ragflow-sdk-myfbee-test 1.0.1.dev0__py3-none-any.whl

ragflow_sdk/__init__.py

@@ -0,0 +1,39 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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.
+#
+
+# Keep runtime validation when beartype is installed, while allowing slim
+# SDK installs to import without the optional checker.
+try:
+    from beartype.claw import beartype_this_package
+
+    beartype_this_package()
+except ImportError:
+    pass
+
+import importlib.metadata
+
+from .ragflow import RAGFlow
+from .modules.dataset import DataSet
+from .modules.chat import Chat
+from .modules.session import Session
+from .modules.document import Document
+from .modules.chunk import Chunk
+from .modules.agent import Agent
+from .modules.memory import Memory
+
+__version__ = importlib.metadata.version("ragflow_sdk")
+
+__all__ = ["RAGFlow", "DataSet", "Chat", "Session", "Document", "Chunk", "Agent", "Memory"]

ragflow_sdk/modules/__init__.py

@@ -0,0 +1,15 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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.
+#

ragflow_sdk/modules/agent.py

@@ -0,0 +1,69 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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 .base import Base
+from .session import Session
+
+
+class Agent(Base):
+    def __init__(self, rag, res_dict):
+        self.id = None
+        self.avatar = None
+        self.canvas_type = None
+        self.description = None
+        self.dsl = None
+        super().__init__(rag, res_dict)
+
+    class Dsl(Base):
+        def __init__(self, rag, res_dict):
+            self.answer = []
+            self.components = {"begin": {"downstream": ["Answer:China"], "obj": {"component_name": "Begin", "params": {}}, "upstream": []}}
+            self.graph = {
+                "edges": [],
+                "nodes": [{"data": {"label": "Begin", "name": "begin"}, "id": "begin", "position": {"x": 50, "y": 200}, "sourcePosition": "left", "targetPosition": "right", "type": "beginNode"}],
+            }
+            self.history = []
+            self.messages = []
+            self.path = []
+            self.reference = []
+            super().__init__(rag, res_dict)
+
+    def create_session(self, **kwargs) -> Session:
+        res = self.post(f"/agents/{self.id}/sessions", json=kwargs)
+        res = res.json()
+        if res.get("code") == 0:
+            return Session(self.rag, res.get("data"))
+        raise Exception(res.get("message"))
+
+    def list_sessions(self, page: int = 1, page_size: int = 30, orderby: str = "create_time", desc: bool = True, id: str = None) -> list[Session]:
+        res = self.get(f"/agents/{self.id}/sessions", {"page": page, "page_size": page_size, "orderby": orderby, "desc": desc, "id": id})
+        res = res.json()
+        if res.get("code") == 0:
+            result_list = []
+            for data in res.get("data"):
+                temp_agent = Session(self.rag, data)
+                result_list.append(temp_agent)
+            return result_list
+        raise Exception(res.get("message"))
+
+    def delete_sessions(self, ids: list[str] | None = None, delete_all: bool = False):
+        payload = {"ids": ids}
+        if delete_all:
+            payload["delete_all"] = True
+        res = self.rm(f"/agents/{self.id}/sessions", payload)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res.get("message"))

ragflow_sdk/modules/base.py

@@ -0,0 +1,62 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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.
+#
+
+
+class Base:
+    def __init__(self, rag, res_dict):
+        self.rag = rag
+        self._update_from_dict(rag, res_dict)
+
+    def _update_from_dict(self, rag, res_dict):
+        for k, v in res_dict.items():
+            if isinstance(v, dict):
+                self.__dict__[k] = Base(rag, v)
+            else:
+                self.__dict__[k] = v
+
+    def to_json(self):
+        pr = {}
+        for name in dir(self):
+            value = getattr(self, name)
+            if not name.startswith("__") and not callable(value) and name != "rag":
+                if isinstance(value, Base):
+                    pr[name] = value.to_json()
+                else:
+                    pr[name] = value
+        return pr
+
+    def post(self, path, json=None, stream=False, files=None):
+        res = self.rag.post(path, json, stream=stream, files=files)
+        return res
+
+    def get(self, path, params=None):
+        res = self.rag.get(path, params)
+        return res
+
+    def rm(self, path, json):
+        res = self.rag.delete(path, json)
+        return res
+
+    def put(self, path, json):
+        res = self.rag.put(path, json)
+        return res
+
+    def patch(self, path, json):
+        res = self.rag.patch(path, json)
+        return res
+
+    def __str__(self):
+        return str(self.to_json())

