rssVideos: Cleverer cleaning

config/scripts/rssVideos

@@ -29,6 +29,7 @@ import yt_dlp as youtube_dl
 
 log = logging.getLogger(__name__)
 
+# TODO Lockfile, or a way to parallel watch and download
 
 def configure_logging(args: configargparse.Namespace) -> None:
     # Configure logging
@@ -95,8 +96,12 @@ class RVElement:
     def guid(self) -> int:
         return int(self.get_tag_data("guid"))
 
+    @property
+    def is_researched(self) -> bool:
+        return "ytdl_infos" in self.__dict__
+
     def salvage_cache(self, cache: "RVElement") -> None:
-        if "ytdl_infos" in cache.__dict__:
+        if cache.is_researched:
             self.__dict__["ytdl_infos"] = cache.__dict__["ytdl_infos"]
             log.debug(f"From cache: {self}")
         if cache.was_downloaded:
@@ -109,7 +114,7 @@ class RVElement:
 
     @property
     def downloaded(self) -> bool:
-        if "ytdl_infos" not in self.__dict__:
+        if not self.is_researched:
             return False
         return os.path.isfile(self.filepath)
 
@@ -167,10 +172,8 @@ class RVElement:
         self.was_downloaded = True
         self.parent.save()
 
-    def act(self) -> None:
+    def preload(self) -> None:
-        if not self.is_video:
+        assert self.is_video
-            log.debug(f"Not a video: {self}")
-            return
         if self.downloaded:
             log.debug(f"Currently downloaded: {self}")
             return
@@ -192,16 +195,22 @@ class RVElement:
 
     def matches_filter(self, args: configargparse.Namespace) -> bool:
         if self.watched:
+            log.debug(f"Already watched: {self}")
             return False
         if args.title and not re.search(args.title, self.title):
+            log.debug(f"Title not matching {args.title}: {self}")
             return False
         if args.guid and not re.search(args.guid, str(self.guid)):
+            log.debug(f"Guid not matching {args.guid}: {self}")
             return False
         if args.link and not re.search(args.link, self.link):
+            log.debug(f"Link not matching {args.link}: {self}")
             return False
-        if args.creator and self.creator and not re.search(args.creator, self.creator):
+        if args.creator and (not self.creator or not re.search(args.creator, self.creator)):
+            log.debug(f"Creator not matching {args.creator}: {self}")
             return False
         if not self.is_video:
+            log.debug(f"Not a video: {self}")
             return False
         if args.duration:
             dur = args.duration
@@ -228,6 +237,7 @@ class RVElement:
 
             duration = int(dur)
             if not comparator(self.duration, duration * multiplier):
+                log.debug(f"Duration {self.duration} not matching {args.duration}: {self}")
                 return False
         return True
 
@@ -244,6 +254,15 @@ class RVElement:
         self.watched = True
         self.parent.save()
 
+    def clean(self) -> None:
+        assert self.is_video
+        log.info(f"Removing gone video: {self.filename}*")
+        for file in os.listdir():
+            if file.startswith(self.filename):
+                log.debug(f"Removing file: {file}")
+                if not self.parent.args.dryrun:
+                    os.unlink(file)
+
 
 class RVDatabase:
     SAVE_FILE = ".cache.p"
@@ -281,6 +300,16 @@ class RVDatabase:
             if el.guid in cache_els:
                 el.salvage_cache(cache_els[el.guid])
 
+    def clean_cache(self, cache: "RVDatabase") -> None:
+        log.debug(f"Cleaning cache")
+        self_els = dict()
+        for self_el in self.elements:
+            self_els[self_el.guid] = self_el
+        for el in cache.elements:
+            if el.guid not in self_els:
+                if el.is_researched and el.is_video:
+                    el.clean()
+
     def import_cache(self, cache: "RVDatabase") -> None:
         log.debug(f"Importing cache")
         self.feed_xml = cache.feed_xml
@@ -314,10 +343,21 @@ class RVDatabase:
                 if file.startswith(filename):
                     break
             else:
-                log.info(f"Removing: {file}")
+                log.info(f"Removing unknown file: {file}")
                 if not self.args.dryrun:
                     os.unlink(file)
 
+    @property
+    def all_researched(self) -> bool:
+        for element in self.elements:
+            if not element.is_researched:
+                return False
+        return True
+
+    def attempt_clean(self) -> None:
+        if self.all_researched:
+            self.clean()
+
     @property
     def ytdl_opts(self) -> dict:
         return {"format": self.args.format, "allsubtitles": self.args.subtitles}
@@ -435,7 +475,7 @@ def get_args() -> configargparse.Namespace:
     parser.add(
         "action",
         nargs="?",
-        choices=("download", "list", "watch", "binge"),
+        choices=("download", "list", "watch", "binge", "clean"),
         default="download",
     )
 
@@ -466,22 +506,24 @@ def main() -> None:
             database.import_cache(cache)
     if cache:
         database.salvage_cache(cache)
-
+        database.clean_cache(cache)
-    if args.action == "download":
+        database.save()
-        # TODO Clean on watch? / cache import with missing video / all researched
-        database.clean()
 
     log.debug(f"Running action")
+    if args.action == "clean":
+        database.clean()
+    else:
+        database.attempt_clean()
         for element in database.filter(args):
             if args.action == "download":
-            element.act()
+                element.preload()
             elif args.action == "list":
                 print(element)
             elif args.action in ("watch", "binge"):
                 element.watch()
             if args.action == "watch":
                 break
-
+        database.attempt_clean()
     database.save()