summaryrefslogtreecommitdiffstats
path: root/src/invidious/helpers/helpers.cr
blob: c887f1a2c961835efcd28f937ce001b5f5d7b7ff (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
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
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
class Config
  YAML.mapping({
    crawl_threads:   Int32,
    channel_threads: Int32,
    feed_threads:    Int32,
    video_threads:   Int32,
    db:              NamedTuple(
      user: String,
      password: String,
      host: String,
      port: Int32,
      dbname: String,
    ),
    dl_api_key:   String?,
    https_only:   Bool?,
    hmac_key:     String?,
    full_refresh: Bool,
    geo_bypass:   Bool,
  })
end

class FilteredCompressHandler < Kemal::Handler
  exclude ["/videoplayback", "/videoplayback/*", "/vi/*", "/api/*", "/ggpht/*"]

  def call(env)
    return call_next env if exclude_match? env

    {% if flag?(:without_zlib) %}
      call_next env
    {% else %}
      request_headers = env.request.headers

      if request_headers.includes_word?("Accept-Encoding", "gzip")
        env.response.headers["Content-Encoding"] = "gzip"
        env.response.output = Gzip::Writer.new(env.response.output, sync_close: true)
      elsif request_headers.includes_word?("Accept-Encoding", "deflate")
        env.response.headers["Content-Encoding"] = "deflate"
        env.response.output = Flate::Writer.new(env.response.output, sync_close: true)
      end

      call_next env
    {% end %}
  end
end

class DenyFrame < Kemal::Handler
  exclude ["/embed/*"]

  def call(env)
    return call_next env if exclude_match? env

    env.response.headers["X-Frame-Options"] = "sameorigin"
    call_next env
  end
end

def rank_videos(db, n, filter, url)
  top = [] of {Float64, String}

  db.query("SELECT id, wilson_score, published FROM videos WHERE views > 5000 ORDER BY published DESC LIMIT 1000") do |rs|
    rs.each do
      id = rs.read(String)
      wilson_score = rs.read(Float64)
      published = rs.read(Time)

      # Exponential decay, older videos tend to rank lower
      temperature = wilson_score * Math.exp(-0.000005*((Time.now - published).total_minutes))
      top << {temperature, id}
    end
  end

  top.sort!

  # Make hottest come first
  top.reverse!
  top = top.map { |a, b| b }

  if filter
    language_list = [] of String
    top.each do |id|
      if language_list.size == n
        break
      else
        client = make_client(url)
        begin
          video = get_video(id, db)
        rescue ex
          next
        end

        if video.language
          language = video.language
        else
          description = XML.parse(video.description)
          content = [video.title, description.content].join(" ")
          content = content[0, 10000]

          results = DetectLanguage.detect(content)
          language = results[0].language

          db.exec("UPDATE videos SET language = $1 WHERE id = $2", language, id)
        end

        if language == "en"
          language_list << id
        end
      end
    end
    return language_list
  else
    return top[0..n - 1]
  end
end

def login_req(login_form, f_req)
  data = {
    "pstMsg"          => "1",
    "checkConnection" => "youtube",
    "checkedDomains"  => "youtube",
    "hl"              => "en",
    "deviceinfo"      => %q([null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]),
    "f.req"           => f_req,
    "flowName"        => "GlifWebSignIn",
    "flowEntry"       => "ServiceLogin",
  }

  data = login_form.merge(data)

  return HTTP::Params.encode(data)
end

def generate_captcha(key)
  minute = Random::Secure.rand(12)
  minute_angle = minute * 30
  minute = minute * 5

  hour = Random::Secure.rand(12)
  hour_angle = hour * 30 + minute_angle.to_f / 12
  if hour == 0
    hour = 12
  end

  clock_svg = <<-END_SVG
  <svg viewBox="0 0 100 100" width="200px">
  <circle cx="50" cy="50" r="45" fill="#eee" stroke="black" stroke-width="2"></circle>
  
