yt_dlp/extractor/pr0gramm.py

   1 import datetime as dt
   2 import functools
   3 import json
   4 import urllib.parse
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     float_or_none,
  10     int_or_none,
  11     make_archive_id,
  12     mimetype2ext,
  13     str_or_none,
  14     urljoin,
  15 )
  16 from ..utils.traversal import traverse_obj
  17
  18
  19 class Pr0grammIE(InfoExtractor):
  20     _VALID_URL = r'https?://pr0gramm\.com\/(?:[^/?#]+/)+(?P<id>[\d]+)(?:[/?#:]|$)'
  21     _TESTS = [{
  22         'url': 'https://pr0gramm.com/new/video/5466437',
  23         'info_dict': {
  24             'id': '5466437',
  25             'ext': 'mp4',
  26             'title': 'pr0gramm-5466437 by g11st',
  27             'tags': ['Neon Genesis Evangelion', 'Touhou Project', 'Fly me to the Moon', 'Marisad', 'Marisa Kirisame', 'video', 'sound', 'Marisa', 'Anime'],
  28             'uploader': 'g11st',
  29             'uploader_id': '394718',
  30             'timestamp': 1671590240,
  31             'upload_date': '20221221',
  32             'like_count': int,
  33             'dislike_count': int,
  34             'age_limit': 0,
  35             'thumbnail': r're:^https://thumb\.pr0gramm\.com/.*\.jpg',
  36             '_old_archive_ids': ['pr0grammstatic 5466437'],
  37         },
  38     }, {
  39         'url': 'https://pr0gramm.com/new/3052805:comment28391322',
  40         'info_dict': {
  41             'id': '3052805',
  42             'ext': 'mp4',
  43             'title': 'pr0gramm-3052805 by Hansking1',
  44             'tags': 'count:15',
  45             'uploader': 'Hansking1',
  46             'uploader_id': '385563',
  47             'timestamp': 1552930408,
  48             'upload_date': '20190318',
  49             'like_count': int,
  50             'dislike_count': int,
  51             'age_limit': 0,
  52             'thumbnail': r're:^https://thumb\.pr0gramm\.com/.*\.jpg',
  53             '_old_archive_ids': ['pr0grammstatic 3052805'],
  54         },
  55     }, {
  56         # Requires verified account
  57         'url': 'https://pr0gramm.com/new/Gianna%20Michaels/5848332',
  58         'info_dict': {
  59             'id': '5848332',
  60             'ext': 'mp4',
  61             'title': 'pr0gramm-5848332 by erd0pfel',
  62             'tags': 'count:18',
  63             'uploader': 'erd0pfel',
  64             'uploader_id': '349094',
  65             'timestamp': 1694489652,
  66             'upload_date': '20230912',
  67             'like_count': int,
  68             'dislike_count': int,
  69             'age_limit': 18,
  70             'thumbnail': r're:^https://thumb\.pr0gramm\.com/.*\.jpg',
  71             '_old_archive_ids': ['pr0grammstatic 5848332'],
  72         },
  73     }, {
  74         'url': 'https://pr0gramm.com/top/5895149',
  75         'info_dict': {
  76             'id': '5895149',
  77             'ext': 'mp4',
  78             'title': 'pr0gramm-5895149 by algoholigSeeManThrower',
  79             'tags': 'count:19',
  80             'uploader': 'algoholigSeeManThrower',
  81             'uploader_id': '457556',
  82             'timestamp': 1697580902,
  83             'upload_date': '20231018',
  84             'like_count': int,
  85             'dislike_count': int,
  86             'age_limit': 0,
  87             'thumbnail': 'https://thumb.pr0gramm.com/2023/10/18/db47bb3db5e1a1b3.jpg',
  88             '_old_archive_ids': ['pr0grammstatic 5895149'],
  89         },
  90     }, {
  91         'url': 'https://pr0gramm.com/static/5466437',
  92         'only_matching': True,
  93     }, {
  94         'url': 'https://pr0gramm.com/new/rowan%20atkinson%20herr%20bohne/3052805',
  95         'only_matching': True,
  96     }, {
  97         'url': 'https://pr0gramm.com/user/froschler/dafur-ist-man-hier/5091290',
  98         'only_matching': True,
  99     }]
 100
 101     BASE_URL = 'https://pr0gramm.com'
 102
 103     @functools.cached_property
 104     def _is_logged_in(self):
 105         return 'pp' in self._get_cookies(self.BASE_URL)
 106
 107     @functools.cached_property
 108     def _maximum_flags(self):
 109         # We need to guess the flags for the content otherwise the api will raise an error
 110         # We can guess the maximum allowed flags for the account from the cookies
 111         # Bitflags are (msbf): pol, nsfp, nsfl, nsfw, sfw
 112         flags = 0b10001
 113         if self._is_logged_in:
 114             flags |= 0b01000