ragflow_sdk/modules/chat.py

@@ -0,0 +1,67 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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 .base import Base
+from .session import Session
+
+
+class Chat(Base):
+    def __init__(self, rag, res_dict):
+        self.id = ""
+        self.name = "assistant"
+        self.icon = ""
+        self.dataset_ids = []
+        self.llm_id = None
+        self.llm_setting = {}
+        self.prompt_config = {}
+        self.similarity_threshold = 0.2
+        self.vector_similarity_weight = 0.3
+        self.top_n = 6
+        self.top_k = 1024
+        self.rerank_id = ""
+        super().__init__(rag, res_dict)
+
+    def update(self, update_message: dict):
+        if not isinstance(update_message, dict):
+            raise Exception("ValueError('`update_message` must be a dict')")
+        res = self.patch(f"/chats/{self.id}", update_message)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+
+    def create_session(self, name: str = "New session") -> Session:
+        res = self.post(f"/chats/{self.id}/sessions", {"name": name})
+        res = res.json()
+        if res.get("code") == 0:
+            return Session(self.rag, res["data"])
+        raise Exception(res.get("message"))
+
+    def list_sessions(self, page: int = 1, page_size: int = 30, orderby: str = "create_time", desc: bool = True, id: str = None, name: str = None, user_id: str = None) -> list[Session]:
+        res = self.get(f"/chats/{self.id}/sessions", {"page": page, "page_size": page_size, "orderby": orderby, "desc": desc, "id": id, "name": name, "user_id": user_id})
+        res = res.json()
+        if res.get("code") == 0:
+            result_list = []
+            for data in res["data"]:
+                result_list.append(Session(self.rag, data))
+            return result_list
+        raise Exception(res["message"])
+
+    def delete_sessions(self, ids: list[str] | None = None, delete_all: bool = False):
+        res = self.rm(f"/chats/{self.id}/sessions", {"ids": ids, "delete_all": delete_all})
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res.get("message"))

ragflow_sdk/modules/chunk.py

@@ -0,0 +1,62 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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 .base import Base
+
+
+class ChunkUpdateError(Exception):
+    def __init__(self, code=None, message=None, details=None):
+        self.code = code
+        self.message = message
+        self.details = details
+        super().__init__(message)
+
+
+class Chunk(Base):
+    def __init__(self, rag, res_dict):
+        self.id = ""
+        self.content = ""
+        self.image_id = ""
+        self.important_keywords = []
+        self.tag_kwd = []
+        self.questions = []
+        self.create_time = ""
+        self.create_timestamp = 0.0
+        self.dataset_id = None
+        self.document_name = ""
+        self.document_keyword = ""
+        self.document_id = ""
+        self.available = True
+        # Additional fields for retrieval results
+        self.similarity = 0.0
+        self.vector_similarity = 0.0
+        self.term_similarity = 0.0
+        self.positions = []
+        self.doc_type = ""
+        for k in list(res_dict.keys()):
+            if k not in self.__dict__:
+                res_dict.pop(k)
+        super().__init__(rag, res_dict)
+
+        # for backward compatibility
+        if not self.document_name:
+            self.document_name = self.document_keyword
+
+    def update(self, update_message: dict):
+        res = self.patch(f"/datasets/{self.dataset_id}/documents/{self.document_id}/chunks/{self.id}", update_message)
+        res = res.json()
+        if res.get("code") != 0:
+            raise ChunkUpdateError(code=res.get("code"), message=res.get("message"), details=res.get("details"))

ragflow_sdk/modules/dataset.py

@@ -0,0 +1,182 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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 typing import Any
+
+from .base import Base
+from .document import Document
+
+
+class DataSet(Base):
+    class ParserConfig(Base):
+        def __init__(self, rag, res_dict):
+            super().__init__(rag, res_dict)
+
+    def __init__(self, rag, res_dict):
+        self.id = ""
+        self.name = ""
+        self.avatar = ""
+        self.tenant_id = None
+        self.description = ""
+        self.embedding_model = ""
+        self.permission = "me"
+        self.document_count = 0
+        self.chunk_count = 0
+        self.chunk_method = "naive"
+        self.parser_config = None
+        self.pagerank = 0
+        for k in list(res_dict.keys()):
+            if k not in self.__dict__:
+                res_dict.pop(k)
+        super().__init__(rag, res_dict)
+
+    def update(self, update_message: dict):
+        res = self.put(f"/datasets/{self.id}", update_message)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+
+        self._update_from_dict(self.rag, res.get("data", {}))
+        return self
+
+    def upload_documents(self, document_list: list[dict]):
+        url = f"/datasets/{self.id}/documents"
+        files = [("file", (ele["display_name"], ele["blob"])) for ele in document_list]
+        res = self.post(path=url, json=None, files=files)
+        res = res.json()
+        if res.get("code") == 0:
+            doc_list = []
+            for doc in res["data"]:
+                document = Document(self.rag, doc)
+                doc_list.append(document)
+            return doc_list
+        raise Exception(res.get("message"))
+
+    def list_documents(
+        self,
+        id: str | None = None,
+        ids: list[str] | None = None,
+        name: str | None = None,
+        keywords: str | None = None,
+        page: int = 1,
+        page_size: int = 30,
+        orderby: str = "create_time",
+        desc: bool = True,
+        create_time_from: int = 0,
+        create_time_to: int = 0,
+    ):
+        # Validate that id and ids are not used together
+        if id and ids:
+            raise ValueError("Cannot use both 'id' and 'ids' parameters at the same time.")
+
+        params = {
+            "id": id,
+            "name": name,
+            "keywords": keywords,
+            "page": page,
+            "page_size": page_size,
+            "orderby": orderby,
+            "desc": desc,
+            "create_time_from": create_time_from,
+            "create_time_to": create_time_to,
+        }
+        # Handle ids parameter - requests expands list values into multiple query params
+        if ids:
+            params["ids"] = ids
+        res = self.get(f"/datasets/{self.id}/documents", params=params)
+        res = res.json()
+        documents = []
+        if res.get("code") == 0:
+            for document in res["data"].get("docs"):
+                documents.append(Document(self.rag, document))
+            return documents
+        raise Exception(res["message"])
+
+    def delete_documents(self, ids: list[str] | None = None, delete_all: bool = False):
+        res = self.rm(f"/datasets/{self.id}/documents", {"ids": ids, "delete_all": delete_all})
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+
+    def _get_documents_status(self, document_ids):
+        import time
+
+        terminal_states = {"DONE", "FAIL", "CANCEL"}
+        interval_sec = 1
+        pending = set(document_ids)
+        finished = []
+        while pending:
+            for doc_id in list(pending):
+
+                def fetch_doc(doc_id: str) -> Document | None:
+                    try:
+                        docs = self.list_documents(id=doc_id)
+                        return docs[0] if docs else None
+                    except Exception:
+                        return None
+
+                doc = fetch_doc(doc_id)
+                if doc is None:
+                    continue
+                if isinstance(doc.run, str) and doc.run.upper() in terminal_states:
+                    finished.append((doc_id, doc.run, doc.chunk_count, doc.token_count))
+                    pending.discard(doc_id)
+                elif float(doc.progress or 0.0) >= 1.0:
+                    finished.append((doc_id, "DONE", doc.chunk_count, doc.token_count))
+                    pending.discard(doc_id)
+            if pending:
+                time.sleep(interval_sec)
+        return finished
+
+    def async_parse_documents(self, document_ids):
+        res = self.post(f"/datasets/{self.id}/chunks", {"document_ids": document_ids})
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res.get("message"))
+
+    def parse_documents(self, document_ids):
+        try:
+            self.async_parse_documents(document_ids)
+            self._get_documents_status(document_ids)
+        except KeyboardInterrupt:
+            self.async_cancel_parse_documents(document_ids)
+
+        return self._get_documents_status(document_ids)
+
+    def async_cancel_parse_documents(self, document_ids):
+        res = self.rm(f"/datasets/{self.id}/chunks", {"document_ids": document_ids})
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res.get("message"))
+
+    def get_auto_metadata(self) -> dict[str, Any]:
+        """
+        Retrieve auto-metadata configuration for a dataset via SDK.
+        """
+        res = self.get(f"/datasets/{self.id}/metadata/config")
+        res = res.json()
+        if res.get("code") == 0:
+            return res["data"]
+        raise Exception(res["message"])
+
+    def update_auto_metadata(self, **config: Any) -> dict[str, Any]:
+        """
+        Update auto-metadata configuration for a dataset via SDK.
+        """
+        res = self.put(f"/datasets/{self.id}/metadata/config", config)
+        res = res.json()
+        if res.get("code") == 0:
+            return res["data"]
+        raise Exception(res["message"])

ragflow_sdk/modules/document.py

@@ -0,0 +1,109 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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.
+#
+
+import logging
+
+from .base import Base
+from .chunk import Chunk
+
+logger = logging.getLogger(__name__)
+
+
+class Document(Base):
+    class ParserConfig(Base):
+        def __init__(self, rag, res_dict):
+            super().__init__(rag, res_dict)
+
+    def __init__(self, rag, res_dict):
+        self.id = ""
+        self.name = ""
+        self.thumbnail = None
+        self.dataset_id = None
+        self.chunk_method = "naive"
+        self.parser_config = {"pages": [[1, 1000000]]}
+        self.source_type = "local"
+        self.type = ""
+        self.created_by = ""
+        self.size = 0
+        self.token_count = 0
+        self.chunk_count = 0
+        self.progress = 0.0
+        self.progress_msg = ""
+        self.process_begin_at = None
+        self.process_duration = 0.0
+        self.run = "0"
+        self.status = "1"
+        self.meta_fields = {}
+        for k in list(res_dict.keys()):
+            if k not in self.__dict__:
+                res_dict.pop(k)
+        super().__init__(rag, res_dict)
+
+    def update(self, update_message: dict):
+        if "meta_fields" in update_message:
+            if not isinstance(update_message["meta_fields"], dict):
+                raise Exception("meta_fields must be a dictionary")
+        res = self.patch(f"/datasets/{self.dataset_id}/documents/{self.id}", update_message)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+
+        self._update_from_dict(self.rag, res.get("data", {}))
+        return self
+
+    def download(self):
+        res = self.get(f"/datasets/{self.dataset_id}/documents/{self.id}")
+        content_disposition = res.headers.get("Content-Disposition", "")
+        if content_disposition.lstrip().lower().startswith("attachment"):
+            logger.debug("Document.download returning attachment content dataset_id=%s document_id=%s", self.dataset_id, self.id)
+            return res.content
+
+        try:
+            response = res.json()
+            if isinstance(response, dict) and "code" in response and response["code"] != 0:
+                raise Exception(response.get("message"))
+        except ValueError:
+            pass
+        logger.debug("Document.download returning non-attachment content dataset_id=%s document_id=%s", self.dataset_id, self.id)
+        return res.content
+
+    def list_chunks(self, page=1, page_size=30, keywords="", id=""):
+        data = {"keywords": keywords, "page": page, "page_size": page_size, "id": id}
+        res = self.get(f"/datasets/{self.dataset_id}/documents/{self.id}/chunks", data)
+        res = res.json()
+        if res.get("code") == 0:
+            chunks = []
+            for data in res["data"].get("chunks"):
+                chunk = Chunk(self.rag, data)
+                chunks.append(chunk)
+            return chunks
+        raise Exception(res.get("message"))
+
+    def add_chunk(self, content: str, important_keywords: list[str] = [], questions: list[str] = [], image_base64: str | None = None, *, tag_kwd: list[str] = []):
+        body = {"content": content, "important_keywords": important_keywords, "tag_kwd": tag_kwd, "questions": questions}
+        if image_base64 is not None:
+            body["image_base64"] = image_base64
+        res = self.post(f"/datasets/{self.dataset_id}/documents/{self.id}/chunks", body)
+        res = res.json()
+        if res.get("code") == 0:
+            return Chunk(self.rag, res["data"].get("chunk"))
+        raise Exception(res.get("message"))
+
+    def delete_chunks(self, ids: list[str] | None = None, delete_all: bool = False):
+        res = self.rm(f"/datasets/{self.dataset_id}/documents/{self.id}/chunks", {"chunk_ids": ids, "delete_all": delete_all})
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res.get("message"))

