From d65753ce059bb3bd5590db75101f3ef18a1ab68c Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Fri, 19 Jul 2024 04:49:09 +1200
Subject: [PATCH 01/20] [GoogleDriveFolder] Fix Extractor

---
 yt_dlp/extractor/googledrive.py | 43 ++++++++-------------------------
 1 file changed, 10 insertions(+), 33 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index dfba2d3ba1..4f4a9cf3f2 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -1,3 +1,4 @@
+import json
 import re
 import urllib.parse
 
@@ -298,46 +299,22 @@ class GoogleDriveFolderIE(InfoExtractor):
         },
         'playlist_count': 3,
     }]
-    _BOUNDARY = '=====vc17a3rwnndj====='
-    _REQUEST = "/drive/v2beta/files?openDrive=true&reason=102&syncType=0&errorRecovery=false&q=trashed%20%3D%20false%20and%20'{folder_id}'%20in%20parents&fields=kind%2CnextPageToken%2Citems(kind%2CmodifiedDate%2CmodifiedByMeDate%2ClastViewedByMeDate%2CfileSize%2Cowners(kind%2CpermissionId%2Cid)%2ClastModifyingUser(kind%2CpermissionId%2Cid)%2ChasThumbnail%2CthumbnailVersion%2Ctitle%2Cid%2CresourceKey%2Cshared%2CsharedWithMeDate%2CuserPermission(role)%2CexplicitlyTrashed%2CmimeType%2CquotaBytesUsed%2Ccopyable%2CfileExtension%2CsharingUser(kind%2CpermissionId%2Cid)%2Cspaces%2Cversion%2CteamDriveId%2ChasAugmentedPermissions%2CcreatedDate%2CtrashingUser(kind%2CpermissionId%2Cid)%2CtrashedDate%2Cparents(id)%2CshortcutDetails(targetId%2CtargetMimeType%2CtargetLookupStatus)%2Ccapabilities(canCopy%2CcanDownload%2CcanEdit%2CcanAddChildren%2CcanDelete%2CcanRemoveChildren%2CcanShare%2CcanTrash%2CcanRename%2CcanReadTeamDrive%2CcanMoveTeamDriveItem)%2Clabels(starred%2Ctrashed%2Crestricted%2Cviewed))%2CincompleteSearch&appDataFilter=NO_APP_DATA&spaces=drive&pageToken={page_token}&maxResults=50&supportsTeamDrives=true&includeItemsFromAllDrives=true&corpora=default&orderBy=folder%2Ctitle_natural%20asc&retryCount=0&key={key} HTTP/1.1"
-    _DATA = f'''--{_BOUNDARY}
-content-type: application/http
-content-transfer-encoding: binary
 
-GET %s
+    def _extract_json(self, idx, webpage):
+        RE = r'AF_initDataCallback\(\{key:\s*([\'"])ds:\s*%d\1,[^}]*data:(?P<data>\[.*?\]),\s*sideChannel:\s\{'
+        return json.loads(self._html_search_regex(RE % idx, webpage, 'JSON', group='data'))
 
---{_BOUNDARY}
-'''
-
-    def _call_api(self, folder_id, key, data, **kwargs):
-        response = self._download_webpage(
-            'https://clients6.google.com/batch/drive/v2beta',
-            folder_id, data=data.encode(),
-            headers={
-                'Content-Type': 'text/plain;charset=UTF-8;',
-                'Origin': 'https://drive.google.com',
-            }, query={
-                '$ct': f'multipart/mixed; boundary="{self._BOUNDARY}"',
-                'key': key,
-            }, **kwargs)
-        return self._search_json('', response, 'api response', folder_id, **kwargs) or {}
-
-    def _get_folder_items(self, folder_id, key):
-        page_token = ''
-        while page_token is not None:
-            request = self._REQUEST.format(folder_id=folder_id, page_token=page_token, key=key)
-            page = self._call_api(folder_id, key, self._DATA % request)
-            yield from page['items']
-            page_token = page.get('nextPageToken')
+    def _get_folder_items(self, results):
+        yield from results
 
     def _real_extract(self, url):
         folder_id = self._match_id(url)
 
         webpage = self._download_webpage(url, folder_id)
-        key = self._search_regex(r'"(\w{39})"', webpage, 'key')
 
-        folder_info = self._call_api(folder_id, key, self._DATA % f'/drive/v2beta/files/{folder_id} HTTP/1.1', fatal=False)
+        title = self._extract_json(0, webpage)[1][2]
+        results_4 = self._extract_json(4, webpage)[-1]
 
         return self.playlist_from_matches(
-            self._get_folder_items(folder_id, key), folder_id, folder_info.get('title'),
-            ie=GoogleDriveIE, getter=lambda item: f'https://drive.google.com/file/d/{item["id"]}')
+            self._get_folder_items(results_4), folder_id, title,
+            ie=GoogleDriveIE, getter=lambda item: f'https://drive.google.com/file/d/{item[0]}')

From 05403ea5ad7e57dcacde98055267285c93a3278b Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Fri, 19 Jul 2024 16:31:08 +1200
Subject: [PATCH 02/20] add tests, fix regex, improve stability.