  <text x="69"     y="20.091" text-anchor="middle" fill="black" font-family="Arial" font-size="10px"> 1</text>
  <text x="82.909" y="34"     text-anchor="middle" fill="black" font-family="Arial" font-size="10px"> 2</text>
  <text x="88"     y="53"     text-anchor="middle" fill="black" font-family="Arial" font-size="10px"> 3</text>
  <text x="82.909" y="72"     text-anchor="middle" fill="black" font-family="Arial" font-size="10px"> 4</text>
  <text x="69"     y="85.909" text-anchor="middle" fill="black" font-family="Arial" font-size="10px"> 5</text>
  <text x="50"     y="91"     text-anchor="middle" fill="black" font-family="Arial" font-size="10px"> 6</text>
  <text x="31"     y="85.909" text-anchor="middle" fill="black" font-family="Arial" font-size="10px"> 7</text>
  <text x="17.091" y="72"     text-anchor="middle" fill="black" font-family="Arial" font-size="10px"> 8</text>
  <text x="12"     y="53"     text-anchor="middle" fill="black" font-family="Arial" font-size="10px"> 9</text>
  <text x="17.091" y="34"     text-anchor="middle" fill="black" font-family="Arial" font-size="10px">10</text>
  <text x="31"     y="20.091" text-anchor="middle" fill="black" font-family="Arial" font-size="10px">11</text>
  <text x="50"     y="15"     text-anchor="middle" fill="black" font-family="Arial" font-size="10px">12</text>

  <circle cx="50" cy="50" r="3" fill="black"></circle>
  <line id="minute" transform="rotate(#{minute_angle}, 50, 50)" x1="50" y1="50" x2="50" y2="16" fill="black" stroke="black" stroke-width="2"></line>
  <line id="hour"   transform="rotate(#{hour_angle}, 50, 50)" x1="50" y1="50" x2="50" y2="24" fill="black" stroke="black" stroke-width="2"></line>
  </svg>
  END_SVG

  challenge = ""
  convert = Process.run(%(convert -density 1200 -resize 400x400 -background none svg:- png:-), shell: true,
    input: IO::Memory.new(clock_svg), output: Process::Redirect::Pipe) do |proc|
    challenge = proc.output.gets_to_end
    challenge = Base64.strict_encode(challenge)
    challenge = "data:image/png;base64,#{challenge}"
  end

  answer = "#{hour}:#{minute.to_s.rjust(2, '0')}"
  token = OpenSSL::HMAC.digest(:sha256, key, answer)
  token = Base64.urlsafe_encode(token)

  return {challenge: challenge, token: token}
end

def html_to_content(description_html)
  if !description_html
    description = ""
    description_html = ""
  else
    description_html = description_html.to_s
    description = description_html.gsub("<br>", "\n")
    description = description.gsub("<br/>", "\n")

    if description.empty?
      description = ""
    else
      description = XML.parse_html(description).content.strip("\n ")
    end
  end

  return description_html, description
end

def extract_videos(nodeset, ucid = nil)
  videos = extract_items(nodeset, ucid)
  videos.select! { |item| !item.is_a?(SearchChannel | SearchPlaylist) }
  videos.map { |video| video.as(SearchVideo) }
end

def extract_items(nodeset, ucid = nil)
  # TODO: Make this a 'common', so it makes more sense to be used here
  items = [] of SearchItem

  nodeset.each do |node|
    anchor = node.xpath_node(%q(.//h3[contains(@class,"yt-lockup-title")]/a))
    if !anchor
      next
    end

    if anchor["href"].starts_with? "https://www.googleadservices.com"
      next
    end

    anchor = node.xpath_node(%q(.//div[contains(@class, "yt-lockup-byline")]/a))
    if !anchor
      author = ""
      author_id = ""
    else
      author = anchor.content.strip
      author_id = anchor["href"].split("/")[-1]
    end

    anchor = node.xpath_node(%q(.//h3[contains(@class, "yt-lockup-title")]/a))
    if !anchor
      next
    end
    title = anchor.content.strip
    id = anchor["href"]

    description_html = node.xpath_node(%q(.//div[contains(@class, "yt-lockup-description")]))
    description_html, description = html_to_content(description_html)

    tile = node.xpath_node(%q(.//div[contains(@class, "yt-lockup-tile")]))
    if !tile
      next
    end

