Use uuid5 associated with file SHA1 instead of randomly generated uuid4

app/util/generate_uuid.py

@@ -0,0 +1,20 @@
+import io
+import pathlib
+import hashlib
+from uuid import UUID, uuid5, NAMESPACE_DNS
+
+NAMESPACE_STR = 'github.com/hv0905/NekoImageGallary'
+
+
+def generate(file_input: pathlib.Path | io.BytesIO) -> UUID:
+    namespace_uuid = uuid5(NAMESPACE_DNS, NAMESPACE_STR)
+    if isinstance(file_input, pathlib.Path):
+        with open(file_input, 'rb') as f:
+            file_content = f.read()
+    elif isinstance(file_input, io.BytesIO):
+        file_input.seek(0)
+        file_content = file_input.read()
+    else:
+        raise ValueError("Unsupported file type. Must be pathlib.Path or io.BytesIO.")
+    file_hash = hashlib.sha1(file_content).hexdigest()
+    return uuid5(namespace_uuid, file_hash)

scripts/local_indexing.py

@@ -1,7 +1,6 @@
 from datetime import datetime
 from pathlib import Path
 from shutil import copy2
-from uuid import uuid4
 
 import PIL
 from PIL import Image
@@ -10,6 +9,7 @@
 from app.Models.img_data import ImageData
 from app.Services.provider import index_service
 from app.config import config
+from app.util import generate_uuid
 from .local_utility import gather_valid_files
 
 
@@ -19,7 +19,7 @@ async def copy_and_index(file_path: Path):
     except PIL.UnidentifiedImageError as e:
         logger.error("Error when opening image {}: {}", file_path, e)
         return
-    image_id = uuid4()
+    image_id = generate_uuid.generate(file_path)
     img_ext = file_path.suffix
     imgdata = ImageData(id=image_id,
                         url=f'/static/{image_id}{img_ext}',