feat: LangGraph工作流核心 — Agent状态/节点/图 + 验证服务 + 知识库

agent/
  state.py: AgentState TypedDict(20字段含意图/压缩/会话/撤销)
  nodes.py: 17个节点函数(生成/修改/验证/纠错/意图分类/压缩/撤销/重置)
  graph.py: 17节点状态图,8意图路由分发

验证服务 validation_service/
  main.py: FastAPI服务,lxml XSD验证 + 结构化检查(字段引用/SQL/尺寸)

数据 data/
  sample_templates/: 4个JRXML示例模板
  corrections/: 3个错误修正案例

脚本 scripts/
  init_kb.py: Chroma知识库初始化
This commit is contained in:
2026-05-14 23:21:10 +08:00
parent 21a5fdf930
commit 4b43c5d3e4
14 changed files with 1375 additions and 0 deletions
+87
View File
@@ -0,0 +1,87 @@
"""初始化 Chroma 知识库,加载示例 JRXML 模板和错误修正案例。
用法: python scripts/init_kb.py
"""
import os
import sys
from pathlib import Path
from dotenv import load_dotenv
sys.path.insert(0, str(Path(__file__).parent.parent))
load_dotenv()
from backend.embeddings import get_embeddings
def load_templates(template_dir: Path) -> list[dict]:
docs = []
for fpath in template_dir.glob('*.jrxml'):
content = fpath.read_text(encoding='utf-8')
name = fpath.stem
docs.append({
'content': content,
'metadata': {
'source': str(fpath),
'type': 'full_report',
'name': name,
},
})
return docs
def load_corrections(corrections_dir: Path) -> list[dict]:
docs = []
for fpath in corrections_dir.glob('*.jrxml'):
content = fpath.read_text(encoding='utf-8')
docs.append({
'content': content,
'metadata': {
'source': str(fpath),
'type': 'correction_case',
'name': fpath.stem,
},
})
return docs
def main():
persist_dir = os.getenv('CHROMA_PERSIST_DIR', './db/chroma')
data_dir = Path(__file__).parent.parent / 'data'
template_dir = data_dir / 'sample_templates'
corrections_dir = data_dir / 'corrections'
docs = []
if template_dir.exists():
docs.extend(load_templates(template_dir))
print(f'{template_dir} 加载了 {len(docs)} 个模板')
if corrections_dir.exists():
corr = load_corrections(corrections_dir)
docs.extend(corr)
print(f'{corrections_dir} 加载了 {len(corr)} 个修正案例')
if not docs:
print('未找到文档,无需索引。')
return
embeddings = get_embeddings()
from langchain_chroma import Chroma
texts = [d['content'] for d in docs]
metadatas = [d['metadata'] for d in docs]
Chroma.from_texts(
texts=texts,
embedding=embeddings,
metadatas=metadatas,
persist_directory=persist_dir,
)
print(f'已将 {len(docs)} 个文档索引到 Chroma,存储位置: {persist_dir}')
if __name__ == '__main__':
main()