feat(llm): support basic rag api(v1) (#64)

* refact: support test wenxin/ollama conn

---------

Co-authored-by: Hongjun Li <[email protected]>
Co-authored-by: imbajin <[email protected]>

.gitignore

@@ -164,3 +164,4 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 .idea/
+*.DS_Store

hugegraph-llm/src/hugegraph_llm/api/exceptions/__init__.py

@@ -0,0 +1,16 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.

hugegraph-llm/src/hugegraph_llm/api/exceptions/rag_exceptions.py

@@ -0,0 +1,37 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+
+from fastapi import HTTPException
+from hugegraph_llm.api.models.rag_response import RAGResponse
+
+
+class ExternalException(HTTPException):
+    def __init__(self):
+        super().__init__(status_code=400, detail="Connect failed with error code -1, please check the input.")
+
+
+class ConnectionFailedException(HTTPException):
+    def __init__(self, status_code: int, message: str):
+        super().__init__(status_code=status_code, detail=message)
+
+
+def generate_response(response: RAGResponse) -> dict:
+    if response.status_code == -1:
+        raise ExternalException()
+    elif not (200 <= response.status_code < 300):
+        raise ConnectionFailedException(response.status_code, response.message)
+    return {"message": "Connection successful. Configured finished."}

hugegraph-llm/src/hugegraph_llm/api/models/__init__.py

@@ -0,0 +1,16 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.

hugegraph-llm/src/hugegraph_llm/api/models/rag_requests.py

@@ -0,0 +1,52 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+
+from pydantic import BaseModel
+from typing import Optional
+
+
+class RAGRequest(BaseModel):
+    query: str
+    raw_llm: Optional[bool] = True
+    vector_only: Optional[bool] = False
+    graph_only: Optional[bool] = False
+    graph_vector: Optional[bool] = False
+
+
+class GraphConfigRequest(BaseModel):
+    ip: str = "127.0.0.1"
+    port: str = "8080"
+    name: str = "hugegraph"
+    user: str = "xxx"
+    pwd: str = "xxx"
+    gs: str = None
+
+
+class LLMConfigRequest(BaseModel):
+    llm_type: str
+    # The common parameters shared by OpenAI, Qianfan Wenxin,
+    # and OLLAMA platforms.
+    api_key: str
+    api_base: str
+    language_model: str
+    # Openai-only properties
+    max_tokens: str = None
+    # qianfan-wenxin-only properties
+    secret_key: str = None
+    # ollama-only properties
+    host: str = None
+    port: str = None

hugegraph-llm/src/hugegraph_llm/api/models/rag_response.py

@@ -0,0 +1,23 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+
+from pydantic import BaseModel
+
+
+class RAGResponse(BaseModel):
+    status_code: int = -1
+    message: str = ""

hugegraph-llm/src/hugegraph_llm/api/rag_api.py

@@ -14,3 +14,53 @@
 # KIND, either express or implied.  See the License for the
 # specific language governing permissions and limitations
 # under the License.
+
+from fastapi import FastAPI, status
+
+from hugegraph_llm.api.models.rag_response import RAGResponse
+from hugegraph_llm.config import settings
+from hugegraph_llm.api.models.rag_requests import RAGRequest, GraphConfigRequest, LLMConfigRequest
+from hugegraph_llm.api.exceptions.rag_exceptions import generate_response
+
+
+def rag_http_api(app: FastAPI, rag_answer_func, apply_graph_conf, apply_llm_conf, apply_embedding_conf):
+    @app.post("/rag", status_code=status.HTTP_200_OK)
+    def rag_answer_api(req: RAGRequest):
+        result = rag_answer_func(req.query, req.raw_llm, req.vector_only, req.graph_only, req.graph_vector)
+        return {
+            key: value
+            for key, value in zip(["raw_llm", "vector_only", "graph_only", "graph_vector"], result)
+            if getattr(req, key)
+        }
+
+    @app.post("/config/graph", status_code=status.HTTP_201_CREATED)
+    def graph_config_api(req: GraphConfigRequest):
+        # Accept status code
+        res = apply_graph_conf(req.ip, req.port, req.name, req.user, req.pwd, req.gs, origin_call="http")
+        return generate_response(RAGResponse(status_code=res, message="Missing Value"))
+
+    @app.post("/config/llm", status_code=status.HTTP_201_CREATED)
+    def llm_config_api(req: LLMConfigRequest):
+        settings.llm_type = req.llm_type
+
+        if req.llm_type == "openai":
+            res = apply_llm_conf(
+                req.api_key, req.api_base, req.language_model, req.max_tokens, origin_call="http"
+            )
+        elif req.llm_type == "qianfan_wenxin":
+            res = apply_llm_conf(req.api_key, req.secret_key, req.language_model, None, origin_call="http")
+        else:
+            res = apply_llm_conf(req.host, req.port, req.language_model, None, origin_call="http")
+        return generate_response(RAGResponse(status_code=res, message="Missing Value"))
+
+    @app.post("/config/embedding", status_code=status.HTTP_201_CREATED)
+    def embedding_config_api(req: LLMConfigRequest):
+        settings.embedding_type = req.llm_type
+
+        if req.llm_type == "openai":
+            res = apply_embedding_conf(req.api_key, req.api_base, req.language_model, origin_call="http")
+        elif req.llm_type == "qianfan_wenxin":
+            res = apply_embedding_conf(req.api_key, req.api_base, None, origin_call="http")
+        else:
+            res = apply_embedding_conf(req.host, req.port, req.language_model, origin_call="http")
+        return generate_response(RAGResponse(status_code=res, message="Missing Value"))

hugegraph-llm/src/hugegraph_llm/config/config.py

@@ -67,7 +67,7 @@ class Config:
     """HugeGraph settings"""
     graph_ip: Optional[str] = "127.0.0.1"
     graph_port: Optional[int] = 8080
-    # graph_space: Optional[str] = "DEFAULT"
+    graph_space: Optional[str] = None
     graph_name: Optional[str] = "hugegraph"
     graph_user: Optional[str] = "admin"
     graph_pwd: Optional[str] = "xxx"