From 29d14ab81772d390fdd97f025ac26dc2d7f5287b Mon Sep 17 00:00:00 2001
From: sansen <sansenhoshi@qq.com>
Date: Fri, 7 Aug 2026 18:02:57 +0800
Subject: [PATCH] =?UTF-8?q?=E4=BF=AE=E5=A4=8D=E5=AD=97=E7=AC=A6=E5=81=8F?=
 =?UTF-8?q?=E7=A7=BB?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 parser.py | 22 ++++++++--------------
 1 file changed, 8 insertions(+), 14 deletions(-)

diff --git a/parser.py b/parser.py
index 2373ab8..25f3474 100644
--- a/parser.py
+++ b/parser.py
@@ -251,8 +251,8 @@ _patterns_dynasty = _build_pattern(with_dynasty=True)
 
 # === 主函数 ===
 def extract_era_years(text):
-    # 去除标记用的方括号（用户标注用，非原文内容）
-    text = text.replace('[', '').replace(']', '')
+    # 去除标记用的方括号/尖括号（用户标注用，非原文内容）
+    text = text.replace('[', '').replace(']', '').replace('【', '').replace('】', '')
     original_text = text
     simplified_text = normalize_chars(cc.convert(text))
     results = []
@@ -381,19 +381,22 @@ def extract_era_years(text):
         year_num = chinese_to_num(year_text)
         if not (1000 <= year_num <= 2100):
             continue
+        # 公元年后紧跟的干支（如"公元一九六〇年庚子"）：无年号可锚定时，
+        # 干支就是年份的干支，并入同一日期单位并校验
+        ganzhi = simplified_text[end:end + 2]
+        if ganzhi in _ganzhi_to_offset:
+            end += 2
         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）"
         result = {
             "原文": raw_fan,
-            "原文(简体)": m.group(0),
+            "原文(简体)": simplified_text[start:end],
             "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:
@@ -635,15 +638,6 @@ 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