Merge 440238ebc9 into 4b5eec0aaa

Fix bug in 720b3dc453

Closes #11623
Authored by: jkruse

yt_dlp/extractor/chaturbate.py

@@ -59,16 +59,15 @@ def _extract_from_api(self, video_id, tld):
                 'Accept': 'application/json',
             }, fatal=False, impersonate=True) or {}
 
-        status = response.get('room_status')
-        if status != 'public':
-            if error := self._ERROR_MAP.get(status):
-                raise ExtractorError(error, expected=True)
-            self.report_warning('Falling back to webpage extraction')
-            return None
-
         m3u8_url = response.get('url')
         if not m3u8_url:
+            status = response.get('room_status')
+            if error := self._ERROR_MAP.get(status):
+                raise ExtractorError(error, expected=True)
+            if status == 'public':
                 self.raise_geo_restricted()
+            self.report_warning(f'Got status "{status}" from API; falling back to webpage extraction')
+            return None
 
         return {
             'id': video_id,

yt_dlp/extractor/lbry.py

@@ -78,7 +78,7 @@ def _parse_stream(self, stream, url):
 
         return info
 
-    def _fetch_page(self, display_id, url, params, page):
+    def _fetch_page(self, display_id, url, params, metapage_nr, page):
         page += 1
         page_params = {
             'no_totals': True,
@@ -86,8 +86,12 @@ def _fetch_page(self, display_id, url, params, page):
             'page_size': self._PAGE_SIZE,
             **params,
         }
+        if metapage_nr == 0:
+            resource = f'page {page}'
+        else:
+            resource = f'page {metapage_nr+1}_{page}'
         result = self._call_api_proxy(
-            'claim_search', display_id, page_params, f'page {page}')
+            'claim_search', display_id, page_params, resource)
         for item in traverse_obj(result, ('items', lambda _, v: v['name'] and v['claim_id'])):
             yield {
                 **self._parse_stream(item, url),
@@ -96,6 +100,32 @@ def _fetch_page(self, display_id, url, params, page):
                 'url': self._permanent_url(url, item['name'], item['claim_id']),
             }
 
+    def _metapage_entries(self, display_id, url, params):
+        if 'release_time' in params:
+            raise ExtractorError('release_time isn\'t allowed because _metapage_entires needs to specify it.')
+        if not ('order_by' in params and params['order_by'] == ['release_time']):
+            raise ExtractorError('videos must be sorted by release_time for _metapage_entries to work.')
+
+        last_metapage = []
+        metapage = OnDemandPagedList(
+            functools.partial(self._fetch_page, display_id, url, params, 0),
+            self._PAGE_SIZE).getslice()
+
+        metapage_nr = 1
+        while len(metapage) > 0:
+            yield from metapage
+
+            next_metapage_params = {
+                **params,
+                'release_time': '<={}'.format(metapage[-1]['release_timestamp']),
+            }
+            last_metapage = metapage
+            metapage = OnDemandPagedList(
+                functools.partial(self._fetch_page, display_id, url, next_metapage_params, metapage_nr),
+                self._PAGE_SIZE).getslice()
+            metapage = [x for x in metapage if x not in last_metapage]
+            metapage_nr += 1
+
     def _playlist_entries(self, url, display_id, claim_param, metadata):
         qs = parse_qs(url)
         content = qs.get('content', [None])[0]
@@ -123,8 +153,12 @@ def _playlist_entries(self, url, display_id, claim_param, metadata):
                 languages.append('none')
             params['any_languages'] = languages
 
+        if qs.get('order', ['new'])[0] == 'new':
+            entries = self._metapage_entries(display_id, url, params)
+        else:
+            self.report_warning('Extraction is limited to 1000 Videos when not sorting by newest.')
             entries = OnDemandPagedList(
-            functools.partial(self._fetch_page, display_id, url, params),
+                functools.partial(self._fetch_page, display_id, url, params, 0),
                 self._PAGE_SIZE)
 
         return self.playlist_result(