---
 yt_dlp/extractor/googledrive.py | 71 +++++++++++++++++++++++++++------
 1 file changed, 58 insertions(+), 13 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 4f4a9cf3f2..7a9d0befb0 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -1,4 +1,3 @@
-import json
 import re
 import urllib.parse
 
@@ -298,23 +297,69 @@ class GoogleDriveFolderIE(InfoExtractor):
             'title': 'Forrest',
         },
         'playlist_count': 3,
+    }, {
+        # Contains various formats
+        'url': 'https://drive.google.com/drive/folders/1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
+        'info_dict': {
+            'id': '1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
+            'title': 'public folder',
+        },
+        'playlist_count': 4,
     }]
+    _JSON_DS_RE = r'key\s*?:\s*?([\'"])ds:\s*?%d\1,[^}]*data:'
+    _JSON_HASH_RE = r'hash\s*?:\s*?([\'"])%d\1,[^}]*data:'
+    _ARRAY_RE = r'\[(?s:.+)\]'
 
-    def _extract_json(self, idx, webpage):
-        RE = r'AF_initDataCallback\(\{key:\s*([\'"])ds:\s*%d\1,[^}]*data:(?P<data>\[.*?\]),\s*sideChannel:\s\{'
-        return json.loads(self._html_search_regex(RE % idx, webpage, 'JSON', group='data'))
+    def _extract_json_ds(self, dsval, webpage, video_id, **kwargs):
+        """
+        Searches for json with the 'ds' value(0~5) from the webpage with regex.
+        Folder info: ds=0; Folder items: ds=4.
+        For example, if the webpage contains the line below, the empty data array
+        can be got by passing dsval=3 to this function.
+            AF_initDataCallback({key: 'ds:3', hash: '2', data:[], sideChannel: {}});
+        """
+        return self._search_json(self._JSON_DS_RE % dsval, webpage,
+                                    f'webpage JSON ds:{dsval}', video_id,
+                                    contains_pattern=self._ARRAY_RE, **kwargs)
 
-    def _get_folder_items(self, results):
-        yield from results
+    def _extract_json_hash(self, hashval, webpage, video_id, **kwargs):
+        """
+        Searches for json with the 'hash' value(1~6) from the webpage with regex.
+        Folder info: hash=1; Folder items: hash=6.
+        For example, if the webpage contains the line below, the empty data array
+        can be got by passing hashval=2 to this function.
+            AF_initDataCallback({key: 'ds:3', hash: '2', data:[], sideChannel: {}});
+        """
+        return self._search_json(self._JSON_HASH_RE % hashval, webpage,
+                                      f'webpage JSON hash:{hashval}', video_id,
+                                      contains_pattern=self._ARRAY_RE, **kwargs)
 
     def _real_extract(self, url):
+        def item_url_getter(item):
+            url_from_0 = f'https://drive.google.com/file/d/{item[0]}'
+            if GoogleDriveIE.suitable(url_from_0):
+                return url_from_0
+            else:
+                for attr in item:
+                    if isinstance(attr, str) and GoogleDriveIE.suitable(attr):
+                        return attr
+            self.write_debug('Failed to extract url!')
+            return None
+
         folder_id = self._match_id(url)
+        headers = self.geo_verification_headers()
 
-        webpage = self._download_webpage(url, folder_id)
+        webpage = self._download_webpage(url, folder_id, headers=headers)
+        json_folder_info = (
+            self._extract_json_ds(0, webpage, folder_id, default=None)
+            or self._extract_json_hash(1, webpage, folder_id)
+        )
+        json_items = (
+            self._extract_json_ds(4, webpage, folder_id, default=None)
+            or self._extract_json_hash(6, webpage, folder_id)
+        )
+        title = json_folder_info[1][2]
+        items = json_items[-1]
 
-        title = self._extract_json(0, webpage)[1][2]
-        results_4 = self._extract_json(4, webpage)[-1]
-
-        return self.playlist_from_matches(
-            self._get_folder_items(results_4), folder_id, title,
-            ie=GoogleDriveIE, getter=lambda item: f'https://drive.google.com/file/d/{item[0]}')
+        return self.playlist_from_matches((item for item in items), folder_id, title,
+                                          ie=GoogleDriveIE, getter=item_url_getter)

From a917af960c8ab3895aaea0f44a4cef2021d418db Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Fri, 19 Jul 2024 16:48:36 +1200
Subject: [PATCH 03/20] fix code formating, fix test folder title

---
 yt_dlp/extractor/googledrive.py | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 7a9d0befb0..d8e64360ef 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -302,7 +302,7 @@ class GoogleDriveFolderIE(InfoExtractor):
         'url': 'https://drive.google.com/drive/folders/1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
         'info_dict': {
             'id': '1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
-            'title': 'public folder',
+            'title': r'], sideChannel: {}});',
         },
         'playlist_count': 4,
     }]
@@ -319,8 +319,8 @@ def _extract_json_ds(self, dsval, webpage, video_id, **kwargs):
             AF_initDataCallback({key: 'ds:3', hash: '2', data:[], sideChannel: {}});
         """
         return self._search_json(self._JSON_DS_RE % dsval, webpage,
-                                    f'webpage JSON ds:{dsval}', video_id,
-                                    contains_pattern=self._ARRAY_RE, **kwargs)
+                                 f'webpage JSON ds:{dsval}', video_id,
+                                 contains_pattern=self._ARRAY_RE, **kwargs)
 
     def _extract_json_hash(self, hashval, webpage, video_id, **kwargs):
         """
@@ -331,8 +331,8 @@ def _extract_json_hash(self, hashval, webpage, video_id, **kwargs):
             AF_initDataCallback({key: 'ds:3', hash: '2', data:[], sideChannel: {}});
         """
         return self._search_json(self._JSON_HASH_RE % hashval, webpage,
-                                      f'webpage JSON hash:{hashval}', video_id,
-                                      contains_pattern=self._ARRAY_RE, **kwargs)
+                                 f'webpage JSON hash:{hashval}', video_id,
+                                 contains_pattern=self._ARRAY_RE, **kwargs)
 
     def _real_extract(self, url):
         def item_url_getter(item):

From 64d4e93516e4ee78c312b7e6d84fb3b975658b8e Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Mon, 22 Jul 2024 16:51:04 +1200
Subject: [PATCH 04/20] add support for subfolders(recursive)

