[extractor/substack] Add extractor (#4011)

Closes #3722
Authored by: elyse0

yt_dlp/extractor/_extractors.py

@@ -1640,6 +1640,7 @@
 from .stretchinternet import StretchInternetIE
 from .stripchat import StripchatIE
 from .stv import STVPlayerIE
+from .substack import SubstackIE
 from .sunporno import SunPornoIE
 from .sverigesradio import (
     SverigesRadioEpisodeIE,

yt_dlp/extractor/generic.py

@@ -69,6 +69,7 @@
 from .sportbox import SportBoxIE
 from .spotify import SpotifyBaseIE
 from .springboardplatform import SpringboardPlatformIE
+from .substack import SubstackIE
 from .svt import SVTIE
 from .teachable import TeachableIE
 from .ted import TedEmbedIE
@@ -2542,7 +2543,34 @@ class GenericIE(InfoExtractor):
                 'timestamp': 1652833414,
                 'age_limit': 0,
             }
-        }, {
+        },
+        {
+            'url': 'https://www.mollymovieclub.com/p/interstellar?s=r#details',
+            'md5': '198bde8bed23d0b23c70725c83c9b6d9',
+            'info_dict': {
+                'id': '53602801',
+                'ext': 'mpga',
+                'title': 'Interstellar',
+                'description': 'Listen now | Episode One',
+                'thumbnail': 'md5:c30d9c83f738e16d8551d7219d321538',
+                'uploader': 'Molly Movie Club',
+                'uploader_id': '839621',
+            },
+        },
+        {
+            'url': 'https://www.blockedandreported.org/p/episode-117-lets-talk-about-depp?s=r',
+            'md5': 'c0cc44ee7415daeed13c26e5b56d6aa0',
+            'info_dict': {
+                'id': '57962052',
+                'ext': 'mpga',
+                'title': 'md5:855b2756f0ee10f6723fa00b16266f8d',
+                'description': 'md5:fe512a5e94136ad260c80bde00ea4eef',
+                'thumbnail': 'md5:2218f27dfe517bb5ac16c47d0aebac59',
+                'uploader': 'Blocked and Reported',
+                'uploader_id': '500230',
+            },
+        },
+        {
             'url': 'https://www.skimag.com/video/ski-people-1980/',
             'info_dict': {
                 'id': 'ski-people-1980',
@@ -3107,6 +3135,11 @@ def _real_extract(self, url):
             # Don't set the extractor because it can be a track url or an album
             return self.url_result(burl)
 
+        # Check for Substack custom domains
+        substack_url = SubstackIE._extract_url(webpage, url)
+        if substack_url:
+            return self.url_result(substack_url, SubstackIE)
+
         # Look for embedded Vevo player
         mobj = re.search(
             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)

yt_dlp/extractor/substack.py

@@ -0,0 +1,100 @@
+import re
+import urllib.parse
+
+from .common import InfoExtractor
+from ..utils import str_or_none, traverse_obj
+
+
+class SubstackIE(InfoExtractor):
+    _VALID_URL = r'https?://(?P<username>[\w-]+)\.substack\.com/p/(?P<id>[\w-]+)'
+    _TESTS = [{
+        'url': 'https://haleynahman.substack.com/p/i-made-a-vlog?s=r',
+        'md5': 'f27e4fc6252001d48d479f45e65cdfd5',
+        'info_dict': {
+            'id': '47660949',
+            'ext': 'mp4',
+            'title': 'I MADE A VLOG',
+            'description': 'md5:10c01ff93439a62e70ce963b2aa0b7f6',
+            'thumbnail': 'md5:bec758a34d8ee9142d43bcebdf33af18',
+            'uploader': 'Maybe Baby',
+            'uploader_id': '33628',
+        }
+    }, {
+        'url': 'https://haleynahman.substack.com/p/-dear-danny-i-found-my-boyfriends?s=r',
+        'md5': '0a63eacec877a1171a62cfa69710fcea',
+        'info_dict': {
+            'id': '51045592',
+            'ext': 'mpga',
+            'title': "🎧 Dear Danny: I found my boyfriend's secret Twitter account",
+            'description': 'md5:a57f2439319e56e0af92dd0c95d75797',
+            'thumbnail': 'md5:daa40b6b79249417c14ff8103db29639',
+            'uploader': 'Maybe Baby',
+            'uploader_id': '33628',
+        }
+    }, {
+        'url': 'https://andrewzimmern.substack.com/p/mussels-with-black-bean-sauce-recipe',
+        'md5': 'fd3c07077b02444ff0130715b5f632bb',
+        'info_dict': {
+            'id': '47368578',
+            'ext': 'mp4',
+            'title': 'Mussels with Black Bean Sauce: Recipe of the Week #7',
+            'description': 'md5:b96234a2906c7d854d5229818d889515',
+            'thumbnail': 'md5:e30bfaa9da40e82aa62354263a9dd232',
+            'uploader': "Andrew Zimmern's Spilled Milk ",
+            'uploader_id': '577659',
+        }
+    }]
+
+    @classmethod
+    def _extract_url(cls, webpage, url):
+        if not re.search(r'<script[^>]+src=["\']https://substackcdn.com/[^"\']+\.js', webpage):
+            return
+
+        mobj = re.search(r'{[^}]*["\']subdomain["\']\s*:\s*["\'](?P<subdomain>[^"]+)', webpage)
+        if mobj:
+            parsed = urllib.parse.urlparse(url)
+            return parsed._replace(netloc=f'{mobj.group("subdomain")}.substack.com').geturl()
+
+    def _extract_video_formats(self, video_id, username):
+        formats, subtitles = [], {}
+        for video_format in ('hls', 'mp4'):
+            video_url = f'https://{username}.substack.com/api/v1/video/upload/{video_id}/src?type={video_format}'
+
+            if video_format == 'hls':
+                fmts, subs = self._extract_m3u8_formats_and_subtitles(video_url, video_id, 'mp4', fatal=False)
+                formats.extend(fmts)
+                self._merge_subtitles(subs, target=subtitles)
+            else:
+                formats.append({
+                    'url': video_url,
+                    'ext': video_format,
+                })
+
+        return formats, subtitles
+
+    def _real_extract(self, url):
+        display_id, username = self._match_valid_url(url).group('id', 'username')
+        webpage = self._download_webpage(url, display_id)
+
+        webpage_info = self._search_json(r'<script[^>]*>\s*window\._preloads\s*=', webpage, 'preloads', display_id)
+
+        post_type = webpage_info['post']['type']
+        formats, subtitles = [], {}
+        if post_type == 'podcast':
+            formats, subtitles = [{'url': webpage_info['post']['podcast_url']}], {}
+        elif post_type == 'video':
+            formats, subtitles = self._extract_video_formats(webpage_info['post']['videoUpload']['id'], username)
+        else:
+            self.raise_no_formats(f'Page type "{post_type}" is not supported')
+
+        self._sort_formats(formats)
+        return {
+            'id': str(webpage_info['post']['id']),
+            'formats': formats,
+            'subtitles': subtitles,
+            'title': traverse_obj(webpage_info, ('post', 'title')),
+            'description': traverse_obj(webpage_info, ('post', 'description')),
+            'thumbnail': traverse_obj(webpage_info, ('post', 'cover_image')),
+            'uploader': traverse_obj(webpage_info, ('pub', 'name')),
+            'uploader_id': str_or_none(traverse_obj(webpage_info, ('post', 'publication_id'))),
+        }