框架: 立卡路补守卫(初卡字段拆分/剥前缀/垃圾拦截)+前缀正则放宽至章号范围变体

.claude/skills/parse-book/scripts/parse_upgrade.py

@@ -251,8 +251,9 @@ def _strip_prefix(s):
     import re as _re
     global WIN_PREFIX_RE
     if WIN_PREFIX_RE is None:
-        # [窗…]/[本窗…] 任意变体全剥（窗29实测模型自造 "[窗本窗]"，仅数字版剥不掉）
-        WIN_PREFIX_RE = _re.compile(r"^(?:\[[窗本][^\]]{0,6}\]\s*)+")
+        # [窗…]/[本窗…] 任意变体全剥（窗29实测模型自造 "[窗本窗]"，仅数字版剥不掉；
+        # 深空实测又造 "[窗387-388]" 章号范围变体，7 字符超旧上限 6——放宽到 12）
+        WIN_PREFIX_RE = _re.compile(r"^(?:\[[窗本][^\]]{0,12}\]\s*)+")
     return WIN_PREFIX_RE.sub("", str(s)).strip()
 
 
@@ -425,10 +426,26 @@ def new_card(conn, work_id, win_no, ent):
         note = _clean_alias(m.group(2))
         if note:
             extra_alias.append(note)
+    # 初卡字段过守卫（深空 4917 实测：立卡路不走 merge_card，粘连/自造前缀/垃圾
+    # 原样入库——更新路守了、立卡路漏了）：列表值逐条拆分、剥前缀、垃圾拦截、带窗号
+    fields0 = {}
+    for k, v in (ent.get("字段", {}) or {}).items():
+        if not isinstance(v, list):
+            fields0[k] = v
+            continue
+        out, seen = [], set()
+        for it in v:
+            for seg in re.split(r"；\s*(?=\[[窗本])", str(it)):
+                core = _strip_prefix(seg)
+                if not core or _is_garbage(core) or core in seen:
+                    continue
+                out.append(f"[窗{win_no}] {core}")
+                seen.add(core)
+        fields0[k] = out
     payload = {"type": ent["型"], "名称": raw,
                "别名": [x for x in ([_clean_alias(a) for a in ent.get("别名", [])] + extra_alias) if x],
                "一句话摘要": ent.get("一句话摘要", ""),
-               "字段": ent.get("字段", {}) or {},
+               "字段": fields0,
                "出场章": sorted(set(ent.get("出场章", []))),
                "来源": f"升格@窗{win_no}", "状态": "草稿",
                "目标库": "本书作品库", "可见范围": "本书私有",