---
 yt_dlp/extractor/googledrive.py | 32 +++++++++++++++++++-------------
 1 file changed, 19 insertions(+), 13 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index d8e64360ef..28e496a458 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -298,13 +298,13 @@ class GoogleDriveFolderIE(InfoExtractor):
         },
         'playlist_count': 3,
     }, {
-        # Contains various formats
+        # Contains various formats and a subfolder
         'url': 'https://drive.google.com/drive/folders/1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
         'info_dict': {
             'id': '1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
             'title': r'], sideChannel: {}});',
         },
-        'playlist_count': 4,
+        'playlist_count': 5,
     }]
     _JSON_DS_RE = r'key\s*?:\s*?([\'"])ds:\s*?%d\1,[^}]*data:'
     _JSON_HASH_RE = r'hash\s*?:\s*?([\'"])%d\1,[^}]*data:'
@@ -335,17 +335,24 @@ def _extract_json_hash(self, hashval, webpage, video_id, **kwargs):
                                  contains_pattern=self._ARRAY_RE, **kwargs)
 
     def _real_extract(self, url):
-        def item_url_getter(item):
-            url_from_0 = f'https://drive.google.com/file/d/{item[0]}'
-            if GoogleDriveIE.suitable(url_from_0):
-                return url_from_0
-            else:
-                for attr in item:
-                    if isinstance(attr, str) and GoogleDriveIE.suitable(attr):
-                        return attr
-            self.write_debug('Failed to extract url!')
+        def item_url_getter(item, video_id):
+            available_IEs = [GoogleDriveFolderIE, GoogleDriveIE]
+            for attr in item:
+                if isinstance(attr, str):
+                    for available_IE in available_IEs:
+                        if available_IE.suitable(attr):
+                            return self.url_result(attr, available_IE, video_id, item[2])
+            self.to_screen(f'Failed to find a suitable extractor for {item[2]}.')
             return None
 
+        def make_playlist(items, playlist_id):
+            entries = []
+            for item in items:
+                entry = item_url_getter(item, playlist_id)
+                if entry:
+                    entries.append(entry)
+            return self.playlist_result(entries, playlist_id, title)
+
         folder_id = self._match_id(url)
         headers = self.geo_verification_headers()
 
@@ -361,5 +368,4 @@ def item_url_getter(item):
         title = json_folder_info[1][2]
         items = json_items[-1]
 
-        return self.playlist_from_matches((item for item in items), folder_id, title,
-                                          ie=GoogleDriveIE, getter=item_url_getter)
+        return make_playlist(items, folder_id)

From cbe698b4b0d7d1b8a9af5bbc3381a82855861f36 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Mon, 22 Jul 2024 17:28:45 +1200
Subject: [PATCH 05/20] Add support for empty folder

---
 yt_dlp/extractor/googledrive.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 28e496a458..950d877f97 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -367,5 +367,7 @@ def make_playlist(items, playlist_id):
         )
         title = json_folder_info[1][2]
         items = json_items[-1]
+        if not isinstance(items, list):
+            return self.playlist_result([], folder_id, title)
 
         return make_playlist(items, folder_id)

From 99628595959370d58a47a1d7c266e1224ad1ada9 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Wed, 24 Jul 2024 23:16:44 +1200
Subject: [PATCH 06/20] Replace function `make_playlist` with a more concise
 generator expression to improve code readability. All tests and code format
 checker are passing         modified:   yt_dlp/extractor/googledrive.py

---
 yt_dlp/extractor/googledrive.py | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 950d877f97..a4f2512119 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -345,14 +345,6 @@ def item_url_getter(item, video_id):
             self.to_screen(f'Failed to find a suitable extractor for {item[2]}.')
             return None
 
-        def make_playlist(items, playlist_id):
-            entries = []
-            for item in items:
-                entry = item_url_getter(item, playlist_id)
-                if entry:
-                    entries.append(entry)
-            return self.playlist_result(entries, playlist_id, title)
-
         folder_id = self._match_id(url)
         headers = self.geo_verification_headers()
 
@@ -365,9 +357,12 @@ def make_playlist(items, playlist_id):
             self._extract_json_ds(4, webpage, folder_id, default=None)
             or self._extract_json_hash(6, webpage, folder_id)
         )
+
         title = json_folder_info[1][2]
         items = json_items[-1]
         if not isinstance(items, list):
             return self.playlist_result([], folder_id, title)
 
-        return make_playlist(items, folder_id)
+        return self.playlist_result(
+            (entry for item in items if (entry := item_url_getter(item, folder_id))),
+            folder_id, title)

From 017997068bb9b181c27e022daa6e184592faebe2 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Sun, 28 Jul 2024 03:18:17 +1200
Subject: [PATCH 07/20] read shortcuts url

---
 yt_dlp/extractor/googledrive.py | 21 +++++++++++++--------
 1 file changed, 13 insertions(+), 8 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index a4f2512119..1a58882944 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -12,6 +12,7 @@
     get_element_html_by_id,
     int_or_none,
     lowercase_escape,
+    traverse_obj,
     try_get,
     update_url_query,
 )
@@ -304,7 +305,7 @@ class GoogleDriveFolderIE(InfoExtractor):
             'id': '1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
             'title': r'], sideChannel: {}});',
         },
-        'playlist_count': 5,
+        'playlist_count': 6,
     }]
     _JSON_DS_RE = r'key\s*?:\s*?([\'"])ds:\s*?%d\1,[^}]*data:'
     _JSON_HASH_RE = r'hash\s*?:\s*?([\'"])%d\1,[^}]*data:'
@@ -337,13 +338,17 @@ def _extract_json_hash(self, hashval, webpage, video_id, **kwargs):
     def _real_extract(self, url):
         def item_url_getter(item, video_id):
             available_IEs = [GoogleDriveFolderIE, GoogleDriveIE]
-            for attr in item:
-                if isinstance(attr, str):
-                    for available_IE in available_IEs:
-                        if available_IE.suitable(attr):
-                            return self.url_result(attr, available_IE, video_id, item[2])
-            self.to_screen(f'Failed to find a suitable extractor for {item[2]}.')
-            return None
+            if 'application/vnd.google-apps.shortcut' in item:
+                entry_url = traverse_obj(
+                    item, (..., ..., lambda _, v: any(ie.suitable(v) for ie in available_IEs),
+                           {str}, any))
+            else:
+                entry_url = traverse_obj(
+                    item, (lambda _, v: any(ie.suitable(v) for ie in available_IEs),
+                           {str}, any))
+            if not entry_url:
+                return None
+            return self.url_result(entry_url, video_id=video_id, video_title=item[2])
 
         folder_id = self._match_id(url)
         headers = self.geo_verification_headers()

