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
2026-05-14 23:21:10 +08:00
parent 21a5fdf930
commit 4b43c5d3e4
14 changed files with 1375 additions and 0 deletions
@@ -0,0 +1,129 @@
+"""JRXML 文件验证服务（FastAPI）。
+
+使用 lxml XML Schema 验证作为 JasperReports 7.0.6 编译验证的第一阶段后备方案。
+要进行完整的编译验证，需要基于 Java 的验证器以及 JasperReports 7.0.6 + JDK 21。
+
+启动: uvicorn validation_service.main:app --port 8001
+"""
+
+import re
+import xml.etree.ElementTree as ET
+from pathlib import Path
+
+from fastapi import FastAPI
+from lxml import etree
+from pydantic import BaseModel
+
+app = FastAPI(title="JRXML 验证服务")
+
+SCHEMA_DIR = Path(__file__).parent / "schemas"
+SCHEMA_FILE = SCHEMA_DIR / "jasperreport_7_0_6.xsd"
+
+
+class ValidationRequest(BaseModel):
+    jrxml: str
+
+
+class ValidationResponse(BaseModel):
+    valid: bool
+    error: str
+
+
+def _check_structural_issues(jrxml: str) -> list[str]:
+    """检查 JRXML 中常见的结构性问题。"""
+    issues = []
+    root = None
+
+    try:
+        root = ET.fromstring(jrxml)
+    except ET.ParseError as e:
+        issues.append(f"XML 解析错误：{e}")
+        return issues
+
+    # 同时处理带命名空间和不带命名空间的元素名
+    ns = "http://jasperreports.sourceforge.net/jasperreports"
+
+    declared_fields = set()
+    for elem in root.iter():
+        tag = elem.tag.split("}")[-1] if "}" in elem.tag else elem.tag
+        if tag == "field":
+            name = elem.get("name")
+            if name:
+                declared_fields.add(name)
+
+    field_expr_pattern = re.compile(r'\$F\{(\w+)\}')
+    for m in field_expr_pattern.finditer(jrxml):
+        field_name = m.group(1)
+        if field_name not in declared_fields:
+            issues.append(
+                f"字段 '{field_name}' 在表达式中使用但未在 <field> 部分声明"
+            )
+
+    query = None
+    for elem in root.iter():
+        tag = elem.tag.split("}")[-1] if "}" in elem.tag else elem.tag
+        if tag == "queryString":
+            query = elem
+            break
+    if query is not None:
+        query_text = "".join(query.itertext()).strip()
+        if not query_text:
+            issues.append("<queryString> 为空 - 请在 CDATA 中添加 SQL 查询")
+        elif not any(kw in query_text.upper() for kw in ["SELECT"]):
+            issues.append("<queryString> 似乎不包含 SQL SELECT 查询")
+
+    if not root.get("pageWidth"):
+        issues.append("缺少 <jasperReport> 上的 pageWidth 属性")
+    if not root.get("pageHeight"):
+        issues.append("缺少 <jasperReport> 上的 pageHeight 属性")
+    if not root.get("name"):
+        issues.append("缺少 <jasperReport> 上的 'name' 属性")
+
+    return issues
+
+
+def _validate_xsd(jrxml: str) -> tuple[bool, str]:
+    """根据 JasperReports XSD schema 验证 JRXML。"""
+    if not SCHEMA_FILE.exists():
+        return True, ""
+
+    try:
+        schema_doc = etree.parse(str(SCHEMA_FILE))
+        xmlschema = etree.XMLSchema(schema_doc)
+        doc = etree.fromstring(jrxml.encode("utf-8"))
+        xmlschema.assertValid(doc)
+        return True, ""
+    except etree.DocumentInvalid as e:
+        return False, str(e)
+    except etree.XMLSchemaError as e:
+        return False, f"Schema 错误：{e}"
+    except Exception as e:
+        return False, f"XML 验证错误：{e}"
+
+
+@app.post("/validate", response_model=ValidationResponse)
+async def validate_jrxml(req: ValidationRequest):
+    jrxml = req.jrxml.strip()
+    if not jrxml:
+        return ValidationResponse(valid=False, error="JRXML 内容为空")
+
+    structural_issues = _check_structural_issues(jrxml)
+    if structural_issues:
+        return ValidationResponse(valid=False, error="; ".join(structural_issues))
+
+    valid, xsd_error = _validate_xsd(jrxml)
+    if not valid:
+        return ValidationResponse(valid=False, error=xsd_error)
+
+    return ValidationResponse(valid=True, error="")
+
+
+@app.get("/health")
+async def health():
+    schema_available = SCHEMA_FILE.exists()
+    return {
+        "status": "ok",
+        "schema_available": schema_available,
+        "validation_type": "XSD" if schema_available else "仅结构检查",
+        "note": "如需完整的 JasperReports 7.0.6 编译验证，请使用基于 Java 的验证器",
+    }
@@ -0,0 +1,6 @@
+@echo off
+echo 正在启动 JRXML 验证服务...
+echo.
+cd /d "%~dp0"
+python -m uvicorn validation_service.main:app --host 0.0.0.0 --port 8001 --reload
+pause