gardel/lightrag
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Remove Oracle storage implementation

Browse Source
This commit is contained in:
yangdx
2025-04-01 18:15:29 +08:00
parent aec302de4e
commit 1e31b26cbe
10 changed files with 21 additions and 1946 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
README-zh.md
View File

@@ -11,7 +11,6 @@
- [X] [2024.12.31]🎯📢LightRAG现在支持[通过文档ID删除](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#delete)。
- [X] [2024.11.25]🎯📢LightRAG现在支持无缝集成[自定义知识图谱](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#insert-custom-kg)，使用户能够用自己的领域专业知识增强系统。
- [X] [2024.11.19]🎯📢LightRAG的综合指南现已在[LearnOpenCV](https://learnopencv.com/lightrag)上发布。非常感谢博客作者。
- [X] [2024.11.12]🎯📢LightRAG现在支持[Oracle Database 23ai的所有存储类型（KV、向量和图）](https://github.com/HKUDS/LightRAG/blob/main/examples/lightrag_oracle_demo.py)。
- [X] [2024.11.11]🎯📢LightRAG现在支持[通过实体名称删除实体](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#delete)。
- [X] [2024.11.09]🎯📢推出[LightRAG Gui](https://lightrag-gui.streamlit.app)，允许您插入、查询、可视化和下载LightRAG知识。
- [X] [2024.11.04]🎯📢现在您可以[使用Neo4J进行存储](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#using-neo4j-for-storage)。
@@ -1037,9 +1036,10 @@ rag.clear_cache(modes=["local"])
| **参数** | **类型** | **说明** | **默认值** |
|--------------|----------|-----------------|-------------|
| **working_dir** | `str` | 存储缓存的目录 | `lightrag_cache+timestamp` |
| **kv_storage** | `str` | 文档和文本块的存储类型。支持的类型：`JsonKVStorage`、`OracleKVStorage` | `JsonKVStorage` |
| **vector_storage** | `str` | 嵌入向量的存储类型。支持的类型：`NanoVectorDBStorage`、`OracleVectorDBStorage` | `NanoVectorDBStorage` |
| **graph_storage** | `str` | 图边和节点的存储类型。支持的类型：`NetworkXStorage`、`Neo4JStorage`、`OracleGraphStorage` | `NetworkXStorage` |
| **kv_storage** | `str` | Storage type for documents and text chunks. Supported types: `JsonKVStorage`,`PGKVStorage`,`RedisKVStorage`,`MongoKVStorage`,`TiDBKVStorage` | `JsonKVStorage` |
| **vector_storage** | `str` | Storage type for embedding vectors. Supported types: `NanoVectorDBStorage`,`PGVectorStorage`,`MilvusVectorDBStorage`,`ChromaVectorDBStorage`,`FaissVectorDBStorage`,`TiDBVectorDBStorage`,`MongoVectorDBStorage`,`QdrantVectorDBStorage` | `NanoVectorDBStorage` |
| **graph_storage** | `str` | Storage type for graph edges and nodes. Supported types: `NetworkXStorage`,`Neo4JStorage`,`PGGraphStorage`,`AGEStorage`,`GremlinStorage` | `NetworkXStorage` |
| **doc_status_storage** | `str` | Storage type for documents process status. Supported types: `JsonDocStatusStorage`,`PGDocStatusStorage`,`MongoDocStatusStorage` | `JsonDocStatusStorage` |
| **chunk_token_size** | `int` | 拆分文档时每个块的最大令牌大小 | `1200` |
| **chunk_overlap_token_size** | `int` | 拆分文档时两个块之间的重叠令牌大小 | `100` |
| **tiktoken_model_name** | `str` | 用于计算令牌数的Tiktoken编码器的模型名称 | `gpt-4o-mini` |

