Sample
Handle thread-local declaration state, failed sig retry poisoning, and reentrant invocations in sorbet-runtime
sha256:92626ce19dbeb23bcf39bacfb6fa5bae6bb0a0e9104b1417876a2f77b4b1f928
PUBLISHED
L3_CONTRACT_PASS
MIT-0
Case
- Goal
- Handle thread-local declaration state, failed sig retry poisoning, and reentrant invocations in sorbet-runtime HOW
- Packages
- sorbet-runtime 0.6.13427
- Environment
- ruby
- Created
- 2026-08-16T17:12:51Z
Commonly assumed
A sig block failing on first invocation can be retried once its dependencies resolve, sig binds to the next lexically defined method across threads, and re-entering a method during sig evaluation executes the underlying method body.
The sample's author recorded this as what a developer or model would expect here. The contract below is what actually ran.
Contract
- Retrying a method whose lazy sig block previously raised an exception fails with RuntimeError: 'A previous invocation... raised, and the current one succeeded' rather than recovering when the sig block now succeeds.
- Declaring sig in one thread and defining the method in another leaves the method completely untyped while leaving a dangling active_declaration that breaks subsequent sig calls in the declaring thread.
- Re-entering a method from inside its own sig block before the first-call wrapper completes raises SystemStackError due to recursive first-call wrapper re-entry.
- Calling sig twice in the same thread without an intervening method definition immediately raises RuntimeError and resets the pending declaration.
Files
- Gemfile
- Gemfile.lock
- NOTES.md
- csx.json
- test/contract.rb
Download the verified artifact (tar.gz) — the exact bytes the contract ran against
Origin Seeder
Verification receipts
- ruby 3 · CONTAINER_RUN · compile:SKIPPED · contract:PASS · load:PASS · resolve:PASS · rubygems@1 · 2026-08-16 · ed25519:d91480838ac982c9