fix: resolve zero-settle rail_key and signer_network from the bound credential (2.5.2) - #82
Merged
Merged
Conversation
…redential No receipt is minted on the $0 carve-out path, so rail_key fell back to the primary MPP rail key and mislabeled Solana zero-settles as Tempo (and vice versa when rails were declared solana-first). zero_amount_carve_out already recovers the signer network from the credential's did:pkh source; map it through _rails_key_for_mppx_method so SettleOutcome.rail_key reflects the rail that was actually bound. The MPP carve-out also hardcoded signer_network to evm, mislabeling Solana signers; surface the carve-out's recovered network on both branches. Bump to 2.5.2. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
lefthook 2.1.10, uvicorn 0.51.0. ty is held at 0.0.56: 0.0.57 newly rejects starlette Headers against the HeadersLike protocol (aip/request.py), a checker-side strictness change to revisit on the next ty release. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
vvillait88
added a commit
that referenced
this pull request
Jul 17, 2026
…l 402 fix (#89) In-range dependency upgrades (uv.lock only; no pyproject constraint changes). fastapi, anyio, asgiref, solders, solana, stripe, sentry-sdk, typer, httpx2/httpcore2, ruff, etc. ty stays pinned at 0.0.56 (deliberate hold; newer ty adds churn-prone diagnostics). Verified locally: ruff check, ruff format --check, ty check (clean), vulture, pytest (1834 passed, 95.49% coverage). 🤖 Generated with [Claude Code](https://claude.com/claude-code) --- **Also unpins `ty`** (`==0.0.56` -> `>=0.0.32`) to match the rest of the python fleet. ty 0.0.60 tightened protocol-assignability checking; starlette 1.3.1 `Headers.get` no longer matched the `HeadersLike` protocol, so `HeadersLike.get` is narrowed to its actual single-arg usage (`get(key, /) -> str | None`). Full gate green at ty 0.0.60: ruff, format, ty, vulture, pytest (1834 passed, 95.49%). --- **Also: malformed-credential 402 fix (parity with node-commerce #82).** Stripe/MPP-team store feedback: a junk or rejected payment credential on POST /purchase returned 400, which x402/MPP clients abort on. Now returns a 402 with a fresh challenge, both for the malformed-shape pre-check (emit a fresh discovery 402, stripping the junk credential so minting/compose take the fresh path; pre_validate + gate/assess stay skipped) and the MPP compose-reject path (return the challenge pympp already emitted as 402). x402 settle-verify failures (forged payTo) still 400. Rides this 2.5.7 release (no extra bump). Gate green: ruff/format/ty/vulture/pytest (1834 passed, 95.49%). --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
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
1 participant
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.
What
Two fixes to the $0 zero-settle carve-out in
Checkout, plus the 2.5.2 version bump and a dev-dep refresh.rail_key resolves from the bound credential. No receipt is minted on the $0 path, so
SettleOutcome.rail_keyfell back to the primary MPP rail key and mislabeled Solana zero-settles as Tempo (and Tempo as Solana when rails were declared solana-first).zero_amount_carve_outalready recovers the signer network from the credential'sdid:pkhsource; the carve-out now maps it through_rails_key_for_mppx_methodso rail_key reflects the rail that was actually bound. Downstream consumers that dispatch on rail_key (network_for_outcome,simulate_deposit_for_outcome) get the correct network as a result.signer_network is no longer hardcoded. The MPP carve-out stamped
signer_network="evm"whenever a signer was recovered, mislabeling Solana signers; both carve-out branches now surface the network the credential actually carries.Dev deps: lefthook 2.1.10, uvicorn 0.51.0. ty is held at 0.0.56; 0.0.57 newly rejects starlette Headers against the HeadersLike protocol, to revisit on the next ty release.
Tests
🤖 Generated with Claude Code