ragflow_sdk/modules/memory.py

@@ -0,0 +1,87 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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 .base import Base
+
+
+class Memory(Base):
+    def __init__(self, rag, res_dict):
+        self.id = ""
+        self.name = ""
+        self.avatar = None
+        self.tenant_id = None
+        self.owner_name = ""
+        self.memory_type = ["raw"]
+        self.storage_type = "table"
+        self.embd_id = ""
+        self.llm_id = ""
+        self.permissions = "me"
+        self.description = ""
+        self.memory_size = 5 * 1024 * 1024
+        self.forgetting_policy = "FIFO"
+        self.temperature = 0.5
+        self.system_prompt = ""
+        self.user_prompt = ""
+        for k in list(res_dict.keys()):
+            if k not in self.__dict__:
+                res_dict.pop(k)
+        super().__init__(rag, res_dict)
+
+    def update(self, update_dict: dict):
+        res = self.put(f"/memories/{self.id}", update_dict)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        self._update_from_dict(self.rag, res.get("data", {}))
+        return self
+
+    def get_config(self):
+        res = self.get(f"/memories/{self.id}/config")
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        self._update_from_dict(self.rag, res.get("data", {}))
+        return self
+
+    def list_memory_messages(self, agent_id: str | list[str] = None, keywords: str = None, page: int = 1, page_size: int = 50):
+        params = {"agent_id": agent_id, "keywords": keywords, "page": page, "page_size": page_size}
+        res = self.get(f"/memories/{self.id}", params)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        return res["data"]
+
+    def forget_message(self, message_id: int):
+        res = self.rm(f"/messages/{self.id}:{message_id}", {})
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        return True
+
+    def update_message_status(self, message_id: int, status: bool):
+        update_message = {"status": status}
+        res = self.put(f"/messages/{self.id}:{message_id}", update_message)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        return True
+
+    def get_message_content(self, message_id: int) -> dict:
+        res = self.get(f"/messages/{self.id}:{message_id}/content")
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        return res["data"]

