tech234a
/
ytcc-archive
mirror of https://github.com/Data-Horde/ytcc-archive
1
0
Fork 0
Code Issues 0 Releases 0 Wiki Activity
archiving community contributions on YouTube: unpublished captions, title and description translations and caption credits
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
106 Commits
7 Branches
348 KiB
 
 
Tree: ed8d73e221
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'ed8d73e221'
${ noResults }
ytcc-archive/worker.py

252 lines
12 KiB
Raw Blame History 

  1. from threading import Thread

  2. import requests

  3. from time import sleep

  4. from os import mkdir, rmdir, listdir, system, environ

  5. from os.path import isdir, isfile, getsize

  6. from json import dumps, loads

  7. 

  8. import signal

  9. 

  10. import tracker

  11. 

  12. from youtube_dl import YoutubeDL

  13. 

  14. from shutil import rmtree, which

  15. 

  16. from queue import Queue

  17. 

  18. from gc import collect

  19. 

  20. from discovery import getmetadata

  21. from export import subprrun

  22. 

  23. #useful Queue example: https://stackoverflow.com/a/54658363

  24. jobs = Queue()

  25. 

  26. langcnt = {}

  27. 

  28. try:

  29.     mkdir("out")

  30. except:

  31.     pass

  32. 

  33. try:

  34.     mkdir("directory")

  35. except:

  36.     pass

  37. 

  38. HEROKU = False

  39. if isfile("../Procfile"):

  40.     HEROKU = True

  41. 

  42. langs = ['ab', 'aa', 'af', 'sq', 'ase', 'am', 'ar', 'arc', 'hy', 'as', 'ay', 'az', 'bn', 'ba', 'eu', 'be', 'bh', 'bi', 'bs', 'br', 

  43.     'bg', 'yue', 'yue-HK', 'ca', 'chr', 'zh-CN', 'zh-HK', 'zh-Hans', 'zh-SG', 'zh-TW', 'zh-Hant', 'cho', 'co', 'hr', 'cs', 'da', 'nl', 

  44.     'nl-BE', 'nl-NL', 'dz', 'en', 'en-CA', 'en-IN', 'en-IE', 'en-GB', 'en-US', 'eo', 'et', 'fo', 'fj', 'fil', 'fi', 'fr', 'fr-BE', 

  45.     'fr-CA', 'fr-FR', 'fr-CH', 'ff', 'gl', 'ka', 'de', 'de-AT', 'de-DE', 'de-CH', 'el', 'kl', 'gn', 'gu', 'ht', 'hak', 'hak-TW', 'ha', 

  46.     'iw', 'hi', 'hi-Latn', 'ho', 'hu', 'is', 'ig', 'id', 'ia', 'ie', 'iu', 'ik', 'ga', 'it', 'ja', 'jv', 'kn', 'ks', 'kk', 'km', 'rw', 

  47.     'tlh', 'ko', 'ku', 'ky', 'lo', 'la', 'lv', 'ln', 'lt', 'lb', 'mk', 'mg', 'ms', 'ml', 'mt', 'mni', 'mi', 'mr', 'mas', 'nan', 

  48.     'nan-TW', 'lus', 'mo', 'mn', 'my', 'na', 'nv', 'ne', 'no', 'oc', 'or', 'om', 'ps', 'fa', 'fa-AF', 'fa-IR', 'pl', 'pt', 'pt-BR', 

  49.     'pt-PT', 'pa', 'qu', 'ro', 'rm', 'rn', 'ru', 'ru-Latn', 'sm', 'sg', 'sa', 'sc', 'gd', 'sr', 'sr-Cyrl', 'sr-Latn', 'sh', 'sdp', 'sn', 

  50.     'scn', 'sd', 'si', 'sk', 'sl', 'so', 'st', 'es', 'es-419', 'es-MX', 'es-ES', 'es-US', 'su', 'sw', 'ss', 'sv', 'tl', 'tg', 'ta', 

  51.     'tt', 'te', 'th', 'bo', 'ti', 'tpi', 'to', 'ts', 'tn', 'tr', 'tk', 'tw', 'uk', 'ur', 'uz', 'vi', 'vo', 'vor', 'cy', 'fy', 'wo', 

  52.     'xh', 'yi', 'yo', 'zu']

  53. 

