feat: data source web node

apps/application/workflow/nodes/data_source_web_node/__init__.py

@@ -0,0 +1,10 @@
+# coding=utf-8
+"""
+@project: MaxKB
+@Author:  虎虎虎
+@file:    __init__.py
+@date:    2026/9/16
+@desc:    Web 站点数据源节点（知识库工作流起始节点之一）
+"""
+
+from .data_source_web_node import DataSourceWebNode

apps/application/workflow/nodes/data_source_web_node/data_source_web_node.py

@@ -0,0 +1,107 @@
+# coding=utf-8
+"""
+@project: MaxKB
+@Author:  虎虎虎
+@file:    data_source_web_node.py
+@date:    2026/9/16
+@desc:    Web 站点数据源节点：知识库工作流的起始节点之一，按根地址抓取站点内容写入 document_list 供下游读取
+"""
+
+import traceback
+
+from django.utils.translation import gettext_lazy as _
+from rest_framework import serializers
+
+from application.workflow.common import WorkflowType
+from application.workflow.i_node import CancelledException, INode
+from common.utils.fork import ChildLink, Fork, ForkManage
+from common.utils.logger import maxkb_logger
+
+
+class DataSourceWebNodeParamsSerializer(serializers.Serializer):
+    source_url = serializers.CharField(required=True, label=_("Web source url"))
+    selector = serializers.CharField(
+        required=False, allow_blank=True, allow_null=True, label=_("Web knowledge selector")
+    )
+
+
+class DataSourceWebNode(INode):
+    serializer_class = DataSourceWebNodeParamsSerializer
+    supported_workflow_type_list = [WorkflowType.KNOWLEDGE]
+    type = "data-source-web-node"
+
+    @staticmethod
+    def get_form_list(node):
+        return [
+            {
+                "field": "source_url",
+                "input_type": "TextInput",
+                "attrs": {"placeholder": _("Please enter the Web root address")},
+                "label": _("Web source url"),
+                "required": True,
+            },
+            {
+                "field": "selector",
+                "input_type": "TextInput",
+                "attrs": {"placeholder": _("The default is body, you can enter .classname/#idname/tagname")},
+                "label": _("Web knowledge selector"),
+                "required": False,
+            },
+        ]
+
+    def _get_collect_handler(self, document_list):
+        def handler(child_link: ChildLink, response: Fork.Response):
+            if response.status == 200:
+                try:
+                    document_name = (
+                        child_link.tag.text
+                        if child_link.tag is not None and len(child_link.tag.text.strip()) > 0
+                        else child_link.url
+                    )
+                    document_list.append({"name": document_name.strip(), "content": response.content})
+                except Exception as e:
+                    maxkb_logger.error(f"{str(e)}:{traceback.format_exc()}")
+            # 已取消则抛出 CancelledException,由引擎结束流程
+            self._check_cancelled()
+
+        return handler
+
+    def execute(self):
+        workflow_params = self.get_workflow_parameters()
+        data_source = workflow_params.get("data_source") or {}
+
+        serializer = self.serializer_class(data=data_source)
+        serializer.is_valid(raise_exception=True)
+        source_url = serializer.validated_data.get("source_url")
+        selector = serializer.validated_data.get("selector") or "body"
+
+        document_list = []
+        collect_handler = self._get_collect_handler(document_list)
+
+        try:
+            ForkManage(source_url, selector.split(" ") if selector else []).fork(3, set(), collect_handler)
+        except CancelledException:
+            raise
+        except Exception as e:
+            maxkb_logger.error(
+                _("data source web node:{node_id} error{error}{traceback}").format(
+                    node_id=self.get_node_id(), error=str(e), traceback=traceback.format_exc()
+                )
+            )
+
+        self.write_context("document_list", document_list)
+        self.write_context("source_url", source_url)
+        self.write_context("selector", selector)
+
+    def get_details(self, index: int = 0, position: dict = None, old_details: dict = None, **kwargs):
+        details = super().get_details(index, position, old_details, **kwargs)
+        details.update(
+            {
+                "document_list": self.get_context("document_list"),
+                "source_url": self.get_context("source_url"),
+                "selector": self.get_context("selector"),
+                "knowledge_base": self.get_workflow_parameters().get("knowledge_base"),
+                "enableException": self.node.properties.get("enableException"),
+            }
+        )
+        return details

ui/src/views/details-demo/index.vue

@@ -167,12 +167,22 @@ const detail = ref<ExecutionNodeDetail[]>([
     is_submit: true,
   }),
 
+  // ===== Web 数据源 =====
+  base(WorkflowNodeType.DataSourceWebNode, 16, 'Web 数据源', {
+    selector: '.content',
+    source_url: 'https://example.com/docs',
+    document_list: [
+      { name: '前言', content: '本文介绍 MaxKB 的安装步骤与系统要求。' },
+      { name: '快速开始', content: '下载镜像后，通过 docker compose 一键启动。' },
+    ],
+  }),
+
   // ===== 循环 =====
   {
     type: WorkflowNodeType.LoopNode,
     name: '循环',
     status: 200,
-    index: 16,
+    index: 17,
     run_time: 4.567,
     loop_type: 'ARRAY',
     loop_node_data: {

ui/src/workflow-canvas/nodes/data-source-web-node/index.ts

@@ -1,4 +1,5 @@
 import DataSourceWebNode from './index.vue'
+import DataSourceWebNodeDetail from './details/index.vue'
 import { WorkflowNodeModel, WorkflowNodeView } from '@/workflow-canvas/core/workflow-node'
 import { WorkflowNodeType } from '@/workflow-canvas/types'
 
@@ -8,4 +9,4 @@ class DataSourceWebNodeView extends WorkflowNodeView {
   }
 }
 
-export default { type: WorkflowNodeType.DataSourceWebNode, model: WorkflowNodeModel, view: DataSourceWebNodeView }
+export default { type: WorkflowNodeType.DataSourceWebNode, model: WorkflowNodeModel, view: DataSourceWebNodeView, details: DataSourceWebNodeDetail }

ui/src/workflow-canvas/nodes/search-knowledge-node/details/index.vue

@@ -17,6 +17,10 @@ interface SearchParagraph {
   document_name?: string
   knowledge_name?: string
   knowledge_type?: string
+  meta?: {
+    source_file_id?: string
+    source_url?: string
+  }
   similarity?: number
   title?: string
 }