8
README.md
View File

@@ -41,7 +41,6 @@
- [X] [2024.12.31]🎯📢LightRAG now supports [deletion by document ID](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#delete).
- [X] [2024.11.25]🎯📢LightRAG now supports seamless integration of [custom knowledge graphs](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#insert-custom-kg), empowering users to enhance the system with their own domain expertise.
- [X] [2024.11.19]🎯📢A comprehensive guide to LightRAG is now available on [LearnOpenCV](https://learnopencv.com/lightrag). Many thanks to the blog author.
- [X] [2024.11.12]🎯📢LightRAG now supports [Oracle Database 23ai for all storage types (KV, vector, and graph)](https://github.com/HKUDS/LightRAG/blob/main/examples/lightrag_oracle_demo.py).
- [X] [2024.11.11]🎯📢LightRAG now supports [deleting entities by their names](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#delete).
- [X] [2024.11.09]🎯📢Introducing the [LightRAG Gui](https://lightrag-gui.streamlit.app), which allows you to insert, query, visualize, and download LightRAG knowledge.
- [X] [2024.11.04]🎯📢You can now [use Neo4J for Storage](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#using-neo4j-for-storage).
@@ -1065,9 +1064,10 @@ Valid modes are:
| **Parameter** | **Type** | **Explanation** | **Default** |
|--------------|----------|-----------------|-------------|
| **working_dir** | `str` | Directory where the cache will be stored | `lightrag_cache+timestamp` |
| **kv_storage** | `str` | Storage type for documents and text chunks. Supported types: `JsonKVStorage`, `OracleKVStorage` | `JsonKVStorage` |
| **vector_storage** | `str` | Storage type for embedding vectors. Supported types: `NanoVectorDBStorage`, `OracleVectorDBStorage` | `NanoVectorDBStorage` |
| **graph_storage** | `str` | Storage type for graph edges and nodes. Supported types: `NetworkXStorage`, `Neo4JStorage`, `OracleGraphStorage` | `NetworkXStorage` |
| **kv_storage** | `str` | Storage type for documents and text chunks. Supported types: `JsonKVStorage`,`PGKVStorage`,`RedisKVStorage`,`MongoKVStorage`,`TiDBKVStorage` | `JsonKVStorage` |
| **vector_storage** | `str` | Storage type for embedding vectors. Supported types: `NanoVectorDBStorage`,`PGVectorStorage`,`MilvusVectorDBStorage`,`ChromaVectorDBStorage`,`FaissVectorDBStorage`,`TiDBVectorDBStorage`,`MongoVectorDBStorage`,`QdrantVectorDBStorage` | `NanoVectorDBStorage` |
| **graph_storage** | `str` | Storage type for graph edges and nodes. Supported types: `NetworkXStorage`,`Neo4JStorage`,`PGGraphStorage`,`AGEStorage`,`GremlinStorage` | `NetworkXStorage` |
| **doc_status_storage** | `str` | Storage type for documents process status. Supported types: `JsonDocStatusStorage`,`PGDocStatusStorage`,`MongoDocStatusStorage` | `JsonDocStatusStorage` |
| **chunk_token_size** | `int` | Maximum token size per chunk when splitting documents | `1200` |
| **chunk_overlap_token_size** | `int` | Overlap token size between two chunks when splitting documents | `100` |
| **tiktoken_model_name** | `str` | Model name for the Tiktoken encoder used to calculate token numbers | `gpt-4o-mini` |

9
config.ini.example
View File

@@ -13,15 +13,6 @@ uri=redis://localhost:6379/1
[qdrant]
uri = http://localhost:16333
[oracle]
dsn = localhost:1521/XEPDB1
user = your_username
password = your_password
config_dir = /path/to/oracle/config
wallet_location = /path/to/wallet # 可选
wallet_password = your_wallet_password # 可选
workspace = default # 可选,默认为default
[tidb]
host = localhost
port = 4000

10
env.example
View File

@@ -109,16 +109,6 @@ LIGHTRAG_VECTOR_STORAGE=NanoVectorDBStorage
LIGHTRAG_GRAPH_STORAGE=NetworkXStorage
LIGHTRAG_DOC_STATUS_STORAGE=JsonDocStatusStorage
### Oracle Database Configuration
ORACLE_DSN=localhost:1521/XEPDB1
ORACLE_USER=your_username
ORACLE_PASSWORD='your_password'
ORACLE_CONFIG_DIR=/path/to/oracle/config
#ORACLE_WALLET_LOCATION=/path/to/wallet
#ORACLE_WALLET_PASSWORD='your_password'
### separating all data from difference Lightrag instances(deprecating)
#ORACLE_WORKSPACE=default
### TiDB Configuration
TIDB_HOST=localhost
TIDB_PORT=4000

267
examples/lightrag_api_oracle_demo.py
View File

@@ -1,267 +0,0 @@
from fastapi import FastAPI, HTTPException, File, UploadFile
from fastapi import Query
from contextlib import asynccontextmanager
from pydantic import BaseModel
from typing import Optional, Any
import sys
import os
from pathlib import Path
import asyncio
import nest_asyncio
from lightrag import LightRAG, QueryParam
from lightrag.llm.openai import openai_complete_if_cache, openai_embed
from lightrag.utils import EmbeddingFunc
import numpy as np
from lightrag.kg.shared_storage import initialize_pipeline_status
print(os.getcwd())
script_directory = Path(__file__).resolve().parent.parent
sys.path.append(os.path.abspath(script_directory))
# Apply nest_asyncio to solve event loop issues
nest_asyncio.apply()
DEFAULT_RAG_DIR = "index_default"
# We use OpenAI compatible API to call LLM on Oracle Cloud
# More docs here https://github.com/jin38324/OCI_GenAI_access_gateway
BASE_URL = "http://xxx.xxx.xxx.xxx:8088/v1/"
APIKEY = "ocigenerativeai"
# Configure working directory
WORKING_DIR = os.environ.get("RAG_DIR", f"{DEFAULT_RAG_DIR}")
print(f"WORKING_DIR: {WORKING_DIR}")
LLM_MODEL = os.environ.get("LLM_MODEL", "cohere.command-r-plus-08-2024")
print(f"LLM_MODEL: {LLM_MODEL}")
EMBEDDING_MODEL = os.environ.get("EMBEDDING_MODEL", "cohere.embed-multilingual-v3.0")
print(f"EMBEDDING_MODEL: {EMBEDDING_MODEL}")
EMBEDDING_MAX_TOKEN_SIZE = int(os.environ.get("EMBEDDING_MAX_TOKEN_SIZE", 512))
print(f"EMBEDDING_MAX_TOKEN_SIZE: {EMBEDDING_MAX_TOKEN_SIZE}")
if not os.path.exists(WORKING_DIR):
os.mkdir(WORKING_DIR)
os.environ["ORACLE_USER"] = ""
os.environ["ORACLE_PASSWORD"] = ""
os.environ["ORACLE_DSN"] = ""
os.environ["ORACLE_CONFIG_DIR"] = "path_to_config_dir"
os.environ["ORACLE_WALLET_LOCATION"] = "path_to_wallet_location"
os.environ["ORACLE_WALLET_PASSWORD"] = "wallet_password"
os.environ["ORACLE_WORKSPACE"] = "company"
async def llm_model_func(
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
) -> str:
return await openai_complete_if_cache(
LLM_MODEL,
prompt,
system_prompt=system_prompt,
history_messages=history_messages,
api_key=APIKEY,
base_url=BASE_URL,
**kwargs,
)
async def embedding_func(texts: list[str]) -> np.ndarray:
return await openai_embed(
texts,
model=EMBEDDING_MODEL,
api_key=APIKEY,
base_url=BASE_URL,
)
async def get_embedding_dim():
test_text = ["This is a test sentence."]
embedding = await embedding_func(test_text)
embedding_dim = embedding.shape[1]
return embedding_dim
async def init():
# Detect embedding dimension
embedding_dimension = await get_embedding_dim()
print(f"Detected embedding dimension: {embedding_dimension}")
# Create Oracle DB connection
# The `config` parameter is the connection configuration of Oracle DB
# More docs here https://python-oracledb.readthedocs.io/en/latest/user_guide/connection_handling.html
# We storage data in unified tables, so we need to set a `workspace` parameter to specify which docs we want to store and query
# Below is an example of how to connect to Oracle Autonomous Database on Oracle Cloud
# Initialize LightRAG
# We use Oracle DB as the KV/vector/graph storage
rag = LightRAG(
enable_llm_cache=False,
working_dir=WORKING_DIR,
chunk_token_size=512,
llm_model_func=llm_model_func,
embedding_func=EmbeddingFunc(
embedding_dim=embedding_dimension,
max_token_size=512,
func=embedding_func,
),
graph_storage="OracleGraphStorage",
kv_storage="OracleKVStorage",
vector_storage="OracleVectorDBStorage",
)
await rag.initialize_storages()
await initialize_pipeline_status()
return rag
# Extract and Insert into LightRAG storage
# with open("./dickens/book.txt", "r", encoding="utf-8") as f:
# await rag.ainsert(f.read())
# # Perform search in different modes
# modes = ["naive", "local", "global", "hybrid"]
# for mode in modes:
# print("="*20, mode, "="*20)
# print(await rag.aquery("这篇文档是关于什么内容的?", param=QueryParam(mode=mode)))
# print("-"*100, "\n")
# Data models
class QueryRequest(BaseModel):
query: str
mode: str = "hybrid"
only_need_context: bool = False
only_need_prompt: bool = False
class DataRequest(BaseModel):
limit: int = 100
class InsertRequest(BaseModel):
text: str
class Response(BaseModel):
status: str
data: Optional[Any] = None
message: Optional[str] = None
# API routes
rag = None
@asynccontextmanager
async def lifespan(app: FastAPI):
global rag
rag = await init()
print("done!")
yield
app = FastAPI(
title="LightRAG API", description="API for RAG operations", lifespan=lifespan
)
@app.post("/query", response_model=Response)
async def query_endpoint(request: QueryRequest):
# try:
# loop = asyncio.get_event_loop()
if request.mode == "naive":
top_k = 3
else:
top_k = 60
result = await rag.aquery(
request.query,
param=QueryParam(
mode=request.mode,
only_need_context=request.only_need_context,
only_need_prompt=request.only_need_prompt,
top_k=top_k,
),
)
return Response(status="success", data=result)
# except Exception as e:
# raise HTTPException(status_code=500, detail=str(e))
@app.get("/data", response_model=Response)
async def query_all_nodes(type: str = Query("nodes"), limit: int = Query(100)):
if type == "nodes":
result = await rag.chunk_entity_relation_graph.get_all_nodes(limit=limit)
elif type == "edges":
result = await rag.chunk_entity_relation_graph.get_all_edges(limit=limit)
elif type == "statistics":
result = await rag.chunk_entity_relation_graph.get_statistics()
return Response(status="success", data=result)
@app.post("/insert", response_model=Response)
async def insert_endpoint(request: InsertRequest):
try:
loop = asyncio.get_event_loop()
await loop.run_in_executor(None, lambda: rag.insert(request.text))
return Response(status="success", message="Text inserted successfully")
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))
@app.post("/insert_file", response_model=Response)
async def insert_file(file: UploadFile = File(...)):
try:
file_content = await file.read()
# Read file content
try:
content = file_content.decode("utf-8")
except UnicodeDecodeError:
# If UTF-8 decoding fails, try other encodings
content = file_content.decode("gbk")
# Insert file content
loop = asyncio.get_event_loop()
await loop.run_in_executor(None, lambda: rag.insert(content))
return Response(
status="success",
message=f"File content from {file.filename} inserted successfully",
)
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))
@app.get("/health")
async def health_check():
return {"status": "healthy"}
if __name__ == "__main__":
import uvicorn
uvicorn.run(app, host="127.0.0.1", port=8020)
# Usage example
# To run the server, use the following command in your terminal:
# python lightrag_api_openai_compatible_demo.py
# Example requests:
# 1. Query:
# curl -X POST "http://127.0.0.1:8020/query" -H "Content-Type: application/json" -d '{"query": "your query here", "mode": "hybrid"}'
# 2. Insert text:
# curl -X POST "http://127.0.0.1:8020/insert" -H "Content-Type: application/json" -d '{"text": "your text here"}'
# 3. Insert file:
# curl -X POST "http://127.0.0.1:8020/insert_file" -H "Content-Type: multipart/form-data" -F "file=@path/to/your/file.txt"
# 4. Health check:
# curl -X GET "http://127.0.0.1:8020/health"

