Sample
Handle middleware stack caching and isolation traps when duplicating Faraday connections across requests or tasks
sha256:ba99fb2bf3dcbb90f9cfeb92673322d9bb85a7bdef5493050f6dd3a779abd62d
PUBLISHED
L3_CONTRACT_PASS
MIT-0
Case
- Goal
- Handle middleware stack caching and isolation traps when duplicating Faraday connections across requests or tasks HOW
- Packages
- faraday 2.14.3
- Environment
- ruby
- Created
- 2026-08-17T02:39:03Z
Commonly assumed
Duplicating an existing Faraday connection with dup creates an unlocked connection that executes subsequently added middleware on its requests.
The sample's author recorded this as what a developer or model would expect here. The contract below is what actually ran.
Contract
- Adding middleware to a duplicated Faraday connection after the original connection has executed a request silently ignores the new middleware during request execution because dup preserves the parent connection's pre-compiled rack app instance.
- assert modifying the middleware stack of a connection after executing a request raises Faraday::RackBuilder::StackLocked
- assert duplicating a connection before any requests are executed successfully compiles and runs newly appended middleware
- assert headers mutated on a duplicated connection are isolated from the parent connection whereas mutating headers on a shared connection affects all callers
- assert nested middleware request callbacks execute in insertion order while response on_complete hooks unwind in reverse order
Files
- Gemfile
- Gemfile.lock
- NOTES.md
- csx.json
- src/connection_factory.rb
- 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 · 2026-08-17 · ed25519:d91480838ac982c9