From c55285bacedbef6c659bc372eefa9ba65d5f9d29 Mon Sep 17 00:00:00 2001 From: "liuchangsheng@wisdomidata.com" Date: Tue, 10 Jun 2025 10:47:22 +0800 Subject: [PATCH] =?UTF-8?q?=E5=88=9D=E5=A7=8B=E5=8C=96=E5=90=88=E5=B9=B6?= =?UTF-8?q?=E5=90=8E=E7=AB=AF=E4=BB=A3=E7=A0=81-=20=E4=BF=AE=E5=A4=8D?= =?UTF-8?q?=E5=90=88=E5=B9=B6BUG?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- api/core/rag/datasource/retrieval_service.py | 8 ++++---- api/core/workflow/nodes/vanna/vanna_node.py | 4 ++-- api/pyproject.toml | 7 +++++-- 3 files changed, 11 insertions(+), 8 deletions(-) diff --git a/api/core/rag/datasource/retrieval_service.py b/api/core/rag/datasource/retrieval_service.py index 93864abbce..fc499713f7 100644 --- a/api/core/rag/datasource/retrieval_service.py +++ b/api/core/rag/datasource/retrieval_service.py @@ -405,6 +405,9 @@ class RetrievalService: record["child_chunks"] = segment_child_map[record["segment"].id].get("child_chunks") # type: ignore record["score"] = segment_child_map[record["segment"].id]["max_score"] + # 高分片段，自动拼接相近的片段 + cls.append_next_segments(records=records,dataset_documents=dataset_documents) + result = [] for record in records: # Extract segment @@ -422,8 +425,6 @@ class RetrievalService: if score_value is not None and isinstance(score_value, int | float | str) else None ) - cls.append_next_segments(records=records,dataset_documents=dataset_documents) - # Create RetrievalSegments object retrieval_segment = RetrievalSegments(segment=segment, child_chunks=child_chunks, score=score) result.append(retrieval_segment) @@ -435,7 +436,6 @@ class RetrievalService: @classmethod def append_next_segments(cls, records: list[dict], dataset_documents : dict): - # import pdb; pdb.set_trace() def filter_record(record): document_id = record["segment"].document_id if document_id in dataset_documents: @@ -521,4 +521,4 @@ class RetrievalService: if document_segment.position == this_positions + 1: next_segment = document_segment break - return next_segment \ No newline at end of file + return next_segment diff --git a/api/core/workflow/nodes/vanna/vanna_node.py b/api/core/workflow/nodes/vanna/vanna_node.py index 56f6e9a320..64d2bdc2aa 100644 --- a/api/core/workflow/nodes/vanna/vanna_node.py +++ b/api/core/workflow/nodes/vanna/vanna_node.py @@ -8,7 +8,7 @@ from core.workflow.entities.node_entities import NodeRunResult from core.workflow.nodes.enums import NodeType from core.workflow.nodes.llm import LLMNode from extensions.utils.vanna_text2sql import VannaServer -from models.workflow import WorkflowNodeExecutionStatus +from core.workflow.entities.workflow_node_execution import WorkflowNodeExecutionStatus from .entities import VannaNodeData @@ -89,4 +89,4 @@ class VannaNode(LLMNode): status=WorkflowNodeExecutionStatus.SUCCEEDED, outputs={"output": sql} ) - + diff --git a/api/pyproject.toml b/api/pyproject.toml index f497274da1..9d41ea502f 100644 --- a/api/pyproject.toml +++ b/api/pyproject.toml @@ -2,8 +2,12 @@ name = "dify-api" dynamic = ["version"] requires-python = ">=3.11,<3.13" - dependencies = [ + "vanna[milvus,mysql,ollama,postgres]>=0.7.9,<0.8.0", + "pymilvus[model]>=2.5.8,<2.6.0", + "pillow>=11.2.1,<12.0.0", + "pymupdf>=1.25.5,<1.26.0", + "kaleido==0.2.1", "authlib==1.3.1", "azure-identity==1.16.1", "beautifulsoup4==4.12.2", @@ -190,7 +194,6 @@ vdb = [ "oracledb==3.0.0", "pgvecto-rs[sqlalchemy]~=0.2.1", "pgvector==0.2.5", - "pymilvus~=2.5.0", "pymochow==1.3.1", "pyobvector~=0.1.6", "qdrant-client==1.9.0",