[BUGFIX] Properly handle UIDs that have misbehaving chars that get sanitized (#870)

Should hopefuly fix both:
- https://github.com/jmbannon/ytdl-sub/issues/843
- https://github.com/jmbannon/ytdl-sub/issues/867

Various sites have unique IDs with characters that get sanitized (i.e. `/`, `:`). `ytdl-sub` was not using sanitized UIDs when looking for files, and would throw an error saying download was missing. This should resolve that issue.

src/ytdl_sub/downloaders/info_json/info_json_downloader.py

@@ -152,7 +152,9 @@ class InfoJsonDownloader(SourcePlugin[InfoJsonDownloaderOptions]):
         for file_name in entry_file_names:
             ext = get_file_extension(file_name)
             file_path = Path(self.output_directory) / file_name
-            working_directory_file_path = Path(self.working_directory) / f"{entry.uid}.{ext}"
+            working_directory_file_path = Path(self.working_directory) / entry.base_filename(
+                ext=ext
+            )
 
             # NFO files will always get rewritten, so ignore
             if ext == "nfo":

src/ytdl_sub/entries/base_entry.py

@@ -8,6 +8,8 @@ from typing import Type
 from typing import TypeVar
 from typing import final
 
+from yt_dlp.utils import sanitize_filename
+
 from ytdl_sub.entries.script.variable_definitions import VARIABLES
 from ytdl_sub.entries.script.variable_definitions import VariableDefinitions
 
@@ -45,6 +47,19 @@ class BaseEntry(ABC):
         """
         return str(self._kwargs[v.uid.metadata_key])
 
+    @property
+    def uid_sanitized(self) -> str:
+        """
+        Sanitized version, used in filenames
+        """
+        return sanitize_filename(self.uid)
+
+    def base_filename(self, ext: str):
+        """
+        The base filename of all yt-dlp downloaded entry files
+        """
+        return f"{self.uid_sanitized}.{ext}"
+
     @property
     def download_archive_extractor(self) -> str:
         """
@@ -124,7 +139,7 @@ class BaseEntry(ABC):
         -------
         The download info json's file name
         """
-        return f"{self.uid}.{self.info_json_ext}"
+        return self.base_filename(ext=self.info_json_ext)
 
     def get_download_info_json_path(self) -> str:
         """

src/ytdl_sub/entries/entry.py

@@ -113,7 +113,8 @@ class Entry(BaseEntry, Scriptable):
         """
         ext = self.try_get(v.ext, str) or self._kwargs[v.ext.metadata_key]
         for possible_ext in [ext, "mkv"]:
-            file_path = str(Path(self.working_directory()) / f"{self.uid}.{possible_ext}")
+            file_name = self.base_filename(ext=possible_ext)
+            file_path = str(Path(self.working_directory()) / file_name)
             if os.path.isfile(file_path):
                 return possible_ext
 
@@ -125,7 +126,7 @@ class Entry(BaseEntry, Scriptable):
         -------
         The entry's file name
         """
-        return f"{self.uid}.{self.ext}"
+        return self.base_filename(ext=self.ext)
 
     def get_download_file_path(self) -> str:
         """Returns the entry's file path to where it was downloaded"""
@@ -137,7 +138,7 @@ class Entry(BaseEntry, Scriptable):
         -------
         The download thumbnail's file name
         """
-        return f"{self.uid}.{self.get(v.thumbnail_ext, str)}"
+        return self.base_filename(ext=self.get(v.thumbnail_ext, str))
 
     def get_download_thumbnail_path(self) -> str:
         """Returns the entry's thumbnail's file path to where it was downloaded"""
@@ -155,7 +156,10 @@ class Entry(BaseEntry, Scriptable):
             possible_thumbnail_exts.add(thumbnail["url"].split(".")[-1])
 
         for ext in possible_thumbnail_exts:
-            possible_thumbnail_path = str(Path(self.working_directory()) / f"{self.uid}.{ext}")
+            possible_thumbnail_filename = self.base_filename(ext=ext)
+            possible_thumbnail_path = str(
+                Path(self.working_directory()) / possible_thumbnail_filename
+            )
             if os.path.isfile(possible_thumbnail_path):
                 return possible_thumbnail_path
 
@@ -202,7 +206,7 @@ class Entry(BaseEntry, Scriptable):
         # HACK: yt-dlp does not record extracted/converted extensions anywhere. If the file is not
         # found, try it using all possible extensions
         if not file_exists:
-            for ext in AUDIO_CODEC_EXTS.union(VIDEO_CODEC_EXTS):
+            for ext in AUDIO_CODEC_EXTS | VIDEO_CODEC_EXTS:
                 if os.path.isfile(self.get_download_file_path().removesuffix(self.ext) + ext):
                     file_exists = True
                     break

src/ytdl_sub/plugins/subtitles.py

@@ -208,7 +208,6 @@ class SubtitlesPlugin(Plugin[SubtitleOptions]):
             file_metadata = FileMetadata(f"Embedded subtitles with lang(s) {', '.join(langs)}")
         if self.plugin_options.subtitles_name:
             for lang in langs:
-                subtitle_file_name = f"{entry.uid}.{lang}.{self.plugin_options.subtitles_type}"
                 output_subtitle_file_name = self.overrides.apply_formatter(
                     formatter=self.plugin_options.subtitles_name,
                     entry=entry,
@@ -216,7 +215,9 @@ class SubtitlesPlugin(Plugin[SubtitleOptions]):
                 )
 
                 self.save_file(
-                    file_name=subtitle_file_name,
+                    file_name=entry.base_filename(
+                        ext=f"{lang}.{self.plugin_options.subtitles_type}"
+                    ),
                     output_file_name=output_subtitle_file_name,
                     entry=entry,
                 )
@@ -225,9 +226,8 @@ class SubtitlesPlugin(Plugin[SubtitleOptions]):
         # Can happen for both file and embedded subs
         for lang in langs:
             for possible_ext in SUBTITLE_EXTENSIONS:
-                possible_subs_file = (
-                    Path(self.working_directory) / f"{entry.uid}.{lang}.{possible_ext}"
-                )
+                possible_subs_filename = entry.base_filename(ext=f"{lang}.{possible_ext}")
+                possible_subs_file = Path(self.working_directory) / possible_subs_filename
                 FileHandler.delete(possible_subs_file)
 
         return file_metadata

src/ytdl_sub/subscriptions/subscription_ytdl_options.py

@@ -56,8 +56,8 @@ class SubscriptionYTDLOptions:
         ytdl-options to apply to every run no matter what
         """
         ytdl_options = {
-            # Download all files in the format of {id}.{ext}
-            "outtmpl": str(Path(self._working_directory) / "%(id)s.%(ext)s"),
+            # Download all files in the format of {id}.{ext}, where id is sanitized
+            "outtmpl": str(Path(self._working_directory) / "%(id)S.%(ext)s"),
             # Always write thumbnails
             "writethumbnail": True,
             "ffmpeg_location": FFMPEG.ffmpeg_path(),