remove functions from reference_playlists_to_library.py move to serverlibary

serverlibrary/__init__.py

@@ -2,3 +2,8 @@ from .core import safe_rename
 from .core import safe_delete
 from .core import get_audio_tracks
 from .core import get_subtitle_tracks
+from .core import words
+from .core import word_similarity
+from .core import score_match_jaccard
+from .core import score_match_fuzzy
+from .core import combined_score

serverlibrary/core.py

@@ -5,6 +5,7 @@ import subprocess
 import json
 import re
 from typing import Any
+from difflib import SequenceMatcher
 
 
 # -----------------------------
@@ -16,7 +17,8 @@ def safe_delete(path: Path):
     else:
         path.unlink()
 
-def safe_rename(src: Path, dst: Path, dry: bool=False, confirm: bool=False, silent=False):
+def safe_rename(src: Path, dst: Path, dry: bool=False, confirm: bool=False,
+                silent=False,overwrite=False) -> (str, bool):
     """
        Safely rename a file or directory without overwriting existing files.
        Automatically appends _1, _2, ... if needed.
@@ -45,14 +47,18 @@ def safe_rename(src: Path, dst: Path, dry: bool=False, confirm: bool=False, sile
         print(ret)
 
     if dry:
-        return
+        return ret, True
 
     if confirm:
         input("Press Enter to confirm...")
 
     dst.parent.mkdir(parents=True, exist_ok=True)
+
+    if dst.exists() and not overwrite:
+        print("File already exists...")
+        return "",False
     src.rename(dst)
-    return ret
+    return ret, True
 
 def get_audio_tracks( source: str, title: int | None = None, handbrake: bool = False) -> list[dict]:
     """
@@ -197,4 +203,42 @@ def get_subtitle_tracks(source):
             "primary": False,
         })
 
-    return tracks
+    return tracks
+
+
+def words(text):
+    return set(filter(None, re.split(r"[^a-z0-9]+", text.lower())))
+
+
+def word_similarity(a: str, b: str) -> float:
+    return SequenceMatcher(None, a, b).ratio()
+
+
+def score_match_jaccard(search_words, candidate_words):
+    intersection = search_words & candidate_words
+    union = search_words | candidate_words
+
+    if not union:
+        return 0.0
+
+    return len(intersection) / len(union)
+
+
+def score_match_fuzzy(search_words, candidate_words):
+    total = 0.0
+
+    for sw in search_words:
+        best = 0.0
+
+        for cw in candidate_words:
+            best = max(best, word_similarity(sw, cw))
+
+        total += best
+
+    #    logger.trace(f"Total: {total}/{len(search_words)}")
+    return total / len(search_words)
+
+
+def combined_score(score_fuzzy: float, score_jaccard: float,
+                   weight_fuzzy: float, weight_jaccard: float) -> float:
+    return score_fuzzy * weight_fuzzy + score_jaccard * weight_jaccard