141
examples/lightrag_oracle_demo.py
View File

@@ -1,141 +0,0 @@
import sys
import os
from pathlib import Path
import asyncio
from lightrag import LightRAG, QueryParam
from lightrag.llm.openai import openai_complete_if_cache, openai_embed
from lightrag.utils import EmbeddingFunc
import numpy as np
from lightrag.kg.shared_storage import initialize_pipeline_status
print(os.getcwd())
script_directory = Path(__file__).resolve().parent.parent
sys.path.append(os.path.abspath(script_directory))
WORKING_DIR = "./dickens"
# We use OpenAI compatible API to call LLM on Oracle Cloud
# More docs here https://github.com/jin38324/OCI_GenAI_access_gateway
BASE_URL = "http://xxx.xxx.xxx.xxx:8088/v1/"
APIKEY = "ocigenerativeai"
CHATMODEL = "cohere.command-r-plus"
EMBEDMODEL = "cohere.embed-multilingual-v3.0"
CHUNK_TOKEN_SIZE = 1024
MAX_TOKENS = 4000
if not os.path.exists(WORKING_DIR):
os.mkdir(WORKING_DIR)
os.environ["ORACLE_USER"] = "username"
os.environ["ORACLE_PASSWORD"] = "xxxxxxxxx"
os.environ["ORACLE_DSN"] = "xxxxxxx_medium"
os.environ["ORACLE_CONFIG_DIR"] = "path_to_config_dir"
os.environ["ORACLE_WALLET_LOCATION"] = "path_to_wallet_location"
os.environ["ORACLE_WALLET_PASSWORD"] = "wallet_password"
os.environ["ORACLE_WORKSPACE"] = "company"
async def llm_model_func(
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
) -> str:
return await openai_complete_if_cache(
CHATMODEL,
prompt,
system_prompt=system_prompt,
history_messages=history_messages,
api_key=APIKEY,
base_url=BASE_URL,
**kwargs,
)
async def embedding_func(texts: list[str]) -> np.ndarray:
return await openai_embed(
texts,
model=EMBEDMODEL,
api_key=APIKEY,
base_url=BASE_URL,
)
async def get_embedding_dim():
test_text = ["This is a test sentence."]
embedding = await embedding_func(test_text)
embedding_dim = embedding.shape[1]
return embedding_dim
async def initialize_rag():
# Detect embedding dimension
embedding_dimension = await get_embedding_dim()
print(f"Detected embedding dimension: {embedding_dimension}")
# Initialize LightRAG
# We use Oracle DB as the KV/vector/graph storage
# You can add `addon_params={"example_number": 1, "language": "Simplfied Chinese"}` to control the prompt
rag = LightRAG(
# log_level="DEBUG",
working_dir=WORKING_DIR,
entity_extract_max_gleaning=1,
enable_llm_cache=True,
enable_llm_cache_for_entity_extract=True,
embedding_cache_config=None, # {"enabled": True,"similarity_threshold": 0.90},
chunk_token_size=CHUNK_TOKEN_SIZE,
llm_model_max_token_size=MAX_TOKENS,
llm_model_func=llm_model_func,
embedding_func=EmbeddingFunc(
embedding_dim=embedding_dimension,
max_token_size=500,
func=embedding_func,
),
graph_storage="OracleGraphStorage",
kv_storage="OracleKVStorage",
vector_storage="OracleVectorDBStorage",
addon_params={
"example_number": 1,
"language": "Simplfied Chinese",
"entity_types": ["organization", "person", "geo", "event"],
"insert_batch_size": 2,
},
)
await rag.initialize_storages()
await initialize_pipeline_status()
return rag
async def main():
try:
# Initialize RAG instance
rag = await initialize_rag()
# Extract and Insert into LightRAG storage
with open(WORKING_DIR + "/docs.txt", "r", encoding="utf-8") as f:
all_text = f.read()
texts = [x for x in all_text.split("\n") if x]
# New mode use pipeline
await rag.apipeline_enqueue_documents(texts)
await rag.apipeline_process_enqueue_documents()
# Old method use ainsert
# await rag.ainsert(texts)
# Perform search in different modes
modes = ["naive", "local", "global", "hybrid"]
for mode in modes:
print("=" * 20, mode, "=" * 20)
print(
await rag.aquery(
"What are the top themes in this story?",
param=QueryParam(mode=mode),
)
)
print("-" * 100, "\n")
except Exception as e:
print(f"An error occurred: {e}")
if __name__ == "__main__":
asyncio.run(main())

