yt_dlp/extractor/pr0gramm.py

   1 import json
   2 from datetime import date
   3 from urllib.parse import unquote
   4
   5 from .common import InfoExtractor
   6 from ..compat import functools
   7 from ..utils import ExtractorError, make_archive_id, urljoin
   8 from ..utils.traversal import traverse_obj
   9
  10
  11 class Pr0grammIE(InfoExtractor):
  12     _VALID_URL = r'https?://pr0gramm\.com\/(?:[^/?#]+/)+(?P<id>[\d]+)(?:[/?#:]|$)'
  13     _TESTS = [{
  14         # Tags require account
  15         'url': 'https://pr0gramm.com/new/video/5466437',
  16         'info_dict': {
  17             'id': '5466437',
  18             'ext': 'mp4',
  19             'title': 'pr0gramm-5466437 by g11st',
  20             'tags': ['Neon Genesis Evangelion', 'Touhou Project', 'Fly me to the Moon', 'Marisad', 'Marisa Kirisame', 'video', 'sound', 'Marisa', 'Anime'],
  21             'uploader': 'g11st',
  22             'uploader_id': 394718,
  23             'upload_timestamp': 1671590240,
  24             'upload_date': '20221221',
  25             'like_count': int,
  26             'dislike_count': int,
  27             'age_limit': 0,
  28             'thumbnail': r're:^https://thumb\.pr0gramm\.com/.*\.jpg',
  29         },
  30     }, {
  31         # Tags require account
  32         'url': 'https://pr0gramm.com/new/3052805:comment28391322',
  33         'info_dict': {
  34             'id': '3052805',
  35             'ext': 'mp4',
  36             'title': 'pr0gramm-3052805 by Hansking1',
  37             'tags': 'count:15',
  38             'uploader': 'Hansking1',
  39             'uploader_id': 385563,
  40             'upload_timestamp': 1552930408,
  41             'upload_date': '20190318',
  42             'like_count': int,
  43             'dislike_count': int,
  44             'age_limit': 0,
  45             'thumbnail': r're:^https://thumb\.pr0gramm\.com/.*\.jpg',
  46         },
  47     }, {
  48         # Requires verified account
  49         'url': 'https://pr0gramm.com/new/Gianna%20Michaels/5848332',
  50         'info_dict': {
  51             'id': '5848332',
  52             'ext': 'mp4',
  53             'title': 'pr0gramm-5848332 by erd0pfel',
  54             'tags': 'count:18',
  55             'uploader': 'erd0pfel',
  56             'uploader_id': 349094,
  57             'upload_timestamp': 1694489652,
  58             'upload_date': '20230912',
  59             'like_count': int,
  60             'dislike_count': int,
  61             'age_limit': 18,
  62             'thumbnail': r're:^https://thumb\.pr0gramm\.com/.*\.jpg',
  63         },
  64     }, {
  65         'url': 'https://pr0gramm.com/static/5466437',
  66         'only_matching': True,
  67     }, {
  68         'url': 'https://pr0gramm.com/new/rowan%20atkinson%20herr%20bohne/3052805',
  69         'only_matching': True,
  70     }, {
  71         'url': 'https://pr0gramm.com/user/froschler/dafur-ist-man-hier/5091290',
  72         'only_matching': True,
  73     }]
  74
  75     BASE_URL = 'https://pr0gramm.com'
  76
  77     @functools.cached_property
  78     def _is_logged_in(self):
  79         return 'pp' in self._get_cookies(self.BASE_URL)
  80
  81     @functools.cached_property
  82     def _maximum_flags(self):
  83         # We need to guess the flags for the content otherwise the api will raise an error
  84         # We can guess the maximum allowed flags for the account from the cookies
  85         # Bitflags are (msbf): nsfp, nsfl, nsfw, sfw
  86         flags = 0b0001
  87         if self._is_logged_in:
  88             flags |= 0b1000
  89             cookies = self._get_cookies(self.BASE_URL)
  90             if 'me' not in cookies:
  91                 self._download_webpage(self.BASE_URL, None, 'Refreshing verification information')
  92             if traverse_obj(cookies, ('me', {lambda x: x.value}, {unquote}, {json.loads}, 'verified')):
  93                 flags |= 0b0110
  94
  95         return flags
  96
  97     def _call_api(self, endpoint, video_id, query={}, note='Downloading API json'):
  98         data = self._download_json(
  99             f'https://pr0gramm.com/api/items/{endpoint}',
 100             video_id, note, query=query, expected_status=403)
 101
 102         error = traverse_obj(data, ('error', {str}))
 103         if error in ('nsfwRequired', 'nsflRequired', 'nsfpRequired', 'verificationRequired'):
 104             if not self._is_logged_in:
 105                 self.raise_login_required()
 106             raise ExtractorError(f'Unverified account cannot access NSFW/NSFL ({error})', expected=True)
 107         elif error:
 108             message = traverse_obj(data, ('msg', {str})) or error
 109             raise ExtractorError(f'API returned error: {message}', expected=True)
 110
 111         return data
 112
 113     def _real_extract(self, url):
 114         video_id = self._match_id(url)
 115         video_info = traverse_obj(
 116             self._call_api('get', video_id, {'id': video_id, 'flags': self._maximum_flags}),
 117             ('items', 0, {dict}))
 118
 119         source = urljoin('https://img.pr0gramm.com', video_info.get('image'))
 120         if not source or not source.endswith('mp4'):
 121             self.raise_no_formats('Could not extract a video', expected=bool(source), video_id=video_id)
 122
 123         tags = None
 124         if self._is_logged_in:
 125             metadata = self._call_api('info', video_id, {'itemId': video_id})
 126             tags = traverse_obj(metadata, ('tags', ..., 'tag', {str}))
 127             # Sorted by "confidence", higher confidence = earlier in list
 128             confidences = traverse_obj(metadata, ('tags', ..., 'confidence', ({int}, {float})))
 129             if confidences:
 130                 tags = [tag for _, tag in sorted(zip(confidences, tags), reverse=True)]
 131
 132         return {
 133             'id': video_id,
 134             'title': f'pr0gramm-{video_id} by {video_info.get("user")}',
 135             'formats': [{
 136                 'url': source,
 137                 'ext': 'mp4',
 138                 **traverse_obj(video_info, {
 139                     'width': ('width', {int}),
 140                     'height': ('height', {int}),
 141                 }),
 142             }],
 143             'tags': tags,
 144             'age_limit': 18 if traverse_obj(video_info, ('flags', {0b110.__and__})) else 0,
 145             '_old_archive_ids': [make_archive_id('Pr0grammStatic', video_id)],
 146             **traverse_obj(video_info, {
 147                 'uploader': ('user', {str}),
 148                 'uploader_id': ('userId', {int}),
 149                 'like_count': ('up', {int}),
 150                 'dislike_count': ('down', {int}),
 151                 'upload_timestamp': ('created', {int}),
 152                 'upload_date': ('created', {int}, {date.fromtimestamp}, {lambda x: x.strftime('%Y%m%d')}),
 153                 'thumbnail': ('thumb', {lambda x: urljoin('https://thumb.pr0gramm.com', x)})
 154             }),
 155         }