From 99d9105f33ea72c58ed98fd0c951a7f50ea2f647 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Sun, 28 Jul 2024 03:36:55 +1200
Subject: [PATCH 08/20] [GoogleDrive] add support for shortcut urls: follow
 redirection Example url:
 https://drive.google.com/file/d/17OrYKQBPRm4J_D1rsGbo4eOmm1_SfoIY/ Redirects
 to(real video):
 https://drive.google.com/file/d/1Jp0I0tS-qMxtXNehGQW5_hWhwgC0FeeB/edit

---
 yt_dlp/extractor/googledrive.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 1a58882944..437b31db9b 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -167,6 +167,10 @@ def _get_automatic_captions(self, video_id, subtitles_id, hl):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+        _, webpage_urlh = self._download_webpage_handle(url, video_id)
+        if webpage_urlh.url != url:
+            return self.url_result(webpage_urlh.url)
+
         video_info = urllib.parse.parse_qs(self._download_webpage(
             'https://drive.google.com/get_video_info',
             video_id, 'Downloading video webpage', query={'docid': video_id}))

From 6e98d99dd5cb171f21fb07832fa62095bf2e55c9 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Sat, 17 Aug 2024 20:21:02 +1200
Subject: [PATCH 09/20] [GoogleDriveFolderIE] improve code readability -
 Combined `_extract_json_ds` and`_extract_json_hash` into one
 method(`_extract_json_meta`) - Improved `item_url_getter`'s traversal path of
 item info - Add notations to improve code readability

---
 yt_dlp/extractor/googledrive.py | 66 ++++++++++++++++-----------------
 1 file changed, 32 insertions(+), 34 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 437b31db9b..80ee06daba 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -303,7 +303,7 @@ class GoogleDriveFolderIE(InfoExtractor):
         },
         'playlist_count': 3,
     }, {
-        # Contains various formats and a subfolder
+        'note': 'Contains various formats and a subfolder, folder name was formerly mismatched',
         'url': 'https://drive.google.com/drive/folders/1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
         'info_dict': {
             'id': '1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
@@ -311,45 +311,43 @@ class GoogleDriveFolderIE(InfoExtractor):
         },
         'playlist_count': 6,
     }]
-    _JSON_DS_RE = r'key\s*?:\s*?([\'"])ds:\s*?%d\1,[^}]*data:'
-    _JSON_HASH_RE = r'hash\s*?:\s*?([\'"])%d\1,[^}]*data:'
-    _ARRAY_RE = r'\[(?s:.+)\]'
 
-    def _extract_json_ds(self, dsval, webpage, video_id, **kwargs):
+    def _extract_json_meta(self, webpage, video_id, dsval=None, hashval=None, name=None, **kwargs):
         """
-        Searches for json with the 'ds' value(0~5) from the webpage with regex.
-        Folder info: ds=0; Folder items: ds=4.
+        Uses regex to search for json metadata with 'ds' value(0-5) or 'hash' value(1-6)
+        from the webpage.
+        Folder info: ds=0, hash=1; Folder items: ds=4, hash=6.
         For example, if the webpage contains the line below, the empty data array
-        can be got by passing dsval=3 to this function.
+        can be got by passing dsval=3 or hashval=2 to this method.
             AF_initDataCallback({key: 'ds:3', hash: '2', data:[], sideChannel: {}});
         """
-        return self._search_json(self._JSON_DS_RE % dsval, webpage,
-                                 f'webpage JSON ds:{dsval}', video_id,
-                                 contains_pattern=self._ARRAY_RE, **kwargs)
-
-    def _extract_json_hash(self, hashval, webpage, video_id, **kwargs):
-        """
-        Searches for json with the 'hash' value(1~6) from the webpage with regex.
-        Folder info: hash=1; Folder items: hash=6.
-        For example, if the webpage contains the line below, the empty data array
-        can be got by passing hashval=2 to this function.
-            AF_initDataCallback({key: 'ds:3', hash: '2', data:[], sideChannel: {}});
-        """
-        return self._search_json(self._JSON_HASH_RE % hashval, webpage,
-                                 f'webpage JSON hash:{hashval}', video_id,
-                                 contains_pattern=self._ARRAY_RE, **kwargs)
+        _ARRAY_RE = r'\[(?s:.+)\]'
+        _META_END_RE = r', sideChannel: \{\}\}\);'  # greedy match to deal with the 2nd test case
+        if dsval:
+            if not name:
+                name = f'webpage JSON metadata ds:{dsval}'
+            return self._search_json(
+                rf'''key\s*?:\s*?(['"])ds:\s*?{dsval}\1,[^\[]*?data:''', webpage, name, video_id,
+                end_pattern=_META_END_RE, contains_pattern=_ARRAY_RE, **kwargs)
+        elif hashval:
+            if not name:
+                name = f'webpage JSON metadata hash:{hashval}'
+            return self._search_json(
+                rf'''hash\s*?:\s*?(['"]){hashval}\1,[^\[]*?data:''', webpage, name, video_id,
+                end_pattern=_META_END_RE, contains_pattern=_ARRAY_RE, **kwargs)
+        return None
 
