Repository navigation
main's provider-set hash is a day and two merges out of date - #3321
Merged
Merged
Conversation
The guard has been red on main since 2026-10-27 and it was right to be. Two merges moved a bench's provider set and neither touched a cache key or this hash: #3311 stopped scoring Mobula on Solana in token-quote-coverage, and #3312 added Perpex to hyperliquid-frontends. Both changed only benchmarks/*.yml, which is precisely the shape this test was written to catch, since a provider set can move without anyone editing the code that caches it. Red CI on main greys out every PR opened against it, so this unblocks the branch as much as it records anything. The four keys are NOT bumped, and that is a decision about timing rather than a weakening of the rule. A bump invalidates entries written before the change. By today the cache has already turned over on its own, which I measured instead of assuming: production serves 106 providers on hyperliquid-frontends with Perpex among them, and /products/perpex answers 200 carrying its hyperliquid-frontends appearance, which is the appearance providers-vP exists to protect. Bumping now would flush healthy entries and reveal nothing. Caught on the day, the keys would have had to move, and for the next change they still do. The comment above the hash says all of that, so the next reader does not take this as permission to skip a bump. Recorded hash moves f355324b6640e7e4 to b767efc1bf40bf99 against 1841 providers. bun test 786 pass 0 fail, tsc clean, eslint clean.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
src/lib/spec-provider-set.test.tshas failed onmainsince 2026-10-27, which greys out every PR opened against it, including #3319.The guard was right. Two merges moved a bench's provider set and neither touched a cache key or the recorded hash:
token-quote-coverage, removing a provider from a sethyperliquid-frontends, adding oneBoth changed only
benchmarks/*.yml, which is exactly the shape this test exists to catch: a provider set can move without anyone touching the code that caches it.Why the four keys are not bumped
This is a decision about timing, not a softening of the rule. A key bump invalidates entries written before the change. By today the cache has already turned over on its own, and I measured that rather than assuming it:
hyperliquid-frontends, Perpex among them/products/perpexanswers 200 and carries itshyperliquid-frontendsappearance, which is the appearanceproviders-vPexists to protectSo bumping now would flush healthy entries and surface nothing. Caught on the day, the keys would have had to move, and for the next change they still do. The comment above the hash spells that out so the next reader does not take it as permission to skip a bump.
f355324b6640e7e4tob767efc1bf40bf99, against 1841 providers.Checks
bun test786 pass 0 fail,tsc --noEmitclean,eslintclean on the touched file.