1.1.0 - automations, clean only mode, bug fixes

server/core/file_scanner.py

@@ -18,6 +18,7 @@ logger = get_logger("FileScanner")
 import sys
 sys.path.insert(0, str(Path(__file__).parent))
 from subtitle_processor import parse_srt, SUBLOGUE_SENTINEL, SUBLOGUE_TOKEN_PATTERN
+from keyword_stripper import get_stripper
 
 
 class FileScanner:
@@ -45,6 +46,7 @@ class FileScanner:
         directory_path: str | Path,
         batch_size: int = DEFAULT_BATCH_SIZE,
         follow_symlinks: bool = False,
+        detect_cleanup_keywords: bool = False,
     ) -> Generator[List[Dict], None, None]:
         """
         Recursively scan a directory tree for .srt files.
@@ -106,8 +108,11 @@ class FileScanner:
             # Plot detection
             # --------------------------------------------
 
+            content = None
             try:
-                plot_marker_count = cls._count_plot_markers(file_path)
+                if detect_cleanup_keywords:
+                    content = file_path.read_text(encoding="utf-8", errors="ignore")
+                plot_marker_count = cls._count_plot_markers(file_path, content=content)
                 has_plot = plot_marker_count > 0
                 logger.debug(
                     "Plot check for %s: %s",
@@ -126,7 +131,7 @@ class FileScanner:
 
             if has_plot:
                 try:
-                    metadata = cls._extract_metadata(file_path)
+                    metadata = cls._extract_metadata(file_path, content=content)
                     logger.debug(
                         "Extracted metadata from %s: %s",
                         file_path.name,
@@ -138,6 +143,13 @@ class FileScanner:
                         file_path.name, e
                     )
 
+            clean_keywords = []
+            if detect_cleanup_keywords and content:
+                try:
+                    clean_keywords = get_stripper().detect_subtitle_watermarks(content)
+                except Exception as e:
+                    logger.debug("Cleanup keyword detection failed: %s", e)
+
             status = "Has Plot" if has_plot else "Not Loaded"
             if plot_marker_count > 1:
                 status = "Duplicate Plot"
@@ -156,6 +168,7 @@ class FileScanner:
                 "imdb_rating": metadata.get("imdb_rating"),
                 "rating": metadata.get("imdb_rating"),
                 "runtime": metadata.get("runtime"),
+                "clean_keywords": clean_keywords,
                 "selected": False,
             })
 
@@ -216,14 +229,15 @@ class FileScanner:
                 )
 
     @classmethod
-    def _count_plot_markers(cls, file_path: Path) -> int:
+    def _count_plot_markers(cls, file_path: Path, content: str | None = None) -> int:
         """
         Count Sublogue plot markers to detect duplicates.
         """
         logger.debug("Scanning for plot markers in %s", file_path.name)
 
         try:
-            content = file_path.read_text(encoding="utf-8", errors="ignore")
+            if content is None:
+                content = file_path.read_text(encoding="utf-8", errors="ignore")
             lower_content = content.lower()
             generated_count = lower_content.count("generated by sublogue")
             if generated_count > 0:
@@ -237,14 +251,15 @@ class FileScanner:
             return 0
 
     @classmethod
-    def _extract_metadata(cls, file_path: Path) -> Dict:
+    def _extract_metadata(cls, file_path: Path, content: str | None = None) -> Dict:
         """
         Extract title, year, rating, runtime, and plot
         from Sublogue-generated subtitles.
         """
         logger.debug("Extracting metadata from %s", file_path.name)
 
-        content = file_path.read_text(encoding="utf-8", errors="ignore")
+        if content is None:
+            content = file_path.read_text(encoding="utf-8", errors="ignore")
         blocks = parse_srt(content)
 
         metadata = {