     def _real_extract(self, url):
         def item_url_getter(item, video_id):
-            available_IEs = [GoogleDriveFolderIE, GoogleDriveIE]
-            if 'application/vnd.google-apps.shortcut' in item:
+            available_IEs = [GoogleDriveFolderIE, GoogleDriveIE]  # subfolder or item
+            if 'application/vnd.google-apps.shortcut' in item:  # extract real link
                 entry_url = traverse_obj(
-                    item, (..., ..., lambda _, v: any(ie.suitable(v) for ie in available_IEs),
-                           {str}, any))
+                    item,
+                    (..., ..., lambda _, v: any(ie.suitable(v) for ie in available_IEs), any))
             else:
                 entry_url = traverse_obj(
-                    item, (lambda _, v: any(ie.suitable(v) for ie in available_IEs),
-                           {str}, any))
+                    item,
+                    (lambda _, v: any(ie.suitable(v) for ie in available_IEs), any))
             if not entry_url:
                 return None
             return self.url_result(entry_url, video_id=video_id, video_title=item[2])
@@ -359,17 +357,17 @@ def item_url_getter(item, video_id):
 
         webpage = self._download_webpage(url, folder_id, headers=headers)
         json_folder_info = (
-            self._extract_json_ds(0, webpage, folder_id, default=None)
-            or self._extract_json_hash(1, webpage, folder_id)
+            self._extract_json_meta(webpage, folder_id, dsval=0, name='folder info', default=None)
+            or self._extract_json_meta(webpage, folder_id, hashval=1)
         )
         json_items = (
-            self._extract_json_ds(4, webpage, folder_id, default=None)
-            or self._extract_json_hash(6, webpage, folder_id)
+            self._extract_json_meta(webpage, folder_id, dsval=4, name='folder items', default=None)
+            or self._extract_json_meta(webpage, folder_id, hashval=6)
         )
 
         title = json_folder_info[1][2]
         items = json_items[-1]
-        if not isinstance(items, list):
+        if not isinstance(items, list):  # empty folder
             return self.playlist_result([], folder_id, title)
 
         return self.playlist_result(

From 8623ada2938124370211dd5682eab281690fdec4 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Sat, 17 Aug 2024 23:49:00 +1200
Subject: [PATCH 10/20] [GoogleDriveFolderIE] Several fixes - Update test:
 GoogleDriveFolder_1 - Raise for log-in required - catch HTTP Error 404

---
 yt_dlp/extractor/googledrive.py | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 80ee06daba..4603611dcc 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -3,6 +3,7 @@
 
 from .common import InfoExtractor
 from .youtube import YoutubeIE
+from ..networking.exceptions import HTTPError
 from ..utils import (
     ExtractorError,
     bug_reports_message,
@@ -303,13 +304,14 @@ class GoogleDriveFolderIE(InfoExtractor):
         },
         'playlist_count': 3,
     }, {
-        'note': 'Contains various formats and a subfolder, folder name was formerly mismatched',
-        'url': 'https://drive.google.com/drive/folders/1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
+        'note': 'Contains various formats and a subfolder, folder name was formerly mismatched.'
+                'also contains loop shortcut, shortcut to non-downloadable files, etc.',
+        'url': 'https://drive.google.com/drive/folders/1jjrhqi94d8TSHSVMSdBjD49MOiHYpHfF',
         'info_dict': {
-            'id': '1CkqRsNlzZ0o3IL083j17s6sH5Q83DcGo',
-            'title': r'], sideChannel: {}});',
+            'id': '1jjrhqi94d8TSHSVMSdBjD49MOiHYpHfF',
+            'title': '], sideChannel: {}});',
         },
-        'playlist_count': 6,
+        'playlist_count': 8,
     }]
 
     def _extract_json_meta(self, webpage, video_id, dsval=None, hashval=None, name=None, **kwargs):
@@ -355,7 +357,14 @@ def item_url_getter(item, video_id):
         folder_id = self._match_id(url)
         headers = self.geo_verification_headers()
 
-        webpage = self._download_webpage(url, folder_id, headers=headers)
+        try:
+            webpage, urlh = self._download_webpage_handle(url, folder_id, headers=headers)
+        except ExtractorError as e:
+            if isinstance(e.cause, HTTPError) and e.cause.status == 404:
+                self.raise_no_formats(e.cause.msg)
+        if urllib.parse.urlparse(urlh.url).netloc == 'accounts.google.com':
+            self.raise_login_required('This video is only available for registered users')
+
         json_folder_info = (
             self._extract_json_meta(webpage, folder_id, dsval=0, name='folder info', default=None)
             or self._extract_json_meta(webpage, folder_id, hashval=1)

From 28ed64d87a4bfb7c27b96124c2dab41e9ba2c614 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Sat, 17 Aug 2024 23:49:56 +1200
Subject: [PATCH 11/20] [GoogleDriveFolderIE] add bare raise in the except
 block

---
 yt_dlp/extractor/googledrive.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 4603611dcc..a77ca24c07 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -362,6 +362,7 @@ def item_url_getter(item, video_id):
         except ExtractorError as e:
             if isinstance(e.cause, HTTPError) and e.cause.status == 404:
                 self.raise_no_formats(e.cause.msg)
+            raise
         if urllib.parse.urlparse(urlh.url).netloc == 'accounts.google.com':
             self.raise_login_required('This video is only available for registered users')
 

From b81a41d5ff33f05379c7ae96997d9d366ff4061b Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Sun, 18 Aug 2024 00:01:03 +1200
Subject: [PATCH 12/20] [GoogleDriveFolderIE] raise for access denied