ragflow_sdk/modules/session.py

@@ -0,0 +1,203 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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.
+#
+
+import json
+import logging
+
+from .base import Base
+
+logger = logging.getLogger(__name__)
+
+
+class Session(Base):
+    def __init__(self, rag, res_dict):
+        self.id = None
+        self.name = "New session"
+        self.messages = [{"role": "assistant", "content": "Hi! I am your assistant, can I help you?"}]
+        for key, value in res_dict.items():
+            if key == "chat_id" and value is not None:
+                self.chat_id = None
+                self.__session_type = "chat"
+            if key == "agent_id" and value is not None:
+                self.agent_id = None
+                self.__session_type = "agent"
+        super().__init__(rag, res_dict)
+
+    def ask(
+        self,
+        question="",
+        stream=False,
+        inputs=None,
+        release=None,
+        return_trace=None,
+        **kwargs,
+    ):
+        """
+        Ask a question to the session.
+
+        Parameters
+        ----------
+        question : str
+            The user's question. May be empty when the agent is driven solely by
+            Begin component inputs.
+        stream : bool
+            If ``True``, yields ``Message`` objects as they arrive (SSE streaming).
+            If ``False``, yields a single ``Message`` with the final answer.
+        inputs : dict, optional
+            Values for variables declared on the agent's **Begin** component. Each
+            value must be a dict containing at least a ``"value"`` key, and may
+            include ``"type"``. Example::
+
+                session.ask(
+                    "",
+                    stream=False,
+                    inputs={"key1": {"type": "line", "value": "hello"}},
+                )
+
+            Only meaningful for agent sessions; ignored for chat sessions.
+        release : bool, optional
+            If ``True``, run against the latest published agent version instead of
+            the editable draft. Only meaningful for agent sessions.
+        return_trace : bool, optional
+            If ``True``, include execution trace information in the response.
+            Only meaningful for agent sessions.
+        **kwargs
+            Additional fields forwarded verbatim to the completion endpoint
+            (e.g. ``session_id``, ``files``, ``user_id``, ``custom_header``).
+            See the HTTP API reference for the full list.
+        """
+        if inputs is not None:
+            kwargs["inputs"] = inputs
+        if release is not None:
+            kwargs["release"] = release
+        if return_trace is not None:
+            kwargs["return_trace"] = return_trace
+
+        if inputs is not None or release is not None or return_trace is not None:
+            logger.debug(
+                "Session.ask explicit-params session_type=%s session_id=%s input_keys=%s release=%s return_trace=%s",
+                self.__session_type,
+                getattr(self, "id", None),
+                list(inputs.keys()) if isinstance(inputs, dict) else None,
+                release,
+                return_trace,
+            )
+
+        if self.__session_type == "agent":
+            res = self._ask_agent(question, stream, **kwargs)
+        elif self.__session_type == "chat":
+            res = self._ask_chat(question, stream, **kwargs)
+        else:
+            raise Exception(f"Unknown session type: {self.__session_type}")
+
+        if stream:
+            for line in res.iter_lines(decode_unicode=True):
+                if not line:
+                    continue  # Skip empty lines
+                line = line.strip()
+                if line.startswith("data:"):
+                    content = line[len("data:") :].strip()
+                    if content == "[DONE]":
+                        break  # End of stream
+                else:
+                    content = line
+
+                try:
+                    json_data = json.loads(content)
+                except json.JSONDecodeError:
+                    continue  # Skip lines that are not valid JSON
+
+                event = json_data.get("event", None)
+                if event:
+                    if self.__session_type == "agent" and event not in {"message", "message_end"}:
+                        continue
+                    if self.__session_type == "chat" and event != "message":
+                        continue
+
+                if self.__session_type == "chat" and json_data.get("data") is True:
+                    return
+                if self.__session_type == "agent":
+                    message = self._structure_answer(json_data)
+                    if event == "message_end":
+                        message.content = ""
+                        if message.reference:
+                            yield message
+                        continue
+                    yield message
+                else:
+                    yield self._structure_answer(json_data["data"])
+        else:
+            try:
+                json_data = res.json()
+            except ValueError:
+                raise Exception(f"Invalid response {res}")
+            yield self._structure_answer(json_data["data"])
+
+    def _structure_answer(self, json_data):
+        answer = ""
+        event = None
+        if self.__session_type == "agent":
+            event = json_data.get("event")
+            json_data = json_data["data"]
+            answer = json_data.get("content", "")
+        elif self.__session_type == "chat":
+            answer = json_data["answer"]
+        reference = json_data.get("reference", {})
+        temp_dict = {"content": answer, "role": "assistant"}
+        if reference and "chunks" in reference:
+            chunks = reference["chunks"]
+            if isinstance(chunks, dict):
+                chunks = list(chunks.values())
+            temp_dict["reference"] = chunks
+            if self.__session_type == "agent":
+                reference_count = len(chunks) if isinstance(chunks, list) else 0
+                logger.debug("Session.ask parsed agent references session_id=%s event=%s reference_count=%s", self.id, event, reference_count)
+        message = Message(self.rag, temp_dict)
+        return message
+
+    def _ask_chat(self, question: str, stream: bool, **kwargs):
+        json_data = {"question": question, "stream": stream, "session_id": self.id}
+        json_data.update(kwargs)
+        res = self.post(f"/chats/{self.chat_id}/completions", json_data, stream=stream)
+        return res
+
+    def _ask_agent(self, question: str, stream: bool, **kwargs):
+        json_data = {
+            "agent_id": self.agent_id,
+            "query": question,
+            "stream": stream,
+            "session_id": self.id,
+            "openai-compatible": False,
+        }
+        json_data.update(kwargs)
+        res = self.post("/agents/chat/completions", json_data, stream=stream)
+        return res
+
+    def update(self, update_message):
+        res = self.patch(f"/chats/{self.chat_id}/sessions/{self.id}", update_message)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res.get("message"))
+
+
+class Message(Base):
+    def __init__(self, rag, res_dict):
+        self.content = "Hi! I am your assistant, can I help you?"
+        self.reference = None
+        self.role = "assistant"
+        self.prompt = None
+        self.id = None
+        super().__init__(rag, res_dict)

