fix(application-form-fill): 代理人字段暂用生产企业信息

review_agent/application_form_fill/services/field_extract.py

@@ -17,6 +17,11 @@ from review_agent.regulatory_review.services.text_extract import extract_text
 
 FIELD_ALIASES = {
     "product_name": ["产品名称"],
+    "applicant_name": ["注册人名称", "生产企业名称", "企业名称", "生产企业"],
+    "applicant_address": ["注册人住所", "生产企业住所", "企业住所", "住所"],
+    "manufacturer_address": ["生产地址", "生产企业地址", "生产场所"],
+    "agent_name": ["代理人名称", "生产企业名称", "企业名称", "生产企业", "注册人名称"],
+    "agent_address": ["代理人住所", "生产企业住所", "企业住所", "住所", "注册人住所"],
     "package_specification": ["包装规格", "规格"],
     "main_components": ["主要组成成分", "主要组成", "组成成分"],
     "intended_use": ["预期用途"],
@@ -41,7 +46,7 @@ def collect_document_texts(summary_batch: FileSummaryBatch) -> dict[str, str]:
 def extract_by_rules(texts: dict[str, str], specs: list[TemplateSpec]) -> dict[str, Any]:
     fields: list[dict[str, Any]] = []
     field_defs = _field_defs(specs)
-    labels = [field["label"] for field in field_defs if field.get("label")]
+    labels = _all_field_labels(field_defs)
     for file_name, text in texts.items():
         source_role = detect_source_role(file_name, text)
         for field in field_defs:
@@ -174,6 +179,15 @@ def _field_aliases(field: dict[str, str]) -> list[str]:
     return result
 
 
+def _all_field_labels(fields: list[dict[str, str]]) -> list[str]:
+    labels: list[str] = []
+    for field in fields:
+        for label in _field_aliases(field):
+            if label not in labels:
+                labels.append(label)
+    return labels
+
+
 def _extract_label_value(text: str, label: str, labels: list[str]) -> tuple[str, str]:
     return _extract_colon_label_value(text, label, labels)