mock youtube.py

.github/workflows/update.yml

@@ -34,14 +34,15 @@ jobs:
             yt_dlp/extractor/commonmistakes.py
             yt_dlp/extractor/commonprotocols.py
             yt_dlp/extractor/extractors.py
+            yt_dlp/extractor/generic.py
             yt_dlp/extractor/openload.py
             yt_dlp/extractor/tver.py
         
       - name: Check status
         run: |
           #don't check modded files
           git update-index --assume-unchanged lib/yt_dlp/extractor/_extractors.py
-          git update-index --assume-unchanged lib/yt_dlp/extractor/generic.py
+          git update-index --assume-unchanged lib/yt_dlp/extractor/youtube.py
           
           if [[ -z $(git status --porcelain) ]]; then
             echo "clean=true" >> $GITHUB_ENV
@@ -59,7 +60,7 @@ jobs:
           
           #don't track modded files
           git update-index --assume-unchanged lib/yt_dlp/extractor/_extractors.py
-          git update-index --assume-unchanged lib/yt_dlp/extractor/generic.py
+          git update-index --assume-unchanged lib/yt_dlp/extractor/youtube.py
           
           git add -A 
           git commit -m "yt_dlp: ${{ env.commit_upstream }}"

README.md

@@ -15,7 +15,7 @@ _By removing all extractors that are irrelevant to TVer.jp, yt-dlp-tver not only
 If this approach ultimately proves to be a poor decision, I can still take solace in the fact that I gained valuable knowledge about GitHub Actions along the way.
 
 ## Modifications
-Besides removing all extractors that aren't necessary for extracting a valid stream manifest URL from TVer.jp, there had to be made some minor changes to [_extractors.py](lib/yt_dlp/extractor/_extractors.py) and (sadly) [generic.py](https://gist.github.com/kuriho/6030dbda5bb07dc8f307d78d2770457e/revisions?diff=unified). 
+Besides removing all extractors that aren't necessary for extracting a valid stream manifest URL from TVer.jp, there had to be made some minor changes to [_extractors.py](lib/yt_dlp/extractor/_extractors.py) and [youtube.py](lib/yt_dlp/extractor/youtube.py). 
 
 ## CI pipeline
 <img src="https://github.com/kuriho/script.module.yt-dlp-tver/blob/master/media/ci.png?raw=true" alt="drawing" width="600"/>

lib/yt_dlp/extractor/generic.py

@@ -6,6 +6,7 @@
 
 from .common import InfoExtractor  # isort: split
 from .commonprotocols import RtmpIE
+from .youtube import YoutubeIE
 from ..compat import compat_etree_fromstring
 from ..utils import (
     KNOWN_EXTENSIONS,
@@ -2634,6 +2635,8 @@ def _extract_embeds(self, url, webpage, *, urlh=None, info_dict={}):
                 if isinstance(src_type, str):
                     src_type = src_type.lower()
                 ext = determine_ext(src).lower()
+                if src_type == 'video/youtube':
+                    return [self.url_result(src, YoutubeIE.ie_key())]
                 if src_type == 'application/dash+xml' or ext == 'mpd':
                     fmts, subs = self._extract_mpd_formats_and_subtitles(
                         src, video_id, mpd_id='dash', fatal=False)
@@ -2703,6 +2706,8 @@ def _extract_embeds(self, url, webpage, *, urlh=None, info_dict={}):
             }, json_ld)]
 
         def check_video(vurl):
+            if YoutubeIE.suitable(vurl):
+                return True
             if RtmpIE.suitable(vurl):
                 return True
             vpath = urllib.parse.urlparse(vurl).path
@@ -2814,6 +2819,10 @@ def filter_video(urls):
             video_url = urllib.parse.urljoin(url, video_url)
             video_id = urllib.parse.unquote(os.path.basename(video_url))
 
+            # Sometimes, jwplayer extraction will result in a YouTube URL
+            if YoutubeIE.suitable(video_url):
+                entries.append(self.url_result(video_url, 'Youtube'))
+                continue
 
             video_id = os.path.splitext(video_id)[0]
             headers = {
@@ -2875,4 +2884,4 @@ def filter_video(urls):
                 # 'url' results don't have a title
                 if e.get('title') is not None:
                     e['title'] = '%s (%d)' % (e['title'], num)
-        return entries
+        return entries

lib/yt_dlp/extractor/youtube.py

@@ -0,0 +1,6 @@
+from .common import InfoExtractor
+
+class YoutubeIE(InfoExtractor):
+    @classmethod
+    def suitable(cls, url):
+        return False