Class: DouyinDownload::Parser

Inherits:
Object
  • Object
show all
Defined in:
lib/douyin_download/parser.rb

Defined Under Namespace

Classes: InvalidDouyinUrlError

Constant Summary collapse

MOBILE_USER_AGENT = 
"Mozilla/5.0 (iPhone; CPU iPhone OS 12_1_4 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/12.0 Mobile/15E148 Safari/604.1"

Instance Method Summary collapse

Constructor Details

#initialize(url) ⇒ Parser

Returns a new instance of Parser.



6
7
8
9
10
11
12
 
# File 'lib/douyin_download/parser.rb', line 6

def initialize(url)
  @url = url
  regexp = /v\.douyin\.com\/[\w\d]+/
  unless  @url =~ regexp
    raise InvalidDouyinUrlError.new("Invalid douyin url: #{@url}")
  end
end

Instance Method Details

#api_play_url ⇒ Object 



36
37
38
 
# File 'lib/douyin_download/parser.rb', line 36

def api_play_url
  info.dig('item_list',0,'video','play_addr','url_list',0)
end

#description ⇒ Object 



32
33
34
 
# File 'lib/douyin_download/parser.rb', line 32

def description
  info.dig('item_list',0,'desc')
end

#douyin_id ⇒ Object 



14
15
16
17
 
# File 'lib/douyin_download/parser.rb', line 14

def douyin_id
  url = @url.gsub /https?:\/\/v\.douyin\.com\//,''
  url.gsub '/',''
end

#download(filename: nil) ⇒ Object 



77
78
79
80
81
82
83
84
85
86
87
88
 
# File 'lib/douyin_download/parser.rb', line 77

def download(filename: nil)
  begin
    body = Faraday.get(play_url, {}, 'User-Agent': MOBILE_USER_AGENT).body
    # filename ||= [douyin_id, description].join("-") + ".mp4"
    filename ||= douyin_id + ".mp4"
    File.open(filename,"wb+"){|f|
      f.write body
    }
    filename
  rescue Exception
  end
end

#download_url ⇒ Object 



28
29
30
 
# File 'lib/douyin_download/parser.rb', line 28

def download_url
  info.dig('item_list',0,'video','download_addr','url_list',0)
end

#info ⇒ Object 



47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
 
# File 'lib/douyin_download/parser.rb', line 47

def info
  # 用移动端设备ua访问该页面
  @info ||= begin
    result = {}
    ua = MOBILE_USER_AGENT
    body = Faraday.get(long_url, {}, 'User-Agent': ua).body
    # puts body
    regexp = /\.init\(\{(.+)\}\)\;\s+\}\)\;/m
    api_url = nil
    if body =~ regexp
      json = "{#{$1}}"
      data = parse_json json
      # pp data

      # 构建api url
      item_ids = data[:itemId]
      dytk = data[:dytk]
      api_url = "https://www.iesdouyin.com/web/api/v2/aweme/iteminfo/?item_ids=#{item_ids}&dytk=#{dytk}"
    else
      # byebug
    end

    if api_url
      body = Faraday.get(api_url).body
      result = JSON.parse body
    end
    result
  end
end

#long_url ⇒ Object 



19
20
21
22
23
24
25
26
 
# File 'lib/douyin_download/parser.rb', line 19

def long_url
  @long_url ||= begin
    # url = "https://v.douyin.com/T4Sjhd/"
    # body = Faraday.get(@url).body
    headers = Faraday.get(@url).headers
    headers['location']
  end
end

#play_url ⇒ Object 



40
41
42
43
44
45
 
# File 'lib/douyin_download/parser.rb', line 40

def play_url
  # 这个是终极的播放url
  # http://v9-dy-y.ixigua.com/390e2ef55562a7f85c79d98dc9814475/5ea127a5/video/tos/cn/tos-cn-ve-15/7131a6682c1040c1aa713f8405c34670/?a=1128&br=0&bt=907&cr=0&cs=0&dr=0&ds=6&er=&l=202004231229000100140472040B062432&lr=&qs=0&rc=M3I7NG87PHVodDMzNmkzM0ApaTM2Ojc3Zzs8N2Y6ZTU7ZGdfMGJsNTNhNmRfLS0tLS9zcy40NDU0YDM2LWA2NC1eLzM6Yw%3D%3D&vl=&vr
  headers = Faraday.get(api_play_url,{},'User-Agent': MOBILE_USER_AGENT).headers
  headers['location']
end