  54. assert which("zip") and which("rsync") and which("curl"), "Please ensure the zip, rsync, and curl commands are installed on your system."

  55. 

  56. #HSID, SSID, SID cookies required

  57. if "HSID" in environ.keys() and "SSID" in environ.keys() and "SID" in environ.keys():

  58.     cookies = {"HSID": environ["HSID"], "SSID": environ["SSID"], "SID": environ["SID"]}

  59. elif isfile("config.json"):

  60.     cookies = loads(open("config.json").read())

  61. else:

  62.     print("HSID, SSID, and SID cookies from youtube.com are required. Specify in config.json or as environment variables.")

  63.     assert False

  64. if not (cookies["HSID"] and cookies["SSID"] and cookies["SID"]):

  65.     print("HSID, SSID, and SID cookies from youtube.com are required. Specify in config.json or as environment variables.")

  66.     assert False

  67. 

  68. mysession = requests.session()

  69. mysession.headers.update({"cookie": "HSID="+cookies["HSID"]+"; SSID="+cookies["SSID"]+"; SID="+cookies["SID"], "Accept-Language": "en-US",})

  70. 

  71. validationtest = mysession.get("https://www.youtube.com/timedtext_editor?action_mde_edit_form=1&v=1iNTtHUwvq4&lang=en&bl=vmp&ui=hd&ref=player&tab=captions&o=U")

  72. 

  73. assert not "accounts.google.com" in validationtest.url, "Please ensure you have correctly specified account cookies."

  74. assert """<button class="yt-uix-button yt-uix-button-size-default yt-uix-button-default yt-uix-button-has-icon" type="button" onclick=";return false;" id="yt-picker-language-button" data-button-action="yt.www.picker.load" data-button-menu-id="arrow-display" data-picker-key="language" data-picker-position="footer" data-button-toggle="true"><span class="yt-uix-button-icon-wrapper"><span class="yt-uix-button-icon yt-uix-button-icon-footer-language yt-sprite"></span></span><span class="yt-uix-button-content">  <span class="yt-picker-button-label">

  75. Language:

  76.   </span>

  77.   English

  78. </span><span class="yt-uix-button-arrow yt-sprite"></span></button>""" in validationtest.text, "Please make sure your YouTube and Google account language is set to English (United States)"

  79. 

  80. del validationtest

  81. 

  82. open("cookies.txt", "w").write("""# HTTP Cookie File

  83. .youtube.com	TRUE	/	FALSE	1663793455	SID	[SID]

  84. .youtube.com	TRUE	/	FALSE	1663793455	HSID	[HSID]

  85. .youtube.com	TRUE	/	TRUE	1663793455	SSID	[SSID]""".replace("[SID]", cookies["SID"]).replace("[HSID]", cookies["HSID"]).replace("[SSID]", cookies["SSID"]))

  86. 

  87. del cookies

  88. 

  89. #Graceful Shutdown

  90. class GracefulKiller:

  91.     kill_now = False

  92.     def __init__(self):

  93.         signal.signal(signal.SIGINT, self.exit_gracefully)

  94.         signal.signal(signal.SIGTERM, self.exit_gracefully)

  95. 

  96.     def exit_gracefully(self, signum, frame):

  97.         print("Graceful exit process initiated, no longer accepting new tasks but finishing existing ones...")

  98.         self.kill_now = True

  99. 

  100. gkiller = GracefulKiller()

  101. 

  102. #microtasks

  103. def threadrunner(jobs: Queue):

  104.     global langcnt

  105.     ydl = YoutubeDL({"extract_flat": "in_playlist", "simulate": True, "skip_download": True, "quiet": True, "cookiefile": "cookies.txt", "source_address": "0.0.0.0", "call_home": False})

  106.     while True:

  107.         if not jobs.empty():

  108.             task, vid, args = jobs.get()

  109. 

  110.             if task == "submitdiscovery":

  111.                 tracker.add_item_to_tracker(args, vid)

  112.             elif task == "discovery":

  113.                 while True:

  114.                     try:

  115.                         info = getmetadata(mysession, str(vid).strip())

  116.                         break

  117.                     except BaseException as e:

  118.                         print(e)

  119.                         print("Error in retrieving information, waiting 30 seconds")

  120.                         sleep(30)

  121.                 if info[0] or info[1]: # ccenabled or creditdata

  122.                     if not isdir("out/"+str(vid).strip()):

  123.                         mkdir("out/"+str(vid).strip())

  124.                 if info[1]:

  125.                     open("out/"+str(vid).strip()+"/"+str(vid).strip()+"_published_credits.json", "w").write(dumps(info[1]))

  126. 

  127.                 if info[0]:

  128.                     langcnt[vid] = 0

  129.                     for langcode in langs:

  130.                         jobs.put(("subtitles", vid, langcode))

  131.                 else:

  132.                     jobs.put(("complete", None, "video:"+vid))

  133. 

  134.                 for videodisc in info[2]:

  135.                     jobs.put(("submitdiscovery", videodisc, tracker.ItemType.Video))

  136.                 for channeldisc in info[3]:

  137.                     jobs.put(("submitdiscovery", channeldisc, tracker.ItemType.Channel))

  138.                 for mixdisc in info[4]:

  139.                     jobs.put(("submitdiscovery", mixdisc, tracker.ItemType.MixPlaylist))

  140.                 for playldisc in info[5]:

  141.                     jobs.put(("submitdiscovery", playldisc, tracker.ItemType.Playlist))

  142. 

  143.             elif task == "subtitles":

  144.                 retval = subprrun(jobs, mysession, args, vid, "default")

  145.                 langcnt[vid] += retval

  146.                 if langcnt[vid] >= 585:

  147.                     jobs.put(("complete", None, "video:"+vid))

  148.             elif task == "subtitles-forceedit-captions":

  149.                 subprrun(jobs, mysession, args, vid, "forceedit-captions")

  150.             elif task == "subtitles-forceedit-metadata":

  151.                 subprrun(jobs, mysession, args, vid, "forceedit-metadata")

  152.             elif task == "channel":

  153.                 try:

  154.                     y = ydl.extract_info("https://www.youtube.com/channel/"+desit.split(":", 1)[1], download=False)

  155.                     for itemyv in y["entries"]:

  156.                         jobs.put(("submitdiscovery", itemyv["id"], tracker.ItemType.Video))

  157.                     jobs.put(("complete", None, "channel:"+args))

  158.                 except:

  159.                     print("YouTube-DL error, ignoring but not marking as complete...", "https://www.youtube.com/channel/"+desit.split(":", 1)[1])

  160.             elif task == "playlist":

  161.                 try:

  162.                     y = ydl.extract_info("https://www.youtube.com/playlist?list="+desit.split(":", 1)[1], download=False)

  163.                     for itemyvp in y["entries"]:

  164.                         jobs.put(("submitdiscovery", itemyvp["id"], tracker.ItemType.Video))

  165.                     jobs.put(("complete", None, "playlist:"+args))

  166.                 except:

  167.                     print("YouTube-DL error, ignoring but not marking as complete...", "https://www.youtube.com/playlist?list="+desit.split(":", 1)[1])

  168.             elif task == "complete":

  169.                 size = 0

  170.                 if ":" in args:

  171.                     if args.split(":", 1)[0] == "video":

  172.                         #check if dir is empty, make zip if needed

  173.                         if isdir("out/"+args.split(":", 1)[1]):

  174.                             if not listdir("out/"+args.split(":", 1)[1]):

  175.                                 rmdir("out/"+args.split(":", 1)[1])

  176.                             else:

  177.                                 #zip it up

  178.                                 if not isdir("directory/"+args.split(":", 1)[1]):

  179.                                     mkdir("directory/"+args.split(":", 1)[1])

  180. 

