Merge pull request #249 from stg-annon/PDT-Fix

[phashDuplicateTagger] Bugfix + Add Compare to Scene paths + Untrack config.py

plugins/phashDuplicateTagger/.gitignore

@@ -0,0 +1 @@
+config.py

plugins/phashDuplicateTagger/config_example.py

@@ -1,3 +1,4 @@
+from pathlib import Path
 import stashapi.log as log
 from stashapi.tools import human_bytes, human_bits
 
@@ -14,6 +15,8 @@ CODEC_PRIORITY = {
     "VC1": 6,
     "SVQ3": 7,
 }
+# Path priority is from highest to lowest and works off the root of the path, to enable add "path" to the PRIORITY list
+PATH_PRIORITY = ["/root/most/important/path","/root/least/important/path"]
 
 KEEP_TAG_NAME = "[PDT: Keep]"
 REMOVE_TAG_NAME = "[PDT: Remove]"
@@ -113,18 +116,18 @@ def compare_size(self, other):
 
 
 def compare_age(self, other):
-    if not (self.mod_time and other.mod_time):
+    if not (self.created_at and other.created_at):
         return
-    if self.mod_time == other.mod_time:
+    if self.created_at == other.created_at:
         return
-    if self.mod_time < other.mod_time:
+    if self.created_at < other.created_at:
         better, worse = self, other
     else:
         worse, better = self, other
     worse.remove_reason = "age"
     return (
         better,
-        f"Choose Oldest: Δ:{worse.mod_time-better.mod_time} | {better.id} older than {worse.id}",
+        f"Choose Oldest: Δ:{worse.created_at-better.created_at} | {better.id} older than {worse.id}",
     )
 
 
@@ -145,3 +148,33 @@ def compare_encoding(self, other):
         self,
         f"Prefer Codec {better.codec}({better.id}) over {worse.codec}({worse.id})",
     )
+
+
+def compare_path(self, other):
+    if PATH_PRIORITY[0] == "/root/most/important/path":
+        return
+    if not self.path or not other.path:
+        return
+    
+    self.path = Path(self.path)
+    other.path = Path(other.path)
+
+    min_score = len(PATH_PRIORITY)
+    self.score = min_score
+    other.score = min_score
+    for score, path in enumerate(PATH_PRIORITY):
+        path = Path(path)
+        if path in self.path.parents:
+            self.score = score
+        if path in other.path.parents:
+            other.score = score
+
+    if self.score == other.score:
+        return
+
+    if self.score < other.score:
+        better, worse = self, other
+    else:
+        worse, better = self, other
+    worse.remove_reason = "filepath"
+    return better, f"Prefer Filepath {PATH_PRIORITY[better.score]} | {better.id} better than {worse.id}"

plugins/phashDuplicateTagger/phashDuplicateTagger.py

@@ -12,8 +12,10 @@ except ModuleNotFoundError:
         "You need to install the stashapi module. (pip install stashapp-tools)",
         file=sys.stderr,
     )
-
-import config
+try:
+    import config
+except ModuleNotFoundError:
+    log.error("Could not import 'config.py' did you copy and rename 'config_example.py'?")
 
 FRAGMENT = json.loads(sys.stdin.read())
 MODE = FRAGMENT["args"]["mode"]
@@ -30,7 +32,7 @@ files {
 	width
 	height
 	bit_rate
-	mod_time
+	created_at
 	duration
 	frame_rate
 	video_codec
@@ -68,10 +70,12 @@ def parse_timestamp(ts, format="%Y-%m-%dT%H:%M:%S%z"):
 class StashScene:
 
     def __init__(self, scene=None) -> None:
+        if len(scene["files"]) != 1:
+            raise Exception(f"Scene has {len(scene['files'])} scene must have one file for comparing")
         file = scene["files"][0]
 
         self.id = int(scene["id"])
-        self.mod_time = parse_timestamp(file["mod_time"])
+        self.created_at = parse_timestamp(file["created_at"])
         if scene.get("date"):
             self.date = parse_timestamp(scene["date"], format="%Y-%m-%d")
         else:
@@ -102,7 +106,7 @@ class StashScene:
         return f"<StashScene ({self.id})>"
 
     def __str__(self) -> str:
-        return f"id:{self.id}, height:{self.height}, size:{human_bytes(self.size)}, file_mod_time:{self.mod_time}, title:{self.title}"
+        return f"id:{self.id}, height:{self.height}, size:{human_bytes(self.size)}, file_created_at:{self.created_at}, title:{self.title}"
 
     def compare(self, other):
         if not (isinstance(other, StashScene)):
@@ -138,9 +142,14 @@ def process_duplicates(distance: PhashDistance = PhashDistance.EXACT):
     log.info(f"Found {total} sets of duplicates.")
 
     for i, group in enumerate(duplicate_list):
-        group = [StashScene(s) for s in group]
+        scene_group = []
+        for s in group:
+            try:
+                scene_group.append(StashScene(s))
+            except Exception as e:
+                log.warning(f"Issue parsing SceneID:{s['id']} - {e}")
         filtered_group = []
-        for scene in group:
+        for scene in scene_group:
             if ignore_tag_id in scene.tag_ids:
                 log.debug(f"Ignore {scene.id} {scene.title}")
             else:
@@ -155,12 +164,12 @@ def process_duplicates(distance: PhashDistance = PhashDistance.EXACT):
 def tag_files(group):
 
     keep_reasons = []
-    keep_scene = None
+    keep_scene = group[0]
 
     total_size = group[0].size
     for scene in group[1:]:
         total_size += scene.size
-        better, msg = scene.compare(group[0])
+        better, msg = scene.compare(keep_scene)
         if better:
             keep_scene = better
             keep_reasons.append(msg)