17
lightrag/api/README-zh.md
View File

@@ -291,11 +291,10 @@ LightRAG 使用 4 种类型的存储用于不同目的：
```
JsonKVStorage JsonFile(默认)
MongoKVStorage MogonDB
RedisKVStorage Redis
TiDBKVStorage TiDB
PGKVStorage Postgres
OracleKVStorage Oracle
RedisKVStorage Redis
MongoKVStorage MogonDB
TiDBKVStorage TiDB
```
* GRAPH_STORAGE 支持的实现名称
@@ -303,25 +302,21 @@ OracleKVStorage Oracle
```
NetworkXStorage NetworkX(默认)
Neo4JStorage Neo4J
MongoGraphStorage MongoDB
TiDBGraphStorage TiDB
PGGraphStorage Postgres
AGEStorage AGE
GremlinStorage Gremlin
PGGraphStorage Postgres
OracleGraphStorage Postgres
```
* VECTOR_STORAGE 支持的实现名称
```
NanoVectorDBStorage NanoVector(默认)
PGVectorStorage Postgres
MilvusVectorDBStorge Milvus
ChromaVectorDBStorage Chroma
TiDBVectorDBStorage TiDB
PGVectorStorage Postgres
FaissVectorDBStorage Faiss
QdrantVectorDBStorage Qdrant
OracleVectorDBStorage Oracle
TiDBVectorDBStorage TiDB
MongoVectorDBStorage MongoDB
```

19
lightrag/api/README.md
View File

@@ -302,11 +302,10 @@ Each storage type have servals implementations:
```
JsonKVStorage JsonFile(default)
MongoKVStorage MogonDB
RedisKVStorage Redis
TiDBKVStorage TiDB
PGKVStorage Postgres
OracleKVStorage Oracle
RedisKVStorage Redis
MongoKVStorage MogonDB
TiDBKVStorage TiDB
```
* GRAPH_STORAGE supported implement-name
@@ -314,25 +313,21 @@ OracleKVStorage Oracle
```
NetworkXStorage NetworkX(defualt)
Neo4JStorage Neo4J
MongoGraphStorage MongoDB
TiDBGraphStorage TiDB
PGGraphStorage Postgres
AGEStorage AGE
GremlinStorage Gremlin
PGGraphStorage Postgres
OracleGraphStorage Postgres
```
* VECTOR_STORAGE supported implement-name
```
NanoVectorDBStorage NanoVector(default)
MilvusVectorDBStorage Milvus
ChromaVectorDBStorage Chroma
TiDBVectorDBStorage TiDB
PGVectorStorage Postgres
MilvusVectorDBStorge Milvus
ChromaVectorDBStorage Chroma
FaissVectorDBStorage Faiss
QdrantVectorDBStorage Qdrant
OracleVectorDBStorage Oracle
TiDBVectorDBStorage TiDB
MongoVectorDBStorage MongoDB
```

