From 915e29379ad18958e3dc9893773834e644d6bcab Mon Sep 17 00:00:00 2001
From: Jesse Bannon <jbann1994@gmail.com>
Date: Mon, 8 Jan 2024 18:53:00 -0800
Subject: [PATCH] [BACKEND] Try to fetch source metadata from uploader_url
 (#886)

Makes it possible to fix https://github.com/jmbannon/ytdl-sub/issues/756

In the case that uploader_url does not equal the input url, keep trying to fetch uploader_url to get all possible metadata that yt-dlp does not provide in a single API call
---
 src/ytdl_sub/downloaders/ytdlp.py    | 20 +++++++++++++++++++-
 src/ytdl_sub/entries/entry_parent.py | 11 +++++++----
 2 files changed, 26 insertions(+), 5 deletions(-)

diff --git a/src/ytdl_sub/downloaders/ytdlp.py b/src/ytdl_sub/downloaders/ytdlp.py
index a7912232..0f43b606 100644
--- a/src/ytdl_sub/downloaders/ytdlp.py
+++ b/src/ytdl_sub/downloaders/ytdlp.py
@@ -216,11 +216,14 @@ class YTDLP:
         **kwargs
             arguments passed directory to YoutubeDL extract_info
         """
+        parent_dict: Dict = {}
         try:
             with cls._listen_and_log_downloaded_info_json(
                 working_directory=working_directory, log_prefix=log_prefix_on_info_json_dl
             ):
-                _ = cls.extract_info(ytdl_options_overrides=ytdl_options_overrides, **kwargs)
+                parent_dict = cls.extract_info(
+                    ytdl_options_overrides=ytdl_options_overrides, **kwargs
+                )
         except RejectedVideoReached:
             cls.logger.debug(
                 "RejectedVideoReached, stopping additional downloads "
@@ -234,4 +237,19 @@ class YTDLP:
         except MaxDownloadsReached:
             cls.logger.info("MaxDownloadsReached, stopping additional downloads.")
 
+        # For YouTube playlists in particular, channel metadata is not fetched. Attempt to get
+        # channel metadata via grabbing uploader_url info json a max of 3 times
+        current_iter = 0
+        url = kwargs.get("url")
+        uploader_url = parent_dict.get("uploader_url")
+        while current_iter < 3 and uploader_url and url != uploader_url:
+            cls.logger.debug("Attempting to get parent metadata from URL %s", uploader_url)
+            parent_dict = cls.extract_info(
+                ytdl_options_overrides=ytdl_options_overrides | {"playlist_items": "0:0"},
+                url=uploader_url,
+            )
+            current_iter += 1
+            url = uploader_url
+            uploader_url = parent_dict.get("uploader_url")
+
         return cls._get_entry_dicts_from_info_json_files(working_directory=working_directory)
diff --git a/src/ytdl_sub/entries/entry_parent.py b/src/ytdl_sub/entries/entry_parent.py
index 6cf7dcb1..232b2ab4 100644
--- a/src/ytdl_sub/entries/entry_parent.py
+++ b/src/ytdl_sub/entries/entry_parent.py
@@ -157,13 +157,16 @@ class EntryParent(BaseEntry):
         def _uid_is_uploader_id(parent: "EntryParent"):
             return parent.uid == parent.uploader_id
 
-        top_level_parents = [
-            parent for parent in parents if parent.num_children() == 0 and _url_matches(parent)
-        ]
+        top_level_parents = [parent for parent in parents if parent.num_children() == 0]
 
         # If more than 1 parent exists, assume the uploader_id is the root parent
         if len(top_level_parents) > 1:
-            top_level_parents = [parent for parent in parents if _uid_is_uploader_id(parent)]
+            top_level_parents = [
+                parent for parent in top_level_parents if _uid_is_uploader_id(parent)
+            ]
+
+        if len(top_level_parents) > 1:
+            top_level_parents = [parent for parent in top_level_parents if _url_matches(parent)]
 
         match len(top_level_parents):
             case 0: