Plugin cleanup and tweaks

plugins/youtube_download/yt_dlp/extractor/cda.py

@@ -1,16 +1,14 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
+import base64
 import codecs
-import re
+import datetime
+import hashlib
+import hmac
 import json
+import random
+import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_chr,
-    compat_ord,
-    compat_urllib_parse_unquote,
-)
+from ..compat import compat_ord, compat_urllib_parse_unquote
 from ..utils import (
     ExtractorError,
     float_or_none,
@@ -19,14 +17,26 @@ from ..utils import (
     multipart_encode,
     parse_duration,
     random_birthday,
-    urljoin,
+    traverse_obj,
+    try_call,
     try_get,
+    urljoin,
 )
 
 
 class CDAIE(InfoExtractor):
     _VALID_URL = r'https?://(?:(?:www\.)?cda\.pl/video|ebd\.cda\.pl/[0-9]+x[0-9]+)/(?P<id>[0-9a-z]+)'
-    _BASE_URL = 'http://www.cda.pl/'
+    _NETRC_MACHINE = 'cdapl'
+
+    _BASE_URL = 'https://www.cda.pl'
+    _BASE_API_URL = 'https://api.cda.pl'
+    _API_HEADERS = {
+        'Accept': 'application/vnd.cda.public+json',
+    }
+    # hardcoded in the app
+    _LOGIN_REQUEST_AUTH = 'Basic YzU3YzBlZDUtYTIzOC00MWQwLWI2NjQtNmZmMWMxY2Y2YzVlOklBTm95QlhRRVR6U09MV1hnV3MwMW0xT2VyNWJNZzV4clRNTXhpNGZJUGVGZ0lWUlo5UGVYTDhtUGZaR1U1U3Q'
+    _BEARER_CACHE = 'cda-bearer'
+
     _TESTS = [{
         'url': 'http://www.cda.pl/video/5749950c',
         'md5': '6f844bf51b15f31fae165365707ae970',
@@ -90,14 +100,110 @@ class CDAIE(InfoExtractor):
                 'Content-Type': content_type,
             }, **kwargs)
 
