Blog

Use const-valued properties as discriminators in a oneOf union to implement the same polymorphic dispatch pattern used by System.Text.Json, illustrated with JSON Patch operations.

JSON Schema's oneOf keyword creates exhaustive discriminated unions, solving the two biggest problems with C# inheritance-based unions: invasiveness and missing-case bugs.

Compose multiple independent JSON Schemas into a single type using allOf — the schema equivalent of implementing multiple interfaces, with Corvus.JsonSchema.

JSON Schema's prefixItems keyword maps to C# ValueTuples. Combined with unevaluatedItems: false, it creates fixed-length, mixed-type arrays with strongly typed accessors.

Fixed-size numeric arrays defined in JSON Schema can be converted to and from Span<T> for use with System.Numerics.Tensors, enabling zero-allocation interop for ML and math workloads.

Model multi-dimensional arrays (rank 2+) in JSON Schema by nesting array definitions with items. Use minItems/maxItems to fix dimensions or allow ragged arrays.

Define typed arrays in JSON Schema using items, minItems, and maxItems. Corvus.JsonSchema generates strongly typed accessors, enumerators, and IEnumerable support for LINQ.

Narrow the validation rules of an existing JSON Schema type by adding tighter constraints on its properties. Unlike OO overrides, constraints compose: both base and derived rules apply.

Extend an open JSON Schema type with new properties using $ref — similar to deriving from a base class in C#. Corvus.JsonSchema combines both schemas into one type.

JSON Schema objects are open by default, allowing extra properties for forwards compatibility — use unevaluatedProperties to close them for strict validation.

Use $ref and $defs to extract shared schema definitions and reuse them across properties. Corvus.JsonSchema generates a single shared .NET type, eliminating duplication.

Add required fields, string length limits, and numeric range constraints to a JSON Schema data object — and see how Corvus.JsonSchema enforces them in C#.

Define a simple data object in JSON Schema and generate strongly typed C# with Corvus.JsonSchema — properties, implicit conversions, and fast serialization.

Corvus.JsonSchema enables safe use of the very high performance JSON parsing offered by .NET's System.Text.Json.

Our experience shows that the the most successful data projects rely heavily on building a multi-disciplinary team.

Build modern interactive web apps with ASP.NET Core, Razor, and HTMX — bridging HTMX hypermedia to Chart.js JSON APIs using HATEOAS-style stateful links.

Architectural Decision Records (ADRs) capture context, options, decisions, and consequences. dotnet-adr is a .NET tool for managing ADRs.

As data and AI become the engine of business change, we need to learn the lessons of the past to avoid expensive failures.

This is a summary of the sessions I attended at SQLