Feature/tweak downgrade logic (#1577)

* tweak downgrade

* fix js sdk

* fix js sdk

* fix upgrade logic

* up

js/sdk/package.json

@@ -1,6 +1,6 @@
 {
   "name": "r2r-js",
-  "version": "0.3.15",
+  "version": "0.3.16",
   "description": "",
   "main": "dist/index.js",
   "browser": "dist/index.browser.js",

js/sdk/src/r2rClient.ts

@@ -1921,41 +1921,24 @@ export class r2rClient {
   /**
    * Search over documents.
    * @param query The query to search for.
-   * @param settings Settings for the document search.
+   * @param vector_search_settings Settings for the document search.
    * @returns A promise that resolves to the response from the server.
    */
   @feature("searchDocuments")
   async searchDocuments(
     query: string,
-    settings?: {
-      searchOverMetadata?: boolean;
-      metadataKeys?: string[];
-      searchOverBody?: boolean;
-      filters?: Record<string, any>;
-      searchFilters?: Record<string, any>;
-      offset?: number;
-      limit?: number;
-      titleWeight?: number;
-      metadataWeight?: number;
-    },
+    vector_search_settings?: VectorSearchSettings | Record<string, any>,
   ): Promise<any> {
     this._ensureAuthenticated();
-
     const json_data: Record<string, any> = {
       query,
-      settings: {
-        search_over_metadata: settings?.searchOverMetadata ?? true,
-        metadata_keys: settings?.metadataKeys ?? ["title"],
-        search_over_body: settings?.searchOverBody ?? false,
-        filters: settings?.filters ?? {},
-        search_filters: settings?.searchFilters ?? {},
-        offset: settings?.offset ?? 0,
-        limit: settings?.limit ?? 10,
-        title_weight: settings?.titleWeight ?? 0.5,
-        metadata_weight: settings?.metadataWeight ?? 0.5,
-      },
+      vector_search_settings,
     };
 
+    Object.keys(json_data).forEach(
+      (key) => json_data[key] === undefined && delete json_data[key],
+    );
+
     return await this._makeRequest("POST", "search_documents", {
       data: json_data,
     });

py/core/main/api/ingestion_router.py

@@ -11,10 +11,10 @@
     Depends,
     File,
     Form,
+    HTTPException,
     Path,
     Query,
     UploadFile,
-    HTTPException,
 )
 from pydantic import Json
 

py/core/main/app.py

@@ -1,11 +1,11 @@
 from typing import Union
 
-from core.base import R2RException
 from fastapi import FastAPI, Request
-from fastapi.responses import JSONResponse
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.openapi.utils import get_openapi
+from fastapi.responses import JSONResponse
 
+from core.base import R2RException
 from core.providers import (
     HatchetOrchestrationProvider,
     SimpleOrchestrationProvider,

py/core/main/app_entry.py

@@ -5,10 +5,11 @@
 from typing import Optional
 
 from apscheduler.schedulers.asyncio import AsyncIOScheduler
-from core.base import R2RException
 from fastapi import FastAPI, Request
-from fastapi.responses import JSONResponse
 from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import JSONResponse
+
+from core.base import R2RException
 
 from .assembly import R2RBuilder, R2RConfig
 

py/core/main/services/ingestion_service.py

@@ -251,6 +251,9 @@ async def augment_document_info(
 
             document_info.summary = response.choices[0].message.content  # type: ignore
 
+            if not document_info.summary:
+                raise ValueError("Expected a generated response.")
+
             embedding = await self.providers.embedding.async_get_embedding(
                 text=document_info.summary,
             )

py/core/main/services/kg_service.py

@@ -3,6 +3,7 @@
 import time
 from typing import AsyncGenerator, Optional
 from uuid import UUID
+
 from fastapi import HTTPException
 
 from core.base import KGExtractionStatus, RunManager

py/core/pipes/kg/deduplication.py

@@ -2,6 +2,7 @@
 import logging
 from typing import Any, Union
 from uuid import UUID
+
 from fastapi import HTTPException
 
 from core.base import AsyncState

py/core/pipes/kg/prompt_tuning.py

@@ -5,6 +5,7 @@
 import logging
 from typing import Any
 from uuid import UUID
+
 from fastapi import HTTPException
 
 from core.base import (

py/core/providers/auth/r2r_auth.py

@@ -1,10 +1,9 @@
 import logging
 import os
 from datetime import datetime, timedelta, timezone
-from fastapi import HTTPException
 
 import jwt
-from fastapi import Depends
+from fastapi import Depends, HTTPException
 from fastapi.security import OAuth2PasswordBearer
 
 from core.base import (

py/core/providers/database/collection.py

@@ -3,6 +3,7 @@
 from datetime import datetime
 from typing import Optional, Union
 from uuid import UUID, uuid4
+
 from fastapi import HTTPException
 
 from core.base import (

py/core/providers/database/document.py

@@ -197,6 +197,12 @@ async def upsert_documents_overview(
                     else:
                         wait_time = 0.1 * (2**retries)  # Exponential backoff
                         await asyncio.sleep(wait_time)
+                except Exception as e:
+                    if 'column "summary"' in str(e):
+                        raise ValueError(
+                            "Document schema is missing 'summary' and 'summary_embedding' columns. Call `r2r db upgrade` to carry out the necessary migration."
+                        )
+                    raise
 
     async def delete_from_documents_overview(
         self, document_id: UUID, version: Optional[str] = None

py/core/providers/database/file.py

@@ -2,9 +2,9 @@
 import logging
 from typing import BinaryIO, Optional, Union
 from uuid import UUID
-from fastapi import HTTPException
 
 import asyncpg
+from fastapi import HTTPException
 
 from core.base import FileHandler, R2RException
 

py/core/providers/database/kg.py

@@ -3,10 +3,10 @@
 import time
 from typing import Any, AsyncGenerator, Optional, Tuple
 from uuid import UUID
-from fastapi import HTTPException
 
 import asyncpg
 from asyncpg.exceptions import PostgresError, UndefinedTableError
+from fastapi import HTTPException
 
 from core.base import (
     CommunityReport,

py/core/providers/database/user.py

@@ -1,6 +1,7 @@
 from datetime import datetime
 from typing import Optional, Union
 from uuid import UUID
+
 from fastapi import HTTPException
 
 from core.base import CryptoProvider, UserHandler