+    def _perform_login(self, username, password):
+        app_version = random.choice((
+            '1.2.88 build 15306',
+            '1.2.174 build 18469',
+        ))
+        android_version = random.randrange(8, 14)
+        phone_model = random.choice((
+            # x-kom.pl top selling Android smartphones, as of 2022-12-26
+            # https://www.x-kom.pl/g-4/c/1590-smartfony-i-telefony.html?f201-system-operacyjny=61322-android
+            'ASUS ZenFone 8',
+            'Motorola edge 20 5G',
+            'Motorola edge 30 neo 5G',
+            'Motorola moto g22',
+            'OnePlus Nord 2T 5G',
+            'Samsung Galaxy A32 SM‑A325F',
+            'Samsung Galaxy M13',
+            'Samsung Galaxy S20 FE 5G',
+            'Xiaomi 11T',
+            'Xiaomi POCO M4 Pro',
+            'Xiaomi Redmi 10',
+            'Xiaomi Redmi 10C',
+            'Xiaomi Redmi 9C NFC',
+            'Xiaomi Redmi Note 10 Pro',
+            'Xiaomi Redmi Note 11 Pro',
+            'Xiaomi Redmi Note 11',
+            'Xiaomi Redmi Note 11S 5G',
+            'Xiaomi Redmi Note 11S',
+            'realme 10',
+            'realme 9 Pro+',
+            'vivo Y33s',
+        ))
+        self._API_HEADERS['User-Agent'] = f'pl.cda 1.0 (version {app_version}; Android {android_version}; {phone_model})'
+
+        cached_bearer = self.cache.load(self._BEARER_CACHE, username) or {}
+        if cached_bearer.get('valid_until', 0) > datetime.datetime.now().timestamp() + 5:
+            self._API_HEADERS['Authorization'] = f'Bearer {cached_bearer["token"]}'
+            return
+
+        password_hash = base64.urlsafe_b64encode(hmac.new(
+            b's01m1Oer5IANoyBXQETzSOLWXgWs01m1Oer5bMg5xrTMMxRZ9Pi4fIPeFgIVRZ9PeXL8mPfXQETZGUAN5StRZ9P',
+            ''.join(f'{bytes((bt & 255, )).hex():0>2}'
+                    for bt in hashlib.md5(password.encode()).digest()).encode(),
+            hashlib.sha256).digest()).decode().replace('=', '')
+
+        token_res = self._download_json(
+            f'{self._BASE_API_URL}/oauth/token', None, 'Logging in', data=b'',
+            headers={**self._API_HEADERS, 'Authorization': self._LOGIN_REQUEST_AUTH},
+            query={
+                'grant_type': 'password',
+                'login': username,
+                'password': password_hash,
+            })
+        self.cache.store(self._BEARER_CACHE, username, {
+            'token': token_res['access_token'],
+            'valid_until': token_res['expires_in'] + datetime.datetime.now().timestamp(),
+        })
+        self._API_HEADERS['Authorization'] = f'Bearer {token_res["access_token"]}'
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
+
+        if 'Authorization' in self._API_HEADERS:
+            return self._api_extract(video_id)
+        else:
+            return self._web_extract(video_id, url)
+
+    def _api_extract(self, video_id):
+        meta = self._download_json(
+            f'{self._BASE_API_URL}/video/{video_id}', video_id, headers=self._API_HEADERS)['video']
+
+        uploader = traverse_obj(meta, 'author', 'login')
+
+        formats = [{
+            'url': quality['file'],
+            'format': quality.get('title'),
+            'resolution': quality.get('name'),
+            'height': try_call(lambda: int(quality['name'][:-1])),
+            'filesize': quality.get('length'),
+        } for quality in meta['qualities'] if quality.get('file')]
+
+        if meta.get('premium') and not meta.get('premium_free') and not formats:
+            raise ExtractorError(
+                'Video requires CDA Premium - subscription needed', expected=True)
+
+        return {
+            'id': video_id,
+            'title': meta.get('title'),
+            'description': meta.get('description'),
+            'uploader': None if uploader == 'anonim' else uploader,
+            'average_rating': float_or_none(meta.get('rating')),
+            'thumbnail': meta.get('thumb'),
+            'formats': formats,
+            'duration': meta.get('duration'),
+            'age_limit': 18 if meta.get('for_adults') else 0,
+            'view_count': meta.get('views'),
+        }
+
+    def _web_extract(self, video_id, url):
         self._set_cookie('cda.pl', 'cda.player', 'html5')
         webpage = self._download_webpage(
-            self._BASE_URL + '/video/' + video_id, video_id)
+            f'{self._BASE_URL}/video/{video_id}/vfilm', video_id)
 
         if 'Ten film jest dostępny dla użytkowników premium' in webpage:
-            raise ExtractorError('This video is only available for premium users.', expected=True)
+            self.raise_login_required('This video is only available for premium users')
 
         if re.search(r'niedostępn[ey] w(?:&nbsp;|\s+)Twoim kraju\s*<', webpage):
             self.raise_geo_restricted()
@@ -147,7 +253,7 @@ class CDAIE(InfoExtractor):
             b = []
             for c in a:
                 f = compat_ord(c)
-                b.append(compat_chr(33 + (f + 14) % 94) if 33 <= f <= 126 else compat_chr(f))
+                b.append(chr(33 + (f + 14) % 94) if 33 <= f <= 126 else chr(f))
             a = ''.join(b)
             a = a.replace('.cda.mp4', '')
             for p in ('.2cda.pl', '.3cda.pl'):
@@ -229,6 +335,4 @@ class CDAIE(InfoExtractor):
 
             extract_format(webpage, resolution)
 
-        self._sort_formats(formats)
-
         return merge_dicts(info_dict, info)