yt_dlp/extractor/pr0gramm.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import merge_dicts
   5
   6
   7 class Pr0grammStaticIE(InfoExtractor):
   8     # Possible urls:
   9     # https://pr0gramm.com/static/5466437
  10     _VALID_URL = r'https?://pr0gramm\.com/static/(?P<id>[0-9]+)'
  11     _TEST = {
  12         'url': 'https://pr0gramm.com/static/5466437',
  13         'md5': '52fa540d70d3edc286846f8ca85938aa',
  14         'info_dict': {
  15             'id': '5466437',
  16             'ext': 'mp4',
  17             'title': 'pr0gramm-5466437 by g11st',
  18             'uploader': 'g11st',
  19             'upload_date': '20221221',
  20         }
  21     }
  22
  23     def _real_extract(self, url):
  24         video_id = self._match_id(url)
  25         webpage = self._download_webpage(url, video_id)
  26
  27         # Fetch media sources
  28         entries = self._parse_html5_media_entries(url, webpage, video_id)
  29         media_info = entries[0]
  30
  31         # Fetch author
  32         uploader = self._html_search_regex(r'by\W+([\w-]+)\W+', webpage, 'uploader')
  33
  34         # Fetch approx upload timestamp from filename
  35         # Have None-defaults in case the extraction fails
  36         uploadDay = None
  37         uploadMon = None
  38         uploadYear = None
  39         uploadTimestr = None
  40         # (//img.pr0gramm.com/2022/12/21/62ae8aa5e2da0ebf.mp4)
  41         m = re.search(r'//img\.pr0gramm\.com/(?P<year>[\d]+)/(?P<mon>[\d]+)/(?P<day>[\d]+)/\w+\.\w{,4}', webpage)
  42
  43         if (m):
  44             # Up to a day of accuracy should suffice...
  45             uploadDay = m.groupdict().get('day')
  46             uploadMon = m.groupdict().get('mon')
  47             uploadYear = m.groupdict().get('year')
  48             uploadTimestr = uploadYear + uploadMon + uploadDay
  49
  50         return merge_dicts({
  51             'id': video_id,
  52             'title': 'pr0gramm-%s%s' % (video_id, (' by ' + uploader) if uploader else ''),
  53             'uploader': uploader,
  54             'upload_date': uploadTimestr
  55         }, media_info)
  56
  57
  58 # This extractor is for the primary url (used for sharing, and appears in the
  59 # location bar) Since this page loads the DOM via JS, yt-dl can't find any
  60 # video information here. So let's redirect to a compatibility version of
  61 # the site, which does contain the <video>-element  by itself,  without requiring
  62 # js to be ran.
  63 class Pr0grammIE(InfoExtractor):
  64     # Possible urls:
  65     # https://pr0gramm.com/new/546637
  66     # https://pr0gramm.com/new/video/546637
  67     # https://pr0gramm.com/top/546637
  68     # https://pr0gramm.com/top/video/546637
  69     # https://pr0gramm.com/user/g11st/uploads/5466437
  70     # https://pr0gramm.com/user/froschler/dafur-ist-man-hier/5091290
  71     # https://pr0gramm.com/user/froschler/reinziehen-1elf/5232030
  72     # https://pr0gramm.com/user/froschler/1elf/5232030
  73     # https://pr0gramm.com/new/5495710:comment62621020 <- this is not the id!
  74     # https://pr0gramm.com/top/fruher war alles damals/5498175
  75
  76     _VALID_URL = r'https?:\/\/pr0gramm\.com\/(?!static/\d+).+?\/(?P<id>[\d]+)(:|$)'
  77     _TEST = {
  78         'url': 'https://pr0gramm.com/new/video/5466437',
  79         'info_dict': {
  80             'id': '5466437',
  81             'ext': 'mp4',
  82             'title': 'pr0gramm-5466437 by g11st',
  83             'uploader': 'g11st',
  84             'upload_date': '20221221',
  85         }
  86     }
  87
  88     def _generic_title():
  89         return "oof"
  90
  91     def _real_extract(self, url):
  92         video_id = self._match_id(url)
  93
  94         return self.url_result(
  95             'https://pr0gramm.com/static/' + video_id,
  96             video_id=video_id,
  97             ie=Pr0grammStaticIE.ie_key())