refactor(parse_markdown): 重构 Markdown 解析逻辑并使用 Pydantic 模型

将 MarkdownNode 类重构为使用 Pydantic 模型，提高代码的可维护性和类型安全性。同时，将解析逻辑与节点操作分离，简化代码结构。

schema/__init__.py

@@ -0,0 +1,4 @@
+from .dataset import *
+from .dataset_generation import APIProvider, LLMResponse, LLMRequest
+from .md_doc import MarkdownNode
+from .prompt_templeta import prompt_templeta

schema/md_doc.py

@@ -0,0 +1,13 @@
+from pydantic import BaseModel, Field
+from typing import List, Optional
+
+class MarkdownNode(BaseModel):
+    level: int = Field(default=0, description="节点层级")
+    title: str = Field(default="Root", description="节点标题")
+    content: Optional[str] = Field(default=None, description="节点内容")
+    children: List['MarkdownNode'] = Field(default_factory=list, description="子节点列表")
+
+    class Config:
+        arbitrary_types_allowed = True
+
+MarkdownNode.model_rebuild()

tools/parse_markdown.py

@@ -1,28 +1,24 @@
 import re
+import sys
+from pathlib import Path
 
-class MarkdownNode:
+# 添加项目根目录到sys.path
-    def __init__(self, level=0, title="Root"):
+sys.path.append(str(Path(__file__).resolve().parent.parent))
-        self.level = level
+from schema import MarkdownNode
-        self.title = title
-        self.content = ""  # 使用字符串存储合并后的内容
-        self.children = []
 
-    def __repr__(self):
+def add_child(parent, child):
-        return f"({self.level}) {self.title}"
+    parent.children.append(child)
 
-    def add_child(self, child):
+def print_tree(node, indent=0):
-        self.children.append(child)
+    prefix = "│  " * (indent - 1) + "└─ " if indent > 0 else ""
-
+    print(f"{prefix}{node.title}")
-    def print_tree(self, indent=0):
+    if node.content:
-        prefix = "│  " * (indent - 1) + "└─ " if indent > 0 else ""
+        content_prefix = "│  " * indent + "├─ [内容]"
-        print(f"{prefix}{self.title}")
+        print(content_prefix)
-        if self.content:
+        for line in node.content.split('\n'):
-            content_prefix = "│  " * indent + "├─ [内容]"
+            print("│  " * indent + "│  " + line)
-            print(content_prefix)
+    for child in node.children:
-            for line in self.content.split('\n'):
+        print_tree(child, indent + 1)
-                print("│  " * indent + "│  " + line)
-        for child in self.children:
-            child.print_tree(indent + 1)
 
 def parse_markdown(markdown):
     lines = markdown.split('\n')
@@ -51,10 +47,10 @@ def parse_markdown(markdown):
         if match:
             level = len(match.group(1))
             title = match.group(2)
-            node = MarkdownNode(level, title)
+            node = MarkdownNode(level=level, title=title, content="", children=[])
             while stack[-1].level >= level:
                 stack.pop()
-            stack[-1].add_child(node)
+            add_child(stack[-1], node)
             stack.append(node)
         else:
             if stack[-1].content:
@@ -65,9 +61,9 @@ def parse_markdown(markdown):
 
 if __name__=="__main__":
     # 从文件读取 Markdown 内容
-    with open("example.md", "r", encoding="utf-8") as f:
+    with open("workdir/example.md", "r", encoding="utf-8") as f:
         markdown = f.read()
 
     # 解析 Markdown 并打印树结构
     root = parse_markdown(markdown)
-    root.print_tree()
+    print_tree(root)