feat: add reindex

memos/cmds/library.py

@@ -584,7 +584,7 @@ def reindex(
                             continue
 
                         update_response = client.post(f"{BASE_URL}/entities/{entity['id']}/last-scan-at")
-                        if update_response.status_code != 200:
+                        if update_response.status_code != 204:
                             print(f"Failed to update last_scan_at for entity {entity['id']}: {update_response.status_code} - {update_response.text}")
                         else:
                             print(f"Updated last_scan_at for entity {entity['id']}")

memos/commands.py

@@ -18,7 +18,7 @@ import sys
 import subprocess
 import platform
 from .cmds.plugin import plugin_app, bind
-from .cmds.library import lib_app, scan, index, watch
+from .cmds.library import lib_app, scan, typesense_index, reindex, watch
 import psutil
 import signal
 from tabulate import tabulate
@@ -139,8 +139,8 @@ def scan_default_library(force: bool = False):
     scan(default_library["id"], plugins=None, folders=None, force=force)
 
 
-@app.command("index")
-def index_default_library(
+@app.command("typesense-index")
+def typsense_index_default_library(
     batchsize: int = typer.Option(
         4, "--batchsize", "-bs", help="Number of entities to index in a batch"
     ),
@@ -164,7 +164,32 @@ def index_default_library(
         print("Default library does not exist.")
         return
 
-    index(default_library["id"], force=force, folders=None, batchsize=batchsize)
+    typesense_index(default_library["id"], force=force, folders=None, batchsize=batchsize)
+
+
+@app.command("reindex")
+def reindex_default_library():
+    """
+    Reindex the default library for memos.
+    """
+    # Get the default library
+    response = httpx.get(f"{BASE_URL}/libraries")
+    if response.status_code != 200:
+        print(f"Failed to retrieve libraries: {response.status_code} - {response.text}")
+        return
+
+    libraries = response.json()
+    default_library = next(
+        (lib for lib in libraries if lib["name"] == settings.default_library), None
+    )
+
+    if not default_library:
+        print("Default library does not exist.")
+        return
+
+    # Reindex the library
+    print(f"Reindexing library: {default_library['name']}")
+    reindex(default_library["id"])
 
 
 @app.command("record")

memos/crud.py

@@ -290,12 +290,15 @@ def update_entity(
     return Entity(**db_entity.__dict__)
 
 
-def touch_entity(entity_id: int, db: Session):
+def touch_entity(entity_id: int, db: Session) -> bool:
     db_entity = db.query(EntityModel).filter(EntityModel.id == entity_id).first()
     if db_entity:
         db_entity.last_scan_at = func.now()
         db.commit()
         db.refresh(db_entity)
+        return True
+    else:
+        return False
         
 
 def update_entity_tags(entity_id: int, tags: List[str], db: Session) -> Entity:

memos/embedding.py

@@ -33,7 +33,7 @@ def init_embedding_model():
         model_dir = settings.embedding.model
         logger.info(f"Using model: {model_dir}")
 
-    model = SentenceTransformer(model_dir, trust_remote_code=True)
+    model = SentenceTransformer(model_dir, trust_remote_code=True, truncate_dim=768)
     model.to(device)
     logger.info(f"Embedding model initialized on device: {device}")
 

memos/plugins/ocr/main.py

@@ -7,7 +7,7 @@ import json
 import base64
 from PIL import Image
 import numpy as np
-from rapidocr_openvino import RapidOCR
+from rapidocr_onnxruntime import RapidOCR
 from concurrent.futures import ThreadPoolExecutor
 from functools import partial
 import yaml

memos/server.py

@@ -372,7 +372,7 @@ async def update_entity(
     return entity
 
 
-@app.post("/entities/{entity_id}/last-scan-at", response_model=Entity, tags=["entity"])
+@app.post("/entities/{entity_id}/last-scan-at", status_code=status.HTTP_204_NO_CONTENT, tags=["entity"])
 def update_entity_last_scan_at(
     entity_id: int,
     db: Session = Depends(get_db)
@@ -380,13 +380,12 @@ def update_entity_last_scan_at(
     """
     Update the last_scan_at timestamp for an entity and trigger update for fts and vec.
     """
-    entity = crud.touch_entity(entity_id, db)
-    if entity is None:
+    succeeded = crud.touch_entity(entity_id, db)
+    if not succeeded:
         raise HTTPException(
             status_code=status.HTTP_404_NOT_FOUND,
             detail="Entity not found",
         )
-    return entity
 
 
 def typesense_required(func):