Conversation
…v6 into PM-5015_ai-only-review
PM-5015 ai only review
…v6 into PM-5203_approval-phase-ai-only
…ding ai decissiosn
What was broken Marathon Match challenge closing could persist winners before all parallel system scoring callbacks had written final summations for every member's latest submission. If an early final summation belonged to a lower-scoring member, that member could be stored as the winner even after later final scores showed a different winner. Root cause (if identifiable) closeMarathonMatch ranked whatever final review summations existed at the time of the close request. It did not verify that every latest member submission had a final/system summation, and it did not limit winner ranking to the latest submission for each member. What was changed Loaded latest non-checkpoint submissions from the review database during Marathon Match close. Winner selection now requires each latest submission to have a matching final summation before the challenge can be completed, and ranks only those latest-submission final summations. When submission rows are unavailable, the logic falls back to the newest final summation per submitter to avoid duplicate placements. Any added/updated tests Added ChallengeService unit coverage for ranking winners from latest member final summations and blocking close while any latest submission is missing a final summation.
…-ai-only PM-5203 approval phase ai only
PM-5161: Guard MM winner selection until final scoring completes
…-ai-only Allow opening approval phase without approval user in AI Only challenges
What was broken Marathon Match close could still persist winners from an incomplete set of final summations when latest submission rows were unavailable to challenge-api. In that fallback path, an older or partial final score could be selected before another submitter's latest known summation had a final score. Root cause (if identifiable) The previous fix guarded the direct latest-submission path, but the no-submission-row fallback still ranked the newest final summation per submitter without checking whether Review API already showed a newer submission-scoped summation for that submitter. What was changed The fallback winner selection now derives the newest submission-scoped summation per submitter from all review summations. If that latest known submission does not have the selected final summation, closeMarathonMatch rejects the close instead of persisting premature winners. Any added/updated tests Added ChallengeService unit coverage for blocking fallback winner selection when a submitter has an older final summation and a newer non-final submission-scoped summation.
PM-5161: Guard MM winner fallback final scoring
Whitelist changes for M2M tokens
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
3 participants
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.
https://topcoder.atlassian.net/browse/PM-5015
This pull request introduces several important updates to support AI-only challenge flows, improve challenge phase handling, and enhance deployment and licensing. The main changes include the introduction of a new AI Review phase and an AI Only Challenge timeline template, logic to identify "stalled" challenges, updates to deployment scripts and CI/CD, and a new open source license.
AI Challenge Flow Enhancements:
config/default.js) to include the ID for the AI Only Challenge timeline template, making it accessible via environment variable or default.Challenge Status Improvements:
stalledproperty to the challenge API response and updating the OpenAPI schema documentation. [1] [2] [3] [4]Deployment and CI/CD:
deploy:devscript topackage.jsonfor tagging and pushing dev builds, streamlining the development deployment workflow..circleci/config.ymlto trigger builds only for tags matching thedev-*pattern, improving CI/CD efficiency for development tags.Access Control and Security:
Licensing:
Apache License 2.0to the repository, clarifying the project's open source license and terms of use.