP::Get.new(uri.request_uri)
req['User-Agent'] = @UserAgent
if hdrs != nil
hdrs.each do |key,val|
req[key] = val
end
end
req['Cookie'] = ck if ck != nil
if @ProxyAddress != nil
#puts @ProxyAddress + ':' + @ProxyPort.to_s
res = Net::HTTP.SOCKSProxy(@ProxyAddress, @ProxyPort).start(uri.host, uri.port) do |http|
http.request(req)
end
else
res = Net::HTTP.start(uri.host, uri.port) do |http|
http.request(req)
end
end
@cookies.SaveCookie(uri.host,res.to_hash['set-cookie'])
if res.kind_of?(Net::HTTPRedirection)
newurl = res['location']
newurl = newurl.sub(/^\//,'http://'+uri.host+'/')
HttpGet(newurl,limit-1)
end
return res
end
end
if __FILE__ == $0
exit if ARGV[0] == nil
exit if ARGV[0].length == 0
#yv = YoutubeVideo.new('http://www.youtube.com/watch?v=DC-pQPq0acs&feature=b-vrec');
#yv = YoutubeVideo.new('http://www.youtube.com/watch?v=dtjrJr1oNmo');
yv = YoutubeVideo.new(ARGV[0])
if yv.IsValid() == false
puts "Invalid url:" + ARGV[0]
exit
end
yv.ParseVideoUri()
yv.DownloadVideos()
yv.DownloadLyrics()
end

0 0