业务本体与标准实体:跨源键值对齐、对账共变算子、健康证据链与实体助手

- 新增标准业务实体(CanonicalEntity)模型、确定性跨源编码翻译、后台候选实体扫描与文档驱动实体智能助手
- 新增指标本体图谱(factor / reconciles_with / upstream_driver)、DAG 防环校验以及 ontology_neighbors / source_health 内置工具
- 新增跨源对账(analysis_reconcile)与滞后共变(analysis_cochange)分析算子,打通 [[h:H1]] 数据源健康证据链、报告导出与本体消融评测
- 数据目录新增「标准业务实体」视图与候选审阅工作台,口径页集成指标本体拓扑面板与端到端浏览器检查
This commit is contained in:
ThearchyHelios
2026-11-12 12:57:53 +08:00
parent 2f90f0eb3a
commit 8d9a712eae
106 changed files with 16654 additions and 295 deletions
+6 -5
View File
@@ -33,14 +33,14 @@ logger = logging.getLogger(__name__)
#: 候选维度「取值不多」的上限:码值表里的取值不超过这么多(取值观察最多也只取这么多,data/catalog_profile_observe)
CANDIDATE_MAX = 50
#: 现象:可比性、基线、趋势
_PHENOMENA = ("comparable", "baseline", "trend")
#: 现象:可比性、基线、趋势、跨源对账、跨源联动
_PHENOMENA = ("comparable", "baseline", "trend", "reconcile", "cochange")
#: 事实表的表类型(数据目录的 kind):流水、日志也是一行一件事
_FACT_KINDS = ("fact", "log")
#: 候选维度的度量类型
_DIMENSION_MEASURES = ("attribute", "status")
#: 拆解的输入快照(下钻的上一层不是快照)
_DATA_ROLES = ("data", "base", "target")
#: 分析算子的输入快照(下钻的上一层不是快照)
_DATA_ROLES = ("data", "base", "target", "primary", "counterpart", "driver")
NOTE_NO_CATALOG = "这次运行读不到数据目录,候选维度核对不了"
NOTE_NO_NOTES = "拆解所用的数据源还没有数据目录,候选维度核对不了"
@@ -288,7 +288,8 @@ def build(catalog: Mapping[str, Any], *, sources: Mapping[str, Mapping[str, Any]
decomposes = [r for _, r in ok if r.get("op") == "decompose"]
excluded = [items[a] for a, r in ok
if (r.get("op") == "decompose" and _code(r) == "dispersed")
or (r.get("op") == "differ" and _code(r) == "no_evidence")]
or (r.get("op") == "differ" and _code(r) == "no_evidence")
or (r.get("op") == "cochange" and _code(r) in ("target_normal", "primary_normal", "independent"))]
tests = sum(1 for _, r in ok if r.get("op") == "differ")
try:
missing, note = _undecomposed([r for r in decomposes if _dims(r)], catalog, sources, load)