fix(application-form-fill): 抽取说明书章节和表格字段

tests/test_regulatory_text_extract.py

@@ -37,3 +37,25 @@ def test_extract_text_reports_unsupported_file(tmp_path):
 
     assert result.status == "unsupported"
     assert result.text == ""
+
+
+def test_extract_text_from_docx_preserves_table_text(tmp_path):
+    from docx import Document
+
+    path = tmp_path / "说明书.docx"
+    document = Document()
+    document.add_paragraph("【主要组成成分】")
+    table = document.add_table(rows=2, cols=2)
+    table.rows[0].cells[0].text = "组分"
+    table.rows[0].cells[1].text = "数量"
+    table.rows[1].cells[0].text = "PCR反应液"
+    table.rows[1].cells[1].text = "1管"
+    document.add_paragraph("【储存条件及有效期】")
+    document.add_paragraph("-20±5℃保存，有效期12个月。")
+    document.save(path)
+
+    result = extract_text(path)
+
+    assert result.status == "success"
+    assert "组分\t数量" in result.text
+    assert result.text.index("PCR反应液") < result.text.index("【储存条件及有效期】")