---
 yt_dlp/extractor/googledrive.py | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index a77ca24c07..a3d645f08c 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -360,8 +360,11 @@ def item_url_getter(item, video_id):
         try:
             webpage, urlh = self._download_webpage_handle(url, folder_id, headers=headers)
         except ExtractorError as e:
-            if isinstance(e.cause, HTTPError) and e.cause.status == 404:
-                self.raise_no_formats(e.cause.msg)
+            if isinstance(e.cause, HTTPError):
+                if e.cause.status == 404:
+                    self.raise_no_formats(e.cause.msg)
+                elif e.cause.status == 403:
+                    self.raise_login_required('Access Denied!')
             raise
         if urllib.parse.urlparse(urlh.url).netloc == 'accounts.google.com':
             self.raise_login_required('This video is only available for registered users')

From 4a76306868e2387dc988edabe20ff4077b5b30c3 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Sun, 18 Aug 2024 03:08:45 +1200
Subject: [PATCH 13/20] [GoogleDriveFolderIE] support private folders and
 my-drive

---
 yt_dlp/extractor/googledrive.py | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index a3d645f08c..f89718b5c1 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -295,7 +295,7 @@ def add_source_format(urlh):
 
 class GoogleDriveFolderIE(InfoExtractor):
     IE_NAME = 'GoogleDrive:Folder'
-    _VALID_URL = r'https?://(?:docs|drive)\.google\.com/drive/folders/(?P<id>[\w-]{28,})'
+    _VALID_URL = r'https?://(?:docs|drive)\.google\.com/drive/(?:folders/(?P<id>[\w-]{28,})|my-drive)'
     _TESTS = [{
         'url': 'https://drive.google.com/drive/folders/1dQ4sx0-__Nvg65rxTSgQrl7VyW_FZ9QI',
         'info_dict': {
@@ -318,7 +318,7 @@ def _extract_json_meta(self, webpage, video_id, dsval=None, hashval=None, name=N
         """
         Uses regex to search for json metadata with 'ds' value(0-5) or 'hash' value(1-6)
         from the webpage.
-        Folder info: ds=0, hash=1; Folder items: ds=4, hash=6.
+        Folder info: ds=0, hash=1; Folder items: ds=4(public folder)/5(private folder), hash=6.
         For example, if the webpage contains the line below, the empty data array
         can be got by passing dsval=3 or hashval=2 to this method.
             AF_initDataCallback({key: 'ds:3', hash: '2', data:[], sideChannel: {}});
@@ -341,6 +341,8 @@ def _extract_json_meta(self, webpage, video_id, dsval=None, hashval=None, name=N
 
     def _real_extract(self, url):
         def item_url_getter(item, video_id):
+            if not isinstance(item, list):
+                return None
             available_IEs = [GoogleDriveFolderIE, GoogleDriveIE]  # subfolder or item
             if 'application/vnd.google-apps.shortcut' in item:  # extract real link
                 entry_url = traverse_obj(
@@ -354,7 +356,7 @@ def item_url_getter(item, video_id):
                 return None
             return self.url_result(entry_url, video_id=video_id, video_title=item[2])
 
-        folder_id = self._match_id(url)
+        folder_id = self._match_id(url) or 'my-drive'
         headers = self.geo_verification_headers()
 
         try:
@@ -370,13 +372,10 @@ def item_url_getter(item, video_id):
             self.raise_login_required('This video is only available for registered users')
 
         json_folder_info = (
-            self._extract_json_meta(webpage, folder_id, dsval=0, name='folder info', default=None)
-            or self._extract_json_meta(webpage, folder_id, hashval=1)
-        )
-        json_items = (
-            self._extract_json_meta(webpage, folder_id, dsval=4, name='folder items', default=None)
-            or self._extract_json_meta(webpage, folder_id, hashval=6)
+            self._extract_json_meta(webpage, folder_id, hashval=1, name='folder info', default=None)
+            or self._extract_json_meta(webpage, folder_id, dsval=0, name='folder info - fallback')
         )
+        json_items = self._extract_json_meta(webpage, folder_id, hashval=6, name='folder items')
 
         title = json_folder_info[1][2]
         items = json_items[-1]

From d133c2c7f707b7dbc00a784c75ca0f2148326f1d Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Sun, 18 Aug 2024 03:15:45 +1200
Subject: [PATCH 14/20] [GoogleDriveFolderIE] revert part of last commit -
 change folder info json metadata extraction default appraoch to ds match

---
 yt_dlp/extractor/googledrive.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index f89718b5c1..1b1e3a2daf 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -372,8 +372,8 @@ def item_url_getter(item, video_id):
             self.raise_login_required('This video is only available for registered users')
 
         json_folder_info = (
-            self._extract_json_meta(webpage, folder_id, hashval=1, name='folder info', default=None)
-            or self._extract_json_meta(webpage, folder_id, dsval=0, name='folder info - fallback')
+            self._extract_json_meta(webpage, folder_id, dsval=0, name='folder info', default=None)
+            or self._extract_json_meta(webpage, folder_id, hashval=1, name='folder info - fallback')
         )
         json_items = self._extract_json_meta(webpage, folder_id, hashval=6, name='folder items')
 

From beb76094fa817376741be87113eb1a28e69d341c Mon Sep 17 00:00:00 2001
From: N/Ame <173015200+grqz@users.noreply.github.com>
Date: Sat, 28 Sep 2024 20:59:50 +1200
Subject: [PATCH 15/20] Update yt_dlp/extractor/googledrive.py

---
 yt_dlp/extractor/googledrive.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 1b1e3a2daf..86900f5767 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -369,7 +369,7 @@ def item_url_getter(item, video_id):
                     self.raise_login_required('Access Denied!')
             raise
         if urllib.parse.urlparse(urlh.url).netloc == 'accounts.google.com':
-            self.raise_login_required('This video is only available for registered users')
+            self.raise_login_required('Access Denied!')
 
         json_folder_info = (
             self._extract_json_meta(webpage, folder_id, dsval=0, name='folder info', default=None)

From 8d827d2460e6eabea92dec7766a5dce9598388bb Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Sun, 29 Sep 2024 00:24:50 +1200
Subject: [PATCH 16/20] Update tests

---
 yt_dlp/extractor/googledrive.py | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 86900f5767..65441c6965 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -53,6 +53,17 @@ class GoogleDriveIE(InfoExtractor):
             'duration': 184,
             'thumbnail': 'https://drive.google.com/thumbnail?id=1IP0o8dHcQrIHGgVyp0Ofvx2cGfLzyO1x',
         },
+    }, {
+        # shortcut url
+        'url': 'https://drive.google.com/file/d/1_n3-8ZwEUV4OniMsLAJ_C1JEjuT2u5Pk/view?usp=drivesdk',
+        'md5': '43d34f7be1acc0262f337a039d1ad12d',
+        'info_dict': {
+            'id': '1J1RCw2jcgUngrZRdpza-IHXYkardZ-4l',
+            'ext': 'webm',
+            'title': 'Forrest walk with Best Mind Refresh Music Mithran [tEvJKrE4cS0].webm',
+            'duration': 512,
+            'thumbnail': 'https://drive.google.com/thumbnail?id=1J1RCw2jcgUngrZRdpza-IHXYkardZ-4l',
+        },
     }, {
         # video can't be watched anonymously due to view count limit reached,
         # but can be downloaded (see https://github.com/ytdl-org/youtube-dl/issues/14046)
@@ -306,7 +317,7 @@ class GoogleDriveFolderIE(InfoExtractor):
     }, {
         'note': 'Contains various formats and a subfolder, folder name was formerly mismatched.'
                 'also contains loop shortcut, shortcut to non-downloadable files, etc.',
-        'url': 'https://drive.google.com/drive/folders/1jjrhqi94d8TSHSVMSdBjD49MOiHYpHfF',
+        'url': 'https://docs.google.com/drive/folders/1jjrhqi94d8TSHSVMSdBjD49MOiHYpHfF',
         'info_dict': {
             'id': '1jjrhqi94d8TSHSVMSdBjD49MOiHYpHfF',
             'title': '], sideChannel: {}});',

From 83e086083577fe50a1d24c34b141006402ce389f Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Mon, 30 Sep 2024 06:35:33 +1300
Subject: [PATCH 17/20] fix my-drive extraction

---
 yt_dlp/extractor/googledrive.py | 33 +++++++++++++++++----------------
 1 file changed, 17 insertions(+), 16 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 65441c6965..8b6ac8d854 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -306,7 +306,7 @@ def add_source_format(urlh):
 
 class GoogleDriveFolderIE(InfoExtractor):
     IE_NAME = 'GoogleDrive:Folder'
-    _VALID_URL = r'https?://(?:docs|drive)\.google\.com/drive/(?:folders/(?P<id>[\w-]{28,})|my-drive)'
+    _VALID_URL = r'https?://(?:docs|drive)\.google\.com/drive/(?:folders/(?P<id>[\w-]{19,})|my-drive)'
     _TESTS = [{
         'url': 'https://drive.google.com/drive/folders/1dQ4sx0-__Nvg65rxTSgQrl7VyW_FZ9QI',
         'info_dict': {
@@ -329,32 +329,36 @@ def _extract_json_meta(self, webpage, video_id, dsval=None, hashval=None, name=N
         """
         Uses regex to search for json metadata with 'ds' value(0-5) or 'hash' value(1-6)
         from the webpage.
-        Folder info: ds=0, hash=1; Folder items: ds=4(public folder)/5(private folder), hash=6.
+        Folder info: ds=0(public folder), hash=1/5;
+        Folder items: ds=4(logged out)/5(logged in), hash=6.
+            public, logged in     info:ds0hash1;      items:ds5hash6
+            public, logged out    info:ds0hash1;      items:ds4hash6
+            my-drive, logged in   info:ds0hash1/4;    items:ds5hash6
+            private, logged in    info:ds0hash1;      items:ds5hash6
         For example, if the webpage contains the line below, the empty data array
         can be got by passing dsval=3 or hashval=2 to this method.
             AF_initDataCallback({key: 'ds:3', hash: '2', data:[], sideChannel: {}});
         """
         _ARRAY_RE = r'\[(?s:.+)\]'
         _META_END_RE = r', sideChannel: \{\}\}\);'  # greedy match to deal with the 2nd test case
-        if dsval:
+        if dsval is not None:
             if not name:
                 name = f'webpage JSON metadata ds:{dsval}'
             return self._search_json(
                 rf'''key\s*?:\s*?(['"])ds:\s*?{dsval}\1,[^\[]*?data:''', webpage, name, video_id,
                 end_pattern=_META_END_RE, contains_pattern=_ARRAY_RE, **kwargs)
-        elif hashval:
+        elif hashval is not None:
             if not name:
                 name = f'webpage JSON metadata hash:{hashval}'
             return self._search_json(
                 rf'''hash\s*?:\s*?(['"]){hashval}\1,[^\[]*?data:''', webpage, name, video_id,
                 end_pattern=_META_END_RE, contains_pattern=_ARRAY_RE, **kwargs)
-        return None
 
     def _real_extract(self, url):
         def item_url_getter(item, video_id):
             if not isinstance(item, list):
                 return None
-            available_IEs = [GoogleDriveFolderIE, GoogleDriveIE]  # subfolder or item
+            available_IEs = (GoogleDriveFolderIE, GoogleDriveIE)  # subfolder or item
             if 'application/vnd.google-apps.shortcut' in item:  # extract real link
                 entry_url = traverse_obj(
                     item,
@@ -377,20 +381,17 @@ def item_url_getter(item, video_id):
                 if e.cause.status == 404:
                     self.raise_no_formats(e.cause.msg)
                 elif e.cause.status == 403:
-                    self.raise_login_required('Access Denied!')
+                    # logged in with an account without access
+                    self.raise_login_required('Access Denied')
             raise
         if urllib.parse.urlparse(urlh.url).netloc == 'accounts.google.com':
-            self.raise_login_required('Access Denied!')
+            # not logged in when visiting a private folder
+            self.raise_login_required('Access Denied')
 
