share/subtitle/youtube.lua

   1 -- libquvi-scripts
   2 -- Copyright (C) 2013  Toni Gundogdu <legatvs@gmail.com>
   3 --
   4 -- This file is part of libquvi-scripts <http://quvi.sourceforge.net/>.
   5 --
   6 -- This program is free software: you can redistribute it and/or
   7 -- modify it under the terms of the GNU Affero General Public
   8 -- License as published by the Free Software Foundation, either
   9 -- version 3 of the License, or (at your option) any later version.
  10 --
  11 -- This program is distributed in the hope that it will be useful,
  12 -- but WITHOUT ANY WARRANTY; without even the implied warranty of
  13 -- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14 -- GNU Affero General Public License for more details.
  15 --
  16 -- You should have received a copy of the GNU Affero General
  17 -- Public License along with this program.  If not, see
  18 -- <http://www.gnu.org/licenses/>.
  19 --
  20
  21 local YouTube = {} -- Utility functions unique to this script
  22
  23 -- Identify the script.
  24 function ident(qargs)
  25   local Y = require 'quvi/youtube'
  26   return Y.ident(qargs)
  27 end
  28
  29 -- Parse subtitle properties.
  30 function parse(qargs)
  31   local Y = require 'quvi/youtube'
  32   local C = require 'quvi/const'
  33
  34   local u = Y.normalize(qargs.input_url)
  35   local v = u:match('v=([%w-_]+)') or error('no match: media ID')
  36
  37   qargs.subtitles = {}
  38
  39   YouTube.tts_get(qargs, v, u, C)
  40   YouTube.cc_get(qargs, v, C)
  41
  42   return qargs
  43 end
  44
  45 --
  46 -- Utility functions
  47 --
  48
  49 -- Extract the TTS (text-to-speech, or transcript).
  50 function YouTube.tts_get(qargs, v, u, C)
  51   local u_fmt = "%s&tlang=%s&type=trackformat=1,&lang=en&kind=asr"
  52   local p = quvi.http.fetch(u).data
  53
  54   local tts_url = p:match('[\'"]TTS_URL[\'"]:%s+[\'"](.-)[\'"]')
  55   if not tts_url then return end
  56
  57   p = nil
  58
  59   tts_url = tts_url:gsub('\\u0026','&')
  60   local U = require 'quvi/util'
  61   tts_url = U.slash_unescape(tts_url)
  62   tts_url = U.unescape(tts_url)
  63
  64   local langs = tts_url:match('asr_langs=(.-)&')
  65   if not langs then return end
  66
  67   local r = {}
  68   for c in langs:gmatch('(%w+)') do
  69     table.insert(r, {
  70       url = string.format(u_fmt, tts_url, c),
  71       translated = '',
  72       id = 'tts_'..c,
  73       original = '',
  74       code = c
  75     })
  76   end
  77
  78   table.insert(qargs.subtitles, {format=C.sif_tt, type=C.st_tts, lang=r})
  79 end
  80
  81 -- Return a new timed-text track URL.
  82 function YouTube.tt_track_new(scheme, v, name, lang)
  83   local t = {
  84     scheme, '://youtube.com/api/timedtext?hl=en&type=track',
  85     '&v=', v, '&name=', name, '&lang=', lang
  86   }
  87   return table.concat(t,'')
  88 end
  89
  90 -- Return a new timed-text list URL.
  91 function YouTube.tt_list_new(scheme, v)
  92   local t = {scheme, '://video.google.com/timedtext?hl=en&type=list&v=', v}
  93   return table.concat(t,'')
  94 end
  95
  96 -- Extract the CC (closed-captions) data.
  97 function YouTube.cc_get(qargs, v, C)
  98   local U = require 'socket.url'
  99   local u = U.parse(qargs.input_url)
 100
 101   local l = YouTube.tt_list_new(u.scheme, v)
 102   local x = quvi.http.fetch(l).data
 103
 104   local L = require 'lxp.lom'
 105   local t = L.parse(x)
 106
 107   local r = {}
 108   for i=1, #t do
 109     if t[i].tag == 'track' then
 110       local lang = t[i].attr['lang_code']
 111       if lang then
 112         local name = t[i].attr['name'] or ''
 113         table.insert(r, {
 114           translated = t[i].attr['lang_translated'] or '',
 115           url = YouTube.tt_track_new(u.scheme, v, name, lang),
 116           original = t[i].attr['lang_original'] or '',
 117           id = table.concat({'cc_', lang}, ''),
 118           code = lang
 119         })
 120       end
 121     end
 122   end
 123   table.insert(qargs.subtitles, {format=C.sif_tt, type=C.st_cc, lang=r})
 124 end
 125
 126 -- vim: set ts=2 sw=2 tw=72 expandtab: