A bit more progress for unified episode identification

src/show.py

@@ -1,11 +1,15 @@
 from concurrent.futures import ProcessPoolExecutor, as_completed
 from dataclasses import dataclass, field
+import math
+import os
+from pathlib import Path
 import re
 import numpy as np
 import cv2
 import logging
+import imdbinfo as imdb
 
-from models import PathInfo
+from models import FileCategory, FileInfo, PathInfo
 import tmdb
 
 
@@ -242,54 +246,45 @@ def match_episodes_to_references(
 
 def process_show(info: PathInfo) -> PathInfo:
     # Gather meta information for identifying episodes
-    episode_durations: set[int] = set()
+    episode_durations: set[int] = (
+        set()
+    )  # TODO: Store duration per specific episode for unique duration identification
+    episode_count = 0  # TODO: Per season
     for ep in imdb.get_all_episodes(info.imdb.imdb_id):
+        episode_count += 1
         if ep.duration:
             episode_durations.add(int(ep.duration / 60))
 
-    logging.debug(episode_durations)
-
-    # Go over all files
-    count = 0
-    for root, dirs, files in os.walk(info.path):
-        for filename in files:
-            filepath = Path(os.path.join(root, filename))
-            file = FileInfo(filepath, info)
-
-            if file.path.suffix != ".mkv" or file.duration_in_seconds is None:
-                continue
-            if (
-                math.floor(file.duration_in_seconds / 60) in episode_durations
-                or math.ceil(file.duration_in_seconds / 60) in episode_durations
-            ):
-                print(f"{filename} {file.duration_in_seconds / 60} EPISODE")
-                count += 1
-            else:
-                print(f"{filename} {file.duration_in_seconds / 60}")
-
-    logging.info(f"Identified [{count}] episodes.")
-    
-    
-    
-    
-    
-    
-    
-    
-    
-    
-    if info.episodes is None:
-        logging.error(
-            "Episodes could not be identified, no reference matching possible."
-        )
-        return info
+    logging.debug(f"Episode durations: {episode_durations}")
 
+    # Get reference images for episode identification
     if info.tmdb_id is None:
         logging.error("TMDB entry not identified, cannot find reference images.")
         return info
-
-    # Match episodes to references
     references = tmdb.download_episode_images(info.tmdb_id)
+
+    # Go over all files
+    for root, dirs, files in os.walk(info.path):
+        # TODO: Detect season in dirs
+
+        for filename in files:
+            filepath = Path(os.path.join(root, filename))
+            file = FileInfo(filepath, info)
+            info.files.append(file)
+
+            # Is file an episode?
+            if file.path.suffix != ".mkv" or file.duration_in_seconds is None:
+                continue
+            elif (
+                math.floor(file.duration_in_seconds / 60) not in episode_durations
+                and math.ceil(file.duration_in_seconds / 60) not in episode_durations
+            ):
+                continue
+
+            # ==== Handle episode ====
+            file.category = FileCategory.EPISODE
+
+            # TODO: Match episode to references
             matches = match_episodes_to_references(
                 [
                     str(f.path.absolute())
@@ -298,7 +293,10 @@ def process_show(info: PathInfo) -> PathInfo:
                 ],
                 references.flatten(),
             )
-    logging.info(matches)
+
+    if info.episodes is None:
+        logging.error("Episodes could not be identified.")
+        return info
 
     # Set new episode names
     if not matches.perfect_match or matches.reference_by_episode is None: