test(data): seam the money path — table-test the ingest outcome matrix (#1009 P0)

[LukasWodka]

Summary

runDataIngest's tail — submit → port-forward → watch → classify → JSON → reclaim — is the path a customer actually lives in, and it had no injection seam, so it was covered by nothing (36.6%). A regression where the CLI previews success but the ingestor writes null/fewer, or where it reclaims the staging copy on a partial failure (deleting the source the user needs to retry), would ship silently.

This is the P0 "seam the money path" item of #1009.

What changed

• Extract runIngestionRun from runDataIngest and route its four cluster-touching steps through package seams — mintIngestorTokenFn / portForwardJobsManagerFn / submitRunFn / cleanStagingFn — mirroring the existing listDatasetsFn seam. The extraction is behavior-identical; runDataIngest just calls it and threads jsonEmitted back for the --output-json error defer.
• Name the reclaim gate — shouldReclaimStaging(status) — the "reclaim ONLY on a clean success" invariant. A detached / partial / failed / errored run keeps the staged source (the Job may still be reading it, or the user wants to retry/inspect). This is the "must NOT reclaim on partial failure" gate #1009 calls out.
• Harden submit.ForwardedConnection.Close to no-op on a zero-value connection (nil stopCh) — it panicked on close(nil) before, which is a latent bug and blocked using a fake in the seam test.

Tests (no cluster needed)

• TestRunIngestionRun_Matrix drives the whole tail through the seams and asserts, per outcome, three things in lockstep: the exit code (5 auth / 8 submit / 9 watch+ingest / 0 success+detached), whether the staging reclaim ran (only on succeeded), and the emitted --output-json status. Includes the mint-fail (5) and port-forward-fail (8) pre-submit returns.
• TestShouldReclaimStaging pins the gate in isolation.
• Extends the existing TestClassifyPushOutcome to cover the exit-5 (auth) and exit-8 (submit) buckets it was missing, plus unknown / nil-result.
• TestSeamsWiredToRealFns guards that no seam is left nil.

Coverage: internal/cli 70.1% → 72.2%; runIngestionRun 94.1%, shouldReclaimStaging 100%, classifyPushOutcome 92.9%. Full suite green.

Type

Test hardening + a latent-panic fix · cli · RFC-0002 foundation. Part of backend#1009 (P0 seam the money path) — independent of the value-level-parity PR (cli#186) and of data-ingestors #340.

Remaining #1009: CI drift wiring (pin the goldens/schema SHA + a per-package coverage floor so this can't rot), the cross-repo taxonomy contract test, one content-compared ingest e2e.

---

Note

Low Risk
Refactor is extraction plus test seams with stated behavior parity; the only production behavior change is the defensive port-forward Close fix.

Overview
Extracts the post-staging ingest tail (token mint → jobs-manager port-forward → submit/watch → classify → --output-json → staging reclaim) into runIngestionRun, with behavior unchanged for runDataIngest aside from threading jsonEmitted back for the early-failure JSON defer.

Cluster-touching steps go through package-level seams (mintIngestorTokenFn, portForwardJobsManagerFn, submitRunFn, cleanStagingFn) so tests can drive the full exit code / JSON status / reclaim matrix without a cluster. The reclaim-only-on-clean-success rule is named shouldReclaimStaging.

submit.ForwardedConnection.Close now no-ops when stopCh is nil (zero-value / test fakes), fixing a latent close(nil) panic.

Adds ingestion_run_test.go (TestRunIngestionRun_Matrix, TestShouldReclaimStaging, TestSeamsWiredToRealFns) and extends TestClassifyPushOutcome for auth (5), submit (8), unknown, and nil-result cases.

^{Reviewed by Cursor Bugbot for commit 0c39dc7. Bugbot is set up for automated code reviews on this repo. Configure here.}

[LukasWodka]

bugbot run

[cursor]

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit 0c39dc7. Configure here.}