UK replay gate: a data-limited inert measure blocks certification of every pe-uk release after #2241
维护者通常 1 天内回复
评估
- 难度
- 4/5
- 预计耗时
- 3-5 天
- 新手友好度
- 35/100
- Issue 类型
- 功能
- 描述清晰度
- 描述清楚
- 活跃度
- 活跃
- 技术栈
- python
调研方向
Start with qualification.py and the scorecard replay path for autumn_budget_2024__sdlt_additional_dwelling_surcharge_2pp, using the cited replay-state receipts as the initial reproduction. Trace how integrity failures and budget_inputs evidence are recorded, then locate the policy configuration for d1344-replay-regressions-v2. Done means the specified reviewed, code/event/measure/year-matched exceptions are narrowly applied only while every required input is missing, with matched regressions remaining review items rather than passes.
由索引模型根据 Issue 内容生成。
描述
No policyengine-uk release after policyengine-uk#2241 can be certified, even by a named reviewer.
What happens. #2241 (merged 15 October) correctly stops charging property stocks as purchases. SDLT higher rates now apply to additional_residential_property_purchased, which no uk-data release provides yet (the imputation is policyengine-uk-data#573). So on candidates 2.134.6 and 2.134.7 (with uk-data 1.58.2), the scorecard replay of autumn_budget_2024__sdlt_additional_dwelling_surcharge_2pp is inert for 2025–2029: base £2.50bn in 2029 under the old bug, head £0. Those 5 construction_inert rows are the receipts' only integrity failures (replay-state receipts b989d471…, 71e40ada…). The same attempts' budget_inputs gate reports the input as newly missing.
Why nothing can qualify. qualification.py fails every receipt whose integrity failed. Under the d1344 rule, a failure can't be accepted on review, and a known gap never excuses a £0 total against a material base value.
Proposed change. Add narrow, receipt-recorded data-limited exceptions:
- a reviewed registry entry matched by code, event, measure key and year;
- it applies only while the same attempt's
budget_inputsevidence reports every required input missing, so it expires once the data provides them; - matched regressions become review items, never a pass;
- the policy moves to
d1344-replay-regressions-v2.
- 主要语言
- Python
- 星标
- 8
- 派生
- 10
- 平均合并
- 21 小时 34 分钟
- 30 天内合并 PR
- 45
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 没有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
PolicyEngine/policyengine.py 的其他 Issue
-
Raise the simulation cache memory warning to 16 GiB可能已有人在做 @anth-volk 于 17 天前认领。 未关闭
难度 2/5 1-3 小时 新手友好度 76/100
PolicyEngine/policyengine.py#549 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 72/100
PolicyEngine/policyengine.py#473 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 68/100
PolicyEngine/policyengine.py#436 · 1 条评论 ·
维护者通常 1 天内回复
-
Certify populace-uk as the UK certified default可能重新可做 关联的 PR 已关闭且未合并。 未关闭
难度 2/5 1-3 小时 新手友好度 72/100
PolicyEngine/policyengine.py#408 ·
维护者通常 1 天内回复
-
Restore ACS-local US regions on the microcosm#1263 rebuild可能已有人在做 @MaxGhenis 于 1 天前认领。 未关闭
难度 5/5 一周以上 新手友好度 25/100
PolicyEngine/policyengine.py#648 ·
维护者通常 1 天内回复
查看 PolicyEngine/policyengine.py 的全部 Issue
相似的 Issue
-
Examples say cnx未关闭documentation embedded
难度 2/5 1-3 小时 新手友好度 68/100
维护者通常 1 天内回复
-
难度 1/5 1 小时以内 新手友好度 88/100
Alishahryar1/free-claude-code#2111 ·
维护者通常 1 天内回复
-
bug
难度 2/5 1-3 小时 新手友好度 68/100
维护者通常 1 天内回复
-
namespace operations
难度 2/5 1-3 小时 新手友好度 72/100
EclipseFdn/open-vsx.org#14373 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 85/100
维护者通常 1 天内回复