增加生成学校易耗品填写的转换程序

src/core/extraction/extractor.py

@@ -115,6 +115,8 @@ def _build_summary(data: dict[str, Any]) -> dict[str, str]:
         "general": "普通发票",
         "payment": "支付记录",
         "application": "出差申请单",
+        "attachments": "其他附件",
+        "note": "易耗品单",
     }
 
     summary = {}
@@ -259,11 +261,12 @@ def extract_invoices(
         - 出差事前申请单 -> 单独存储，不参与匹配
 
     Returns:
-        (payment_records, applications, groups):
-            - payment_records: 支付记录列表（仅包含真实发票，不含申请单）
-            - applications: 出差事前申请单列表（单独存储，不参与支付匹配）
+        元组 (payment_records, applications, groups):
+            - payment_records: 支付记录列表（每项含关联发票 _matched_invoices）
+            - applications: 出差事前申请单列表
             - groups: 按文档类型分组的字典
-                {'travel': [差旅发票], 'general': [普通发票], 'application': [出差事前申请单]}
+                {'travel': [差旅发票], 'general': [普通发票], 'application': [出差事前申请单],
+                 'attachments': [其他附件]}
 
     Raises:
         ExtractionError: 当所有文件提取均失败时抛出，携带失败文件列表和原始错误。
@@ -280,13 +283,14 @@ def extract_invoices(
     all_files = _find_all_files(directory)
     if not all_files:
         log.warning("未找到支持的文件")
-        return [], [], {"travel": [], "general": [], "application": []}
+        return [], [], {"travel": [], "general": [], "application": [], "attachments": []}
 
     log.info(f"发现 {len(all_files)} 个文件")
 
     all_invoices = []
     all_cards = []
     applications = []
+    attachments = []
     # 记录失败文件及其错误信息
     failed_files: list[tuple[str, str]] = []
 
@@ -318,6 +322,9 @@ def extract_invoices(
             elif inv_type == "payment":
                 all_cards.append(result)
                 log.info(f"[{inv_type}] 已解析: {file_path.name}")
+            elif inv_type in ("attachments", "note"):
+                attachments.append(result)
+                log.info(f"[{inv_type}] 已解析: {file_path.name}")
             elif result.get("invoice_number"):
                 all_invoices.append(result)
                 log.info(f"[{inv_type}] 已解析: {file_path.name}")
@@ -334,7 +341,9 @@ def extract_invoices(
             details=error_details,
         )
 
-    log.info(f"分类结果: 发票 {len(all_invoices)} 张, 支付记录 {len(all_cards)} 条, 申请单 {len(applications)} 份")
+    log.info(
+        f"分类结果: 发票 {len(all_invoices)} 张, 支付记录 {len(all_cards)} 条, 申请单 {len(applications)} 份, 附件 {len(attachments)} 个"
+    )
 
     if not all_invoices:
         log.warning("未成功解析任何发票")
@@ -349,10 +358,12 @@ def extract_invoices(
     all_documents.extend(applications)
 
     groups = classify_invoice_batch(all_documents)
+    groups["attachments"] = attachments
     log.info(
         f"文档分类: 差旅发票 {len(groups['travel'])} 张, "
         f"普通发票 {len(groups['general'])} 张, "
-        f"出差事前申请单 {len(groups['application'])} 份"
+        f"出差事前申请单 {len(groups['application'])} 份, "
+        f"附件 {len(groups['attachments'])} 个"
     )
 
     return payment_records, applications, groups