    case tile["class"]
    when .includes? "yt-lockup-playlist"
      plid = HTTP::Params.parse(URI.parse(id).query.not_nil!)["list"]

      anchor = node.xpath_node(%q(.//div[contains(@class, "yt-lockup-meta")]/a))

      if !anchor
        anchor = node.xpath_node(%q(.//ul[@class="yt-lockup-meta-info"]/li/a))
      end

      video_count = node.xpath_node(%q(.//span[@class="formatted-video-count-label"]/b))
      if video_count
        video_count = video_count.content

        if video_count == "50+"
          author = "YouTube"
          author_id = "UC-9-kyTW8ZkZNDHQJ6FgpwQ"
          video_count = video_count.rchop("+")
        end

        video_count = video_count.to_i?
      end
      video_count ||= 0

      videos = [] of SearchPlaylistVideo
      node.xpath_nodes(%q(.//*[contains(@class, "yt-lockup-playlist-items")]/li)).each do |video|
        anchor = video.xpath_node(%q(.//a))
        if anchor
          video_title = anchor.content.strip
          id = HTTP::Params.parse(URI.parse(anchor["href"]).query.not_nil!)["v"]
        end
        video_title ||= ""
        id ||= ""

        anchor = video.xpath_node(%q(.//span/span))
        if anchor
          length_seconds = decode_length_seconds(anchor.content)
        end
        length_seconds ||= 0

        videos << SearchPlaylistVideo.new(
          video_title,
          id,
          length_seconds
        )
      end

      items << SearchPlaylist.new(
        title,
        plid,
        author,
        author_id,
        video_count,
        videos
      )
    when .includes? "yt-lockup-channel"
      author = title.strip
      ucid = id.split("/")[-1]

      author_thumbnail = node.xpath_node(%q(.//div/span/img)).try &.["data-thumb"]?
      author_thumbnail ||= node.xpath_node(%q(.//div/span/img)).try &.["src"]
      author_thumbnail ||= ""

      subscriber_count = node.xpath_node(%q(.//span[contains(@class, "yt-subscriber-count")])).try &.["title"].delete(",").to_i?
      subscriber_count ||= 0

      video_count = node.xpath_node(%q(.//ul[@class="yt-lockup-meta-info"]/li)).try &.content.split(" ")[0].delete(",").to_i?
      video_count ||= 0

      items << SearchChannel.new(
        author,
        ucid,
        author_thumbnail,
        subscriber_count,
        video_count,
        description,
        description_html
      )
    else
      id = id.lchop("/watch?v=")

      metadata = node.xpath_nodes(%q(.//div[contains(@class,"yt-lockup-meta")]/ul/li))

      begin
        published = decode_date(metadata[0].content.lchop("Streamed ").lchop("Starts "))
      rescue ex
      end
      begin
        published ||= Time.epoch(metadata[0].xpath_node(%q(.//span)).not_nil!["data-timestamp"].to_i64)
      rescue ex
      end
      published ||= Time.now

      begin
        view_count = metadata[0].content.rchop(" watching").delete(",").try &.to_i64?
      rescue ex
      end
      begin
        view_count ||= metadata.try &.[1].content.delete("No views,").try &.to_i64?
      rescue ex
      end
      view_count ||= 0_i64

      length_seconds = node.xpath_node(%q(.//span[@class="video-time"]))
      if length_seconds
        length_seconds = decode_length_seconds(length_seconds.content)
      else
        length_seconds = -1
      end

      live_now = node.xpath_node(%q(.//span[contains(@class, "yt-badge-live")]))
      if live_now
        live_now = true
      else
        live_now = false
      end

      if node.xpath_node(%q(.//span[text()="Premium"]))
        premium = true
      else
        premium = false
      end

      if node.xpath_node(%q(.//span[contains(text(), "Get YouTube Premium")]))
        paid = true
      else
        paid = false
      end

      items << SearchVideo.new(
        title,
        id,
        author,
        author_id,
        published,
        view_count,
        description,
        description_html,
        length_seconds,
        live_now,
        paid,
        premium
      )
    end
  end

  return items
end