ragflow_sdk/ragflow.py

@@ -0,0 +1,398 @@
+#
+#  Copyright 2024 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed 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 typing import Optional, Any
+
+import requests
+
+from .modules.agent import Agent
+from .modules.chat import Chat
+from .modules.chunk import Chunk
+from .modules.dataset import DataSet
+from .modules.memory import Memory
+
+
+class RAGFlow:
+    def __init__(self, api_key, base_url, version="v1"):
+        """
+        api_url: http://<host_address>/api/v1
+        """
+        self.user_key = api_key
+        self.api_url = f"{base_url}/api/{version}"
+        self.authorization_header = {"Authorization": "{} {}".format("Bearer", self.user_key)}
+
+    def post(self, path, json=None, stream=False, files=None):
+        res = requests.post(url=self.api_url + path, json=json, headers=self.authorization_header, stream=stream, files=files)
+        return res
+
+    def get(self, path, params=None, json=None):
+        res = requests.get(url=self.api_url + path, params=params, headers=self.authorization_header, json=json)
+        return res
+
+    def delete(self, path, json):
+        res = requests.delete(url=self.api_url + path, json=json, headers=self.authorization_header)
+        return res
+
+    def put(self, path, json):
+        res = requests.put(url=self.api_url + path, json=json, headers=self.authorization_header)
+        return res
+
+    def patch(self, path, json):
+        res = requests.patch(url=self.api_url + path, json=json, headers=self.authorization_header)
+        return res
+
+    def create_dataset(
+        self,
+        name: str,
+        avatar: Optional[str] = None,
+        description: Optional[str] = None,
+        embedding_model: Optional[str] = None,
+        permission: str = "me",
+        chunk_method: str = "naive",
+        parser_config: Optional[DataSet.ParserConfig] = None,
+        auto_metadata_config: Optional[dict[str, Any]] = None,
+    ) -> DataSet:
+        payload = {
+            "name": name,
+            "avatar": avatar,
+            "description": description,
+            "embedding_model": embedding_model,
+            "permission": permission,
+            "chunk_method": chunk_method,
+        }
+        if parser_config is not None:
+            payload["parser_config"] = parser_config.to_json()
+        if auto_metadata_config is not None:
+            payload["auto_metadata_config"] = auto_metadata_config
+
+        res = self.post("/datasets", payload)
+        res = res.json()
+        if res.get("code") == 0:
+            return DataSet(self, res["data"])
+        raise Exception(res["message"])
+
+    def delete_datasets(self, ids: list[str] | None = None, delete_all: bool = False):
+        res = self.delete("/datasets", {"ids": ids, "delete_all": delete_all})
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+
+    def get_dataset(self, name: str):
+        _list = self.list_datasets(name=name)
+        if len(_list) > 0:
+            return _list[0]
+        raise Exception("Dataset %s not found" % name)
+
+    def list_datasets(
+        self, page: int = 1, page_size: int = 30, orderby: str = "create_time", desc: bool = True, id: str | None = None, ids: list[str] | None = None, name: str | None = None
+    ) -> list[DataSet]:
+        if id and ids:
+            raise ValueError("Cannot use both 'id' and 'ids' parameters at the same time.")
+
+        res = self.get(
+            "/datasets",
+            {
+                "page": page,
+                "page_size": page_size,
+                "orderby": orderby,
+                "desc": desc,
+                "id": id,
+                "ids": ids,
+                "name": name,
+            },
+        )
+        res = res.json()
+        result_list = []
+        if res.get("code") == 0:
+            for data in res["data"]:
+                result_list.append(DataSet(self, data))
+            return result_list
+        raise Exception(res["message"])
+
+    def create_chat(
+        self,
+        name: str,
+        icon: str = "",
+        dataset_ids: list[str] | None = None,
+        llm_id: str | None = None,
+        llm_setting: dict | None = None,
+        prompt_config: dict | None = None,
+        **kwargs,
+    ) -> Chat:
+        payload = {"name": name, "icon": icon, "dataset_ids": dataset_ids or []}
+        if llm_id is not None:
+            payload["llm_id"] = llm_id
+        if llm_setting is not None:
+            payload["llm_setting"] = llm_setting
+        if prompt_config is not None:
+            payload["prompt_config"] = prompt_config
+        payload.update(kwargs)
+        res = self.post("/chats", payload)
+        res = res.json()
+        if res.get("code") == 0:
+            return Chat(self, res["data"])
+        raise Exception(res["message"])
+
+    def delete_chats(self, ids: list[str] | None = None, delete_all: bool = False):
+        res = self.delete("/chats", {"ids": ids, "delete_all": delete_all})
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+
+    def get_chat(self, chat_id: str) -> Chat:
+        res = self.get(f"/chats/{chat_id}")
+        res = res.json()
+        if res.get("code") == 0:
+            return Chat(self, res["data"])
+        raise Exception(res["message"])
+
+    def list_chats(
+        self,
+        page: int = 1,
+        page_size: int = 30,
+        orderby: str = "create_time",
+        desc: bool = True,
+        id: str | None = None,
+        name: str | None = None,
+        keywords: str | None = None,
+        owner_ids: str | list[str] | None = None,
+    ) -> list[Chat]:
+        res = self.get(
+            "/chats",
+            {
+                "page": page,
+                "page_size": page_size,
+                "orderby": orderby,
+                "desc": desc,
+                "id": id,
+                "name": name,
+                "keywords": keywords,
+                "owner_ids": owner_ids,
+            },
+        )
+        res = res.json()
+        result_list = []
+        if res.get("code") == 0:
+            for data in res["data"]["chats"]:
+                result_list.append(Chat(self, data))
+            return result_list
+        raise Exception(res["message"])
+
+    def retrieve(
+        self,
+        dataset_ids,
+        document_ids=None,
+        question="",
+        page=1,
+        page_size=30,
+        similarity_threshold=0.2,
+        vector_similarity_weight=0.3,
+        top_k=1024,
+        rerank_id: str | None = None,
+        keyword: bool = False,
+        cross_languages: list[str] | None = None,
+        metadata_condition: dict | None = None,