-        json_folder_info = (
-            self._extract_json_meta(webpage, folder_id, dsval=0, name='folder info', default=None)
-            or self._extract_json_meta(webpage, folder_id, hashval=1, name='folder info - fallback')
-        )
-        json_items = self._extract_json_meta(webpage, folder_id, hashval=6, name='folder items')
+        title = self._extract_json_meta(webpage, folder_id, dsval=0, name='folder info')[1][2]
+        items = self._extract_json_meta(webpage, folder_id, hashval=6, name='folder items')[-1]
 
-        title = json_folder_info[1][2]
-        items = json_items[-1]
-        if not isinstance(items, list):  # empty folder
+        if items is False:  # empty folder
             return self.playlist_result([], folder_id, title)
 
         return self.playlist_result(

From 23ea25196deb91cfbf69ce2fd4bd56f7557f6d4b Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Mon, 30 Sep 2024 12:24:13 +1300
Subject: [PATCH 18/20] update docstring

---
 yt_dlp/extractor/googledrive.py | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 8b6ac8d854..6399c66ab8 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -331,10 +331,9 @@ def _extract_json_meta(self, webpage, video_id, dsval=None, hashval=None, name=N
         from the webpage.
         Folder info: ds=0(public folder), hash=1/5;
         Folder items: ds=4(logged out)/5(logged in), hash=6.
-            public, logged in     info:ds0hash1;      items:ds5hash6
-            public, logged out    info:ds0hash1;      items:ds4hash6
-            my-drive, logged in   info:ds0hash1/4;    items:ds5hash6
-            private, logged in    info:ds0hash1;      items:ds5hash6
+            logged out  info:ds0hash1;          items:ds4hash6
+            logged in   info:ds0hash1;          items:ds5hash6
+            my-drive    info:ds0hash1/ds0hash4; items:ds5hash6
         For example, if the webpage contains the line below, the empty data array
         can be got by passing dsval=3 or hashval=2 to this method.
             AF_initDataCallback({key: 'ds:3', hash: '2', data:[], sideChannel: {}});

From 3582a238a09ee929b75b2030a99146f6b4aa5b23 Mon Sep 17 00:00:00 2001
From: grqx <zenithwzj@gmail.com>
Date: Tue, 1 Oct 2024 23:51:00 +1300
Subject: [PATCH 19/20] update url and video id instead of returning a url
 result

---
 yt_dlp/extractor/googledrive.py | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 6399c66ab8..33ac5cb8a5 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -179,9 +179,16 @@ def _get_automatic_captions(self, video_id, subtitles_id, hl):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        _, webpage_urlh = self._download_webpage_handle(url, video_id)
+        try:
+            _, webpage_urlh = self._download_webpage_handle(url, video_id)
+        except ExtractorError as e:
+            if isinstance(e.cause, HTTPError):
+                if e.cause.status in (401, 403):
+                    self.raise_login_required('Access Denied')
+                raise
         if webpage_urlh.url != url:
-            return self.url_result(webpage_urlh.url)
+            url = webpage_urlh.url
+            video_id = self._match_id(url)
 
         video_info = urllib.parse.parse_qs(self._download_webpage(
             'https://drive.google.com/get_video_info',
@@ -378,7 +385,7 @@ def item_url_getter(item, video_id):
         except ExtractorError as e:
             if isinstance(e.cause, HTTPError):
                 if e.cause.status == 404:
-                    self.raise_no_formats(e.cause.msg)
+                    self.raise_no_formats(e.cause.msg, expected=True)
                 elif e.cause.status == 403:
                     # logged in with an account without access
                     self.raise_login_required('Access Denied')

From 1ef35f1c0001c59e4694e9cdd19873bab16c3489 Mon Sep 17 00:00:00 2001
From: grqx_termux <zenithwzj@gmail.com>
Date: Wed, 2 Oct 2024 17:52:29 +1300
Subject: [PATCH 20/20] update docstring

---
 yt_dlp/extractor/googledrive.py | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/yt_dlp/extractor/googledrive.py b/yt_dlp/extractor/googledrive.py
index 33ac5cb8a5..0594cbae13 100644
--- a/yt_dlp/extractor/googledrive.py
+++ b/yt_dlp/extractor/googledrive.py
@@ -336,11 +336,9 @@ def _extract_json_meta(self, webpage, video_id, dsval=None, hashval=None, name=N
         """
         Uses regex to search for json metadata with 'ds' value(0-5) or 'hash' value(1-6)
         from the webpage.
-        Folder info: ds=0(public folder), hash=1/5;
-        Folder items: ds=4(logged out)/5(logged in), hash=6.
-            logged out  info:ds0hash1;          items:ds4hash6
-            logged in   info:ds0hash1;          items:ds5hash6
-            my-drive    info:ds0hash1/ds0hash4; items:ds5hash6
+            logged out  folder info:ds0hash1;          items:ds4hash6
+            logged in   folder info:ds0hash1;          items:ds5hash6
+            my-drive    folder info:ds0hash1/ds0hash4; items:ds5hash6
         For example, if the webpage contains the line below, the empty data array
         can be got by passing dsval=3 or hashval=2 to this method.
             AF_initDataCallback({key: 'ds:3', hash: '2', data:[], sideChannel: {}});