[CI bisect — do not review] INR schema change minus rail property#650
[CI bisect — do not review] INR schema change minus rail property#650ls-bolt[bot] wants to merge 1 commit into
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub. 2 Skipped Deployments
|
✱ Stainless preview builds for gridThis PR will update the cli csharp go kotlin openapi php python ruby typescript Edit this comment to update them. They will appear in their respective SDK's changelogs. ✅ grid-openapi studio · code · diff
✅ grid-ruby studio · code · diff
✅ grid-go studio · code · diff
✅ grid-kotlin studio · code · diff
|
13aa9e8 to
848999a
Compare
|
Round 1 result: csharp still failed with the full change minus Round 2 (this push): branch now carries only the |
848999a to
912ad46
Compare
|
Round 2 result: csharp failed on the enum-only diff (+NEFT/+RTGS, 2 lines) — a shape that already exists on EUR/DKK ( Round 3 (this push): zero schema changes — a one-word edit to |
…iption), no structural changes (do not review)
912ad46 to
b1f940c
Compare
|
Round 3 result: csharp passed on the Round 4 (this push): a field-description-only change (GTQ
|
|
Round 4 result: csharp failed on a description-only change (GTQ The Stainless csharp target fails its build CI for ANY freshly-generated code right now, regardless of content. Diffs that don't change generated csharp (docs-only PRs, #646 is exonerated — its content is fine (all real spec gates green). The csharp target is unpublished ( Closing this diagnostic PR. |
…#671) ## Summary Removes the csharp target from `.stainless/stainless.yml` so the `Build SDKs for pull request / preview` check stops failing on every spec-changing PR. **Why every spec PR is red**: Stainless's csharp build CI is broken upstream for ANY freshly generated code (content-independent, proven by bisect on #650, breakage window ~Jul 3). The preview check gates on a per-language head-vs-base regression, and with csharp in main's config the base build resolves to a cached pre-breakage csharp *success* — so every fresh head failure counts as a regression → red. **Why this fixes it**: with csharp out of the config on main, base builds have no cached csharp success to regress from — matching outcomes on base and head → "not a regression" → green. Proven end-to-end by probe #659: the exact previously-failing change class (doc-comment-only codegen delta) ran against a csharp-free base and `preview` **passed**. Full mechanics: [investigation writeup](#657 (comment)). **No shipping impact**: the csharp target is unpublished (`production_repo: null`, `publish.nuget: false`). Re-add the block when Stainless fixes their csharp build CI upstream. ## Sequencing (for #657 and other open spec PRs) 1. Merge this PR — **its own `preview` will be red** (its base is current main); merge through it (precedent: #633, #637). 2. Merge/rebase main into open spec PRs (e.g. #657) — their next `preview` run goes green. ## Verification - `make lint` passes; the file matches the exact removal previously validated on the #657 branch (commit 069ad9a) and by probe #659. --- 🤖 [thundering-graviton-3](https://zeus.dev.dev.sparkinfra.net/#/arc?id=thundering-graviton)[(#3)](https://zeus.dev.dev.sparkinfra.net/#/instance?id=thundering-graviton-3) | [Feedback](https://zeus.dev.dev.sparkinfra.net/feedback) Original PR: #663 Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

Throwaway diagnostic PR — do not review, will be closed shortly.
Bisecting the chronic
Build SDKs for pull request / previewcsharp failure on #646 (and #644): this branch carries the exact INR schema change from #646 minus the newrailproperty, to determine whetherrailis what breaks the Stainless csharp SDK build.railis the breakerContext: #646's preview failed twice with
csharp: Your SDK build had a failure in the build CI job, which is a regression from the base state, as did auto-sync #644 which carries the same INR fields.Slack thread: https://lightsparkgroup.slack.com/archives/D0BCM1RUHHU/p1783372549749679
🤖 bonded-titan(#1) | Feedback