25
lightrag/kg/__init__.py
View File

@@ -6,7 +6,6 @@ STORAGE_IMPLEMENTATIONS = {
"RedisKVStorage",
"TiDBKVStorage",
"PGKVStorage",
"OracleKVStorage",
],
"required_methods": ["get_by_id", "upsert"],
},
@@ -19,7 +18,6 @@ STORAGE_IMPLEMENTATIONS = {
"AGEStorage",
"GremlinStorage",
"PGGraphStorage",
# "OracleGraphStorage",
],
"required_methods": ["upsert_node", "upsert_edge"],
},
@@ -32,7 +30,6 @@ STORAGE_IMPLEMENTATIONS = {
"PGVectorStorage",
"FaissVectorDBStorage",
"QdrantVectorDBStorage",
"OracleVectorDBStorage",
"MongoVectorDBStorage",
],
"required_methods": ["query", "upsert"],
@@ -41,7 +38,6 @@ STORAGE_IMPLEMENTATIONS = {
"implementations": [
"JsonDocStatusStorage",
"PGDocStatusStorage",
"PGDocStatusStorage",
"MongoDocStatusStorage",
],
"required_methods": ["get_docs_by_status"],
@@ -56,12 +52,6 @@ STORAGE_ENV_REQUIREMENTS: dict[str, list[str]] = {
"RedisKVStorage": ["REDIS_URI"],
"TiDBKVStorage": ["TIDB_USER", "TIDB_PASSWORD", "TIDB_DATABASE"],
"PGKVStorage": ["POSTGRES_USER", "POSTGRES_PASSWORD", "POSTGRES_DATABASE"],
"OracleKVStorage": [
"ORACLE_DSN",
"ORACLE_USER",
"ORACLE_PASSWORD",
"ORACLE_CONFIG_DIR",
],
# Graph Storage Implementations
"NetworkXStorage": [],
"Neo4JStorage": ["NEO4J_URI", "NEO4J_USERNAME", "NEO4J_PASSWORD"],
@@ -78,12 +68,6 @@ STORAGE_ENV_REQUIREMENTS: dict[str, list[str]] = {
"POSTGRES_PASSWORD",
"POSTGRES_DATABASE",
],
"OracleGraphStorage": [
"ORACLE_DSN",
"ORACLE_USER",
"ORACLE_PASSWORD",
"ORACLE_CONFIG_DIR",
],
# Vector Storage Implementations
"NanoVectorDBStorage": [],
"MilvusVectorDBStorage": [],
@@ -92,12 +76,6 @@ STORAGE_ENV_REQUIREMENTS: dict[str, list[str]] = {
"PGVectorStorage": ["POSTGRES_USER", "POSTGRES_PASSWORD", "POSTGRES_DATABASE"],
"FaissVectorDBStorage": [],
"QdrantVectorDBStorage": ["QDRANT_URL"], # QDRANT_API_KEY has default value None
"OracleVectorDBStorage": [
"ORACLE_DSN",
"ORACLE_USER",
"ORACLE_PASSWORD",
"ORACLE_CONFIG_DIR",
],
"MongoVectorDBStorage": [],
# Document Status Storage Implementations
"JsonDocStatusStorage": [],
@@ -112,9 +90,6 @@ STORAGES = {
"NanoVectorDBStorage": ".kg.nano_vector_db_impl",
"JsonDocStatusStorage": ".kg.json_doc_status_impl",
"Neo4JStorage": ".kg.neo4j_impl",
"OracleKVStorage": ".kg.oracle_impl",
"OracleGraphStorage": ".kg.oracle_impl",
"OracleVectorDBStorage": ".kg.oracle_impl",
"MilvusVectorDBStorage": ".kg.milvus_impl",
"MongoKVStorage": ".kg.mongo_impl",
"MongoDocStatusStorage": ".kg.mongo_impl",

1463
lightrag/kg/oracle_impl.py
View File

File diff suppressed because it is too large Load Diff