+        use_kg: bool = False,
+        toc_enhance: bool = False,
+    ):
+        if document_ids is None:
+            document_ids = []
+        data_json = {
+            "page": page,
+            "page_size": page_size,
+            "similarity_threshold": similarity_threshold,
+            "vector_similarity_weight": vector_similarity_weight,
+            "top_k": top_k,
+            "rerank_id": rerank_id,
+            "keyword": keyword,
+            "question": question,
+            "dataset_ids": dataset_ids,
+            "document_ids": document_ids,
+            "cross_languages": cross_languages,
+            "metadata_condition": metadata_condition,
+            "use_kg": use_kg,
+            "toc_enhance": toc_enhance,
+        }
+        # Send a POST request to the backend service (using requests library as an example, actual implementation may vary)
+        res = self.post("/retrieval", json=data_json)
+        res = res.json()
+        if res.get("code") == 0:
+            chunks = []
+            for chunk_data in res["data"].get("chunks"):
+                chunk = Chunk(self, chunk_data)
+                chunks.append(chunk)
+            return chunks
+        raise Exception(res.get("message"))
+
+    def list_agents(self, page: int = 1, page_size: int = 30, orderby: str = "update_time", desc: bool = True) -> list[Agent]:
+        res = self.get(
+            "/agents",
+            {
+                "page": page,
+                "page_size": page_size,
+                "orderby": orderby,
+                "desc": desc,
+            },
+        )
+        res = res.json()
+        result_list = []
+        if res.get("code") == 0:
+            data = res.get("data") or {}
+            data_list = data.get("canvas", [])
+            for data in data_list:
+                result_list.append(Agent(self, data))
+            return result_list
+        raise Exception(res["message"])
+
+    def get_agent(self, agent_id: str) -> Agent:
+        res = self.get(f"/agents/{agent_id}")
+        res = res.json()
+        if res.get("code") == 0:
+            return Agent(self, res["data"])
+        raise Exception(res["message"])
+
+    def create_agent(
+        self,
+        title: str,
+        dsl: dict,
+        description: str | None = None,
+        canvas_type: str | None = None,
+    ) -> None:
+        req = {"title": title, "dsl": dsl}
+
+        if description is not None:
+            req["description"] = description
+
+        if canvas_type is not None:
+            req["canvas_type"] = canvas_type
+
+        res = self.post("/agents", req)
+        res = res.json()
+
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+
+    def update_agent(
+        self,
+        agent_id: str,
+        title: str | None = None,
+        description: str | None = None,
+        dsl: dict | None = None,
+        canvas_type: str | None = None,
+    ) -> None:
+        req = {}
+
+        if title is not None:
+            req["title"] = title
+
+        if description is not None:
+            req["description"] = description
+
+        if dsl is not None:
+            req["dsl"] = dsl
+
+        if canvas_type is not None:
+            req["canvas_type"] = canvas_type
+
+        res = self.put(f"/agents/{agent_id}", req)
+        res = res.json()
+
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+
+    def delete_agent(self, agent_id: str) -> None:
+        res = self.delete(f"/agents/{agent_id}", {})
+        res = res.json()
+
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+
+    def create_memory(self, name: str, memory_type: list[str], embd_id: str, llm_id: str):
+        payload = {"name": name, "memory_type": memory_type, "embd_id": embd_id, "llm_id": llm_id}
+        res = self.post("/memories", payload)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        return Memory(self, res["data"])
+
+    def list_memory(self, page: int = 1, page_size: int = 50, tenant_id: str | list[str] = None, memory_type: str | list[str] = None, storage_type: str = None, keywords: str = None) -> dict:
+        res = self.get(
+            "/memories",
+            {
+                "page": page,
+                "page_size": page_size,
+                "tenant_id": tenant_id,
+                "memory_type": memory_type,
+                "storage_type": storage_type,
+                "keywords": keywords,
+            },
+        )
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        result_list = []
+        for data in res["data"]["memory_list"]:
+            result_list.append(Memory(self, data))
+        return {"code": res.get("code", 0), "message": res.get("message"), "memory_list": result_list, "total_count": res["data"]["total_count"]}
+
+    def delete_memory(self, memory_id: str):
+        res = self.delete(f"/memories/{memory_id}", {})
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+
+    def add_message(self, memory_id: list[str], agent_id: str, session_id: str, user_input: str, agent_response: str, user_id: str = "") -> str:
+        """Append messages to memories; ``user_id`` is forwarded only for API-key auth (external subject)."""
+        payload = {"memory_id": memory_id, "agent_id": agent_id, "session_id": session_id, "user_input": user_input, "agent_response": agent_response, "user_id": user_id}
+        res = self.post("/messages", payload)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        return res["message"]
+
+    def search_message(
+        self,
+        query: str,
+        memory_id: list[str],
+        agent_id: str = None,
+        session_id: str = None,
+        user_id: str = None,
+        similarity_threshold: float = 0.2,
+        keywords_similarity_weight: float = 0.7,
+        top_n: int = 10,
+    ) -> list[dict]:
+        params = {
+            "query": query,
+            "memory_id": memory_id,
+            "agent_id": agent_id,
+            "session_id": session_id,
+            "user_id": user_id,
+            "similarity_threshold": similarity_threshold,
+            "keywords_similarity_weight": keywords_similarity_weight,
+            "top_n": top_n,
+        }
+        res = self.get("/messages/search", params)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        return res["data"]
+
+    def get_recent_messages(self, memory_id: list[str], agent_id: str = None, session_id: str = None, limit: int = 10) -> list[dict]:
+        params = {"memory_id": memory_id, "agent_id": agent_id, "session_id": session_id, "limit": limit}
+        res = self.get("/messages", params)
+        res = res.json()
+        if res.get("code") != 0:
+            raise Exception(res["message"])
+        return res["data"]

