Skip to content

release: ai 809c821 to production - #199

Merged
jm-rivera merged 1 commit into
mainfrom
release/2026-11-14-3
Nov 14, 2026
Merged

jm-rivera merged 1 commit into
mainfrom
release/2026-11-14-3

Conversation

@jm-rivera

Copy link
Copy Markdown
Contributor

Releases ai at 809c821 to production, cut on release/2026-11-14-3: one pull request after the release at 33f312a.

  • fix(compiler): read a family a declared composite sums as one measure #198 reads a family dfo-apis marks sums_components as one summed measure, so a ranking over DAC2a's net bilateral plus imputed multilateral composite compiles. This fixes a plan card's provider switch to that option, which failed in production with "a calculation was asked of data in a shape it cannot read". A refused plan revision now logs its code, constraint and reason.

Release order: production apis must serve sums_components first (DevelopmentFinanceObservatory/apis release of 2f3b925); without it the harness reads every family as today. No migration, setting or secret change.

Testing: CI passed on #198, and Deploy dev passed on 809c821. On dev Studio, "Which recipients got the most net ODA from France in 2023? Top 10." with the plan card's provider switched to "net bilateral plus imputed multilateral as published, by convention" applies the switch and draws France's top 10 under that count. Repeat on production after both deploys.

…#198)

* fix(compiler): read a family a declared composite sums as one measure

A plan card ranking France's DAC2a aid to recipients could not switch to "net bilateral plus imputed multilateral as published, by convention": the compiler read the composite's two StatVars as two measures and refused the ranking with `input_shape_unsupported`. dfo-apis adds a stacked composite's components by declaration and now says so in `resolved.stat_vars.sums_components`. `source_from_spec_block` carries the flag onto `SelectedSource.sums_components`, and `default_reading` reads such a family summed unless an open axis cannot be summed. A block without the flag reads as before.

A refused plan revision now logs its code and the compiler's raw detail before the reason is reworded for the user, so a refused switch can be diagnosed from the logs.

* fix(harness): log a refused revision's code, constraint and a capped reason only

The detail dict can hold several issues naming the user's entities. The log keeps what diagnosis needs: the refusal code, the compile issue code and the compiler's reason, cut to 500 characters.
@jm-rivera
jm-rivera merged commit f6b1381 into main Nov 14, 2026
11 checks passed
@jm-rivera
jm-rivera deleted the release/2026-11-14-3 branch November 14, 2026 17:04

This branch was successfully deployed

1 active deployment
dev — 809c8213 Deployed Nov 14, 2026 by jm-rivera via deploy (dev) / deploy #94
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant