Repository navigation
Qualify current-version restore and collection costs - #116
Merged
Merged
Conversation
added 9 commits
October 20, 2026 17:57
# Conflicts: # docs/plans/lab-workflow-reliability/execution-ledger.md # docs/plans/lab-workflow-reliability/plan.md
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars. ⚙️ Run configuration
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
MrScripty
marked this pull request as ready for review
October 20, 2026 23:26
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This follow-up qualifies current-version coordinator, timing and publication backup/restore integrity, and measures the current collection facade. It updates the existing plan and acceptance criteria to the user's revised support policy.
Old projects, clients and catalog layouts have no backward-compatibility guarantee. Owners must perform their own one-time migration or adaptation to the formats required by the new Tuldok version. Intentional breaking changes do not require old-client adapters or an automatic migration system. Current-version correctness, schema validation, honest unsupported-version errors, scientific meaning, custody/accounting and backup/restore integrity remain required. This policy amendment introduces no new schema number or wire format.
PR115 merged first as
d33145170250445e630c8378681b0a81f7c8dce2after all four applicable checks passed. This follow-up changes tests, existing plan/acceptance documentation and scoped evidence. Application, schema, README and workflow bytes remain unchanged.Validation and evidence:
00d32f8: 36 pairs, 72 arms and 12 fresh processes. All 27 twenty-item pairs were faster with batch collection; all nine single-item controls were slower. Raw order, resource, HTTP, owner and cleanup observations are retained. Individual collection stays the default; no automatic threshold or general speed claim is introduced.cd420c44neutral HTTP/CLI qualification: 12 cases passed in 97.352 seconds, with zero failures, errors or skips. Optional classifier comparison was explicitly excluded locally.9d699cc154e9dde89fe26df16ad57c421e68d734, tree9f6850b84da60c4463b7bb98e0c22790353fa4b9. Receipt SHA256:7bde9c541af2bf9cd7d2a87e02f319ad74649a0bc655dcd774079a0060df9a06.The completed pinned old-application witness is preserved as historical, non-gating evidence. Its newly added helper and invocation were removed from active qualification; the retained-run qualifier is byte-identical to its pre-existing
cd420c44form. The superseded lifecycle run was cancelled after its old-version checkpoint; that cancellation is not counted as a pass. Existing historical regression tests are not expanded.The scope update and revised plan supersede compatibility promises in historical reports.
A hosted raw-log download returned Forbidden; run/check/step metadata does not establish unobserved test counts or zero skips. This does not claim full C1/T1/M1/T5/T8, hardware or science acceptance. PUB-1 still needs the original real failure, custody, profile and output inventory. SCI-1 still needs a real comparison pair, criterion authority and environment/procedure.