ragflow_sdk_myfbee_test-1.0.1.dev0.dist-info/METADATA

@@ -0,0 +1,10 @@
+Metadata-Version: 2.4
+Name: ragflow-sdk-myfbee-test
+Version: 1.0.1.dev0
+Summary: Python client sdk of [RAGFlow](https://github.com/infiniflow/ragflow). RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding.
+Author-email: Zhichang Yu <yuzhichang@gmail.com>
+License: Apache License, Version 2.0
+Requires-Python: <3.14,>=3.13
+Description-Content-Type: text/markdown
+Requires-Dist: requests<3.0.0,>=2.30.0
+Requires-Dist: beartype<1.0.0,>=0.20.0

ragflow_sdk_myfbee_test-1.0.1.dev0.dist-info/RECORD

@@ -0,0 +1,15 @@
+ragflow_sdk/__init__.py,sha256=ql_QJdp4vDpQ82flx5mdaQVX-Jn1ExmN_lYfPKGuaS4,1319
+ragflow_sdk/ragflow.py,sha256=lVbpO1JHr8KNB9aDpkdfq967tbuXkIG-N7x4dPJo0oQ,13954
+ragflow_sdk/modules/__init__.py,sha256=UQpEQcTBGyFrQsFf8jsVbreWSQA39-TPtgjjtoOpyQY,622
+ragflow_sdk/modules/agent.py,sha256=ocygkPNRm44GKVozpSCe6eOvkXhMao6axpFMmZPd92E,2793
+ragflow_sdk/modules/base.py,sha256=uBGd7yUFqfobmdhsWmP-nqNFmYwvcW_juT1Gu3-Qw-4,1918
+ragflow_sdk/modules/chat.py,sha256=v7hljH8DVBNmhJjKw7F9y-lhtu-CzZkTGuCiMpXaeak,2656
+ragflow_sdk/modules/chunk.py,sha256=HAj7uRkaVTOow9Ozc_BlbLH-2p1zel-ucBsZs1T2VXc,2137
+ragflow_sdk/modules/dataset.py,sha256=3JN_HWEu7h-IK9lUdSLkIWEw9ZyHgbUo86BugxlroUI,6503
+ragflow_sdk/modules/document.py,sha256=tVk1bCuoOEOAPKHD_ULx0smZ04A52pmY2OGSF_6WAkU,4412
+ragflow_sdk/modules/memory.py,sha256=6Gp140KJPd8mB0hdkIf0rCc_RjRqoA7S6d4glYu4pcE,3114
+ragflow_sdk/modules/session.py,sha256=pRpIFuIX6TuF-EZcaYLzZLFDA1QmVkIClavXhER_EUY,7898
+ragflow_sdk_myfbee_test-1.0.1.dev0.dist-info/METADATA,sha256=l0i24q3_1VVOzqOMFVLvEAEAqz3951lOnOcKSy6o4Hk,494
+ragflow_sdk_myfbee_test-1.0.1.dev0.dist-info/WHEEL,sha256=YVMoNqKzERt-wjUZwJ33xBGAwnFl-4cqbYkTtWa4itE,91
+ragflow_sdk_myfbee_test-1.0.1.dev0.dist-info/top_level.txt,sha256=ep9RG2ggpS-qOwUXbtEbjfVyi6mVgQ407otRgO-xNdA,12
+ragflow_sdk_myfbee_test-1.0.1.dev0.dist-info/RECORD,,

ragflow_sdk_myfbee_test-1.0.1.dev0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (84.0.0)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

ragflow_sdk_myfbee_test-1.0.1.dev0.dist-info/top_level.txt

@@ -0,0 +1 @@
+ragflow_sdk