diff --git a/parser.py b/parser.py index f3a09a6..2373ab8 100644 --- a/parser.py +++ b/parser.py @@ -384,14 +384,21 @@ def extract_era_years(text): covered.update(range(start, end)) raw_fan = original_text[start:end] era_type = "公元纪年(中华人民共和国)" if year_num >= 1949 else "公元纪年(近代/其他)" - results.append({ + result = { "原文": raw_fan, "原文(简体)": m.group(0), "类型": era_type, "公元": year_num, "公元中文": f"{arabic_to_chinese_year(year_num)}年", "位置": {"起始": start, "结束": end}, - }) + } + # 公元年后紧跟的干支(如"公元一九六〇年庚子")用于校验年份一致性 + ganzhi = ''.join(ch for ch in simplified_text[end:end + 4] if ch not in '【】').strip()[:2] + if ganzhi in _ganzhi_to_offset: + result["干支"] = ganzhi + if gregorian_to_ganzhi(year_num) != ganzhi: + result["干支不符"] = True + results.append(result) # === 3. 干支纪年(如"清雍正乙巳十一月十二午时生")=== def _process_ganzhi_match(m): @@ -628,6 +635,15 @@ def extract_era_years(text): if result.get("公元") is not None: result.setdefault("干支", gregorian_to_ganzhi(result["公元"])) + # === 收尾:紧邻的【】标注(用户标记的日期跨度)纳入原文 === + for result in results: + start, end = result["位置"]["起始"], result["位置"]["结束"] + if (start > 0 and simplified_text[start - 1] == '【' + and end < len(simplified_text) and simplified_text[end] == '】'): + result["位置"] = {"起始": start - 1, "结束": end + 1} + result["原文"] = original_text[start - 1:end + 1] + result["原文(简体)"] = simplified_text[start - 1:end + 1] + return results