 115             cookies = self._get_cookies(self.BASE_URL)
 116             if 'me' not in cookies:
 117                 self._download_webpage(self.BASE_URL, None, 'Refreshing verification information')
 118             if traverse_obj(cookies, ('me', {lambda x: x.value}, {urllib.parse.unquote}, {json.loads}, 'verified')):
 119                 flags |= 0b00110
 120
 121         return flags
 122
 123     def _call_api(self, endpoint, video_id, query={}, note='Downloading API json'):
 124         data = self._download_json(
 125             f'https://pr0gramm.com/api/items/{endpoint}',
 126             video_id, note, query=query, expected_status=403)
 127
 128         error = traverse_obj(data, ('error', {str}))
 129         if error in ('nsfwRequired', 'nsflRequired', 'nsfpRequired', 'verificationRequired'):
 130             if not self._is_logged_in:
 131                 self.raise_login_required()
 132             raise ExtractorError(f'Unverified account cannot access NSFW/NSFL ({error})', expected=True)
 133         elif error:
 134             message = traverse_obj(data, ('msg', {str})) or error
 135             raise ExtractorError(f'API returned error: {message}', expected=True)
 136
 137         return data
 138
 139     @staticmethod
 140     def _create_source_url(path):
 141         return urljoin('https://img.pr0gramm.com', path)
 142
 143     def _real_extract(self, url):
 144         video_id = self._match_id(url)
 145         video_info = traverse_obj(
 146             self._call_api('get', video_id, {'id': video_id, 'flags': self._maximum_flags}),
 147             ('items', 0, {dict}))
 148
 149         source = video_info.get('image')
 150         if not source or not source.endswith('mp4'):
 151             self.raise_no_formats('Could not extract a video', expected=bool(source), video_id=video_id)
 152
 153         metadata = self._call_api('info', video_id, {'itemId': video_id}, note='Downloading tags')
 154         tags = traverse_obj(metadata, ('tags', ..., 'tag', {str}))
 155         # Sorted by "confidence", higher confidence = earlier in list
 156         confidences = traverse_obj(metadata, ('tags', ..., 'confidence', ({int}, {float})))
 157         if confidences:
 158             tags = [tag for _, tag in sorted(zip(confidences, tags), reverse=True)]
 159
 160         formats = traverse_obj(video_info, ('variants', ..., {
 161             'format_id': ('name', {str}),
 162             'url': ('path', {self._create_source_url}),
 163             'ext': ('mimeType', {mimetype2ext}),
 164             'vcodec': ('codec', {str}),
 165             'width': ('width', {int_or_none}),
 166             'height': ('height', {int_or_none}),
 167             'bitrate': ('bitRate', {float_or_none}),
 168             'filesize': ('fileSize', {int_or_none}),
 169         })) if video_info.get('variants') else [{
 170             'ext': 'mp4',
 171             'format_id': 'source',
 172             **traverse_obj(video_info, {
 173                 'url': ('image', {self._create_source_url}),
 174                 'width': ('width', {int_or_none}),
 175                 'height': ('height', {int_or_none}),
 176             }),
 177         }]
 178
 179         subtitles = {}
 180         for subtitle in traverse_obj(video_info, ('subtitles', lambda _, v: v['language'])):
 181             subtitles.setdefault(subtitle['language'], []).append(traverse_obj(subtitle, {
 182                 'url': ('path', {self._create_source_url}),
 183                 'note': ('label', {str}),
 184             }))
 185
 186         return {
 187             'id': video_id,
 188             'title': f'pr0gramm-{video_id} by {video_info.get("user")}',
 189             'tags': tags,
 190             'formats': formats,
 191             'subtitles': subtitles,
 192             'age_limit': 18 if traverse_obj(video_info, ('flags', {0b110.__and__})) else 0,
 193             '_old_archive_ids': [make_archive_id('Pr0grammStatic', video_id)],
 194             **traverse_obj(video_info, {
 195                 'uploader': ('user', {str}),
 196                 'uploader_id': ('userId', {str_or_none}),
 197                 'like_count': ('up', {int}),
 198                 'dislike_count': ('down', {int}),
 199                 'timestamp': ('created', {int}),
 200                 'upload_date': ('created', {int}, {dt.date.fromtimestamp}, {lambda x: x.strftime('%Y%m%d')}),
 201                 'thumbnail': ('thumb', {lambda x: urljoin('https://thumb.pr0gramm.com', x)}),
 202             }),
 203         }