Newest first, grouped by year. Search or filter by topic.
Inflow's monolith failed startup when two modules disagreed about the shape of a shared event. The microservices branch deletes that check from exactly the boundary that became remote - and it turns out the deletion was not optional.
Inflow's boot-time contract verifier is excellent and almost unused. A census of every consumer-side type copy in the estate shows four protected and fourteen not - and the two failure modes are a startup crash and a silently null property.
Inflow lets each module own a private copy of every event it consumes, then type-checks those copies against the producer at application start. It is consumer-driven contract testing with no broker, no schema registry and about 280 lines - and the check is weaker than the ceremony suggests.
The Trill monolith ships a startup-time validator that compares independently declared copies of the same message and crashes the app on a mismatch - the most original idea in either build, adopted by four declarations out of twenty-two.
The last commit of ModularMonolith renames Name to ConferenceName in the subscriber's copy of its one integration event - and the JSON translator has delivered null into that field ever since. Anatomy of a silent contract break.
Orders and Parcels keep the estate's only cross-service contract honest with Pactify - a pact written to a relative path six directories above the repository, verified against a seeded Huge Weapon, and checked for shape only. Consumer-driven contracts with no broker, read from source.
The estate's full who-publishes-what matrix, built from the source - and the one broken cell in it. Orders' copied ParcelDeleted event declares the deliveries exchange while Parcels publishes on parcels, so Orders never learns a parcel was deleted, and neither the Pact tests nor the hand-maintained registry can catch it.
Pactify.AspNetCore in one sitting: an attribute, an assembly scan, and a /pacts/{provider} endpoint that serves your contract classes as live JSON - plus what a runtime contracts endpoint is really for.
The series retrospective - what a thirty-file library taught me that a full-featured one couldn't, Pactify's honest limit list versus PactNet, the broker maturity ladder, and the org-shaped question hiding under the tooling.
A line-by-line read of Pactify's matching engine - top-level-only body comparison, header names that ignore your casing option, nested objects that can never match, and two format-string bugs that crash the error path.
How a pact travels from the consumer's pipeline to the provider's - Pactify's file and HTTP publishers, Pact Broker URL anatomy, and the unchecked response status that makes a failed publish silently pass your build.
The provider side of Pactify - how CreateFor<Startup> hosts your whole API in memory with TestServer, why interactions verify concurrently, the SmartFormat trick behind endpoint templating, and the discovery that Given(...) never runs.
An annotated tour of the JSON document Pactify produces - the serializer settings that shape it, the snake_case field inherited from Pact v1, and the unusual decision to ship the consumer's matching options inside the contract itself.
A walk through Pactify's PactMaker fluent API - the two-interface builder trick that hides Build() from you, why the consumer test always passes, and the uninitialized-object surprise inside WithBody<T>().
Unit tests mock what you assume, end-to-end tests tell you too late - consumer-driven contract testing closes the gap, and a small open-source .NET library called Pactify is the clearest way I know to learn how it works. Part 1 of a series.