  181.                                 while not isfile("directory/"+args.split(":", 1)[1]+"/"+args.split(":", 1)[1]+".zip"):

  182.                                     print("Attempting to zip item...")

  183.                                     system("zip -9 -r -j directory/"+args.split(":", 1)[1]+"/"+args.split(":", 1)[1]+".zip out/"+args.split(":", 1)[1])

  184. 

  185.                                 #get a target

  186.                                 targetloc = None

  187.                                 while not targetloc:

  188.                                     targetloc = tracker.request_upload_target()

  189.                                     if targetloc:

  190.                                         break

  191.                                     else:

  192.                                         print("Waiting 5 minutes...")

  193.                                         sleep(300)

  194. 

  195.                                 if targetloc.startswith("rsync"):

  196.                                     system("rsync -rltv --timeout=300 --contimeout=300 --progress --bwlimit 0 --recursive --partial --partial-dir .rsync-tmp --min-size 1 --no-compress --compress-level 0 directory/"+args.split(":", 1)[1]+"/ "+targetloc)

  197.                                 elif targetloc.startswith("http"):

  198.                                     system("curl -F "+args.split(":", 1)[1]+".zip=@directory/"+args.split(":", 1)[1]+"/"+args.split(":", 1)[1]+".zip "+targetloc)

  199. 

  200. 

  201.                                 size = getsize("directory/"+args.split(":", 1)[1]+"/"+args.split(":", 1)[1]+".zip")

  202.                                 #cleanup

  203.                                 try:

  204.                                     del langcnt[args.split(":", 1)[1]]

  205.                                     rmtree("directory/"+args.split(":", 1)[1]+"/")

  206.                                     rmdir("directory/"+args.split(":", 1)[1]+"/")

  207.                                     rmtree("out/"+args.split(":", 1)[1]+"/")

  208.                                     rmdir("out/"+args.split(":", 1)[1]+"/")

  209.                                 except:

  210.                                     pass

  211.                 tracker.mark_item_as_done(args, size)

  212.             jobs.task_done()

  213.         else:

  214.             if not gkiller.kill_now:

  215.                 # get a new task from tracker

  216.                 collect() #cleanup

  217.                 desit = tracker.request_item_from_tracker()

  218.                 print("New task:", desit)

  219.                 if desit:

  220.                     if desit.split(":", 1)[0] == "video":

  221.                         jobs.put(("discovery", desit.split(":", 1)[1], None))

  222.                     elif desit.split(":", 1)[0] == "channel":

  223.                         jobs.put(("channel", None, desit.split(":", 1)[1]))

  224.                     elif desit.split(":", 1)[0] == "playlist":

  225.                         jobs.put(("playlist", None, desit.split(":", 1)[1]))

  226.                     else:

  227.                         print("Ignoring item for now", desit)

  228.                 else:

  229.                     print("Ignoring item for now", desit)

  230.             else:

  231.                 break

  232.     

  233. 

  234. threads = []

  235. 

  236. THREADCNT = 50

  237. if HEROKU:

  238.     THREADCNT = 20

  239. #now create the rest of the threads

  240. for i in range(THREADCNT):

  241.     runthread = Thread(target=threadrunner, args=(jobs,))

  242.     runthread.start()

  243.     threads.append(runthread)

  244.     del runthread

  245. 

  246. #https://stackoverflow.com/a/11968881

  247. for x in threads:

  248.     x.join()

  249.     threads.remove(x)

  250.     del x

  251. 

  252. print("Exiting...")