feat: PubMed搜索合规审计 Phase 5-7 全部修复

P1-2: [SB][STAT][UID][DEP][MESH]字段支持
P1-4: [MH:noexp]解析+抑制树展开
P3: 前端date_to/P3-1、CJK高亮/P3-3、extractPlainText/P3-4、precision_mode移除/P3-6、loadMore sort/P3-5
P4-1: 词数上限50→100
P4-2: retracted参数清理

backend/app/services/pubmed_query_parser.py

@@ -23,7 +23,7 @@ from dataclasses import dataclass, field
 from enum import Enum, auto
 
 # ─── 查询复杂度限制 ───
-MAX_TERMS = 50       # 与 API 层 check_query_complexity 的 50 词限制对齐
+MAX_TERMS = 100      # P4-1: 放宽到 100 词（原 50 词）
 MAX_PAREN_DEPTH = 10  # 括号嵌套最深层数
 
 
@@ -42,6 +42,7 @@ _FIELD_TAG_MAP: dict[str, str] = {
     "LAU": "author",
     "TW": "all",
     "OT": "all",
+    "MESH": "MH",            # P1-2: [MESH] 是 [MH] 的别名
     # 新增标量字段
     "LA": "language",
     "VI": "volume",
@@ -55,23 +56,30 @@ _SPECIAL_FIELDS = {
     "MH", "MAJR", "PT", "DP", "PMID", "DOI",
     # 日期字段
     "EDAT", "CRDT", "MHDA", "LR", "DCOM",
+    # P1-2: 电子出版日期
+    "DEP",
     # JSONB 字段
     "GR", "SH", "RN", "NM", "SI", "PA",
     # 待补充抽取的字段（标记为特殊以便未来实现）
     "AUID", "COIS", "ED", "IR", "PS", "PUBN", "TT",
+    # P1-2: 新字段
+    "SB", "STAT", "UID",
 }
 
 # 支持日期范围语法的字段
-_DATE_RANGE_FIELDS = {"DP", "EDAT", "CRDT", "MHDA", "LR", "DCOM"}
+_DATE_RANGE_FIELDS = {"DP", "EDAT", "CRDT", "MHDA", "LR", "DCOM", "DEP"}
 
 # 所有合法字段标签（PubMed 全量字段）
 # P1-1: 移除了 BOOK/FILTER/ISBN（未实现，降级为 plain text 不如报错透明）
 _ALL_FIELD_TAGS = {
     "AB", "AD", "AU", "CN", "FAU", "AUID", "LAU", "COIS",
     "DCOM", "CRDT", "EDAT", "MHDA", "LR", "DP", "DOI",
+    "DEP",                               # P1-2: Date of Electronic Publication
     "RN", "ED", "GR", "IR", "IP",
     "TA", "JT", "LA", "LID", "MAJR", "SH", "MH", "MH:NOEXP", "OT", "PG",
     "PA", "PT", "PMID", "PUBN", "SI", "PS", "NM", "TW",
+    "SB", "STAT", "UID",                 # P1-2: Subset, Status, UID
+    "MESH",                              # P1-2: [MH] 别名
     "TI", "TIAB", "TT", "VI",
 }
 
@@ -187,6 +195,8 @@ class ParsedPubmedQuery:
     lr_to: str | None = None
     dcom_from: str | None = None                                   # [DCOM]
     dcom_to: str | None = None
+    dep_from: str | None = None                                     # P1-2: [DEP]
+    dep_to: str | None = None
     # 特殊字段（存储 Term 以保留 is_not 标志）
     mesh_terms: list[Term] = field(default_factory=list)           # [MH] (was list[str])
     majr_terms: list[Term] = field(default_factory=list)           # [MAJR] (was list[str])
@@ -207,6 +217,9 @@ class ParsedPubmedQuery:
     auid_terms: list[Term] = field(default_factory=list)             # [AUID]
     cois_terms: list[Term] = field(default_factory=list)             # [COIS]
     tt_terms: list[Term] = field(default_factory=list)               # [TT]
+    sb_terms: list[Term] = field(default_factory=list)               # P1-2: [SB] Subset
+    stat_terms: list[Term] = field(default_factory=list)             # P1-2: [STAT] Status
+    uid_terms: list[Term] = field(default_factory=list)              # P1-2: [UID] PMID/DOI 统一
     plain_terms: list[Term] = field(default_factory=list)          # no field tag
     boolean_operator: str = "and"                                  # "and" | "or" | "mixed"
     has_not: bool = False                                          # contains NOT
@@ -349,6 +362,8 @@ class PubmedQueryParser:
             pass
         elif term.field == "__RANGE_DCOM__":
             pass
+        elif term.field == "__RANGE_DEP__":
+            pass
         elif term.field == "ED":
             result.ed_terms.append(term)
         elif term.field == "IR":
@@ -363,6 +378,12 @@ class PubmedQueryParser:
             result.cois_terms.append(term)
         elif term.field == "TT":
             result.tt_terms.append(term)
+        elif term.field == "SB":
+            result.sb_terms.append(term)
+        elif term.field == "STAT":
+            result.stat_terms.append(term)
+        elif term.field == "UID":
+            result.uid_terms.append(term)
         else:
             result.plain_terms.append(term)
 
@@ -494,6 +515,7 @@ class PubmedQueryParser:
                 "MHDA": ("mhda_from", "mhda_to", None, None, "__RANGE_MHDA__"),
                 "LR": ("lr_from", "lr_to", None, None, "__RANGE_LR__"),
                 "DCOM": ("dcom_from", "dcom_to", None, None, "__RANGE_DCOM__"),
+                "DEP": ("dep_from", "dep_to", None, None, "__RANGE_DEP__"),  # P1-2
             }
             date_attr, date_attr_to, yr_from_attr, yr_to_attr, marker_field = attr_map[field]
             # 反向范围自动交换（如 2026:2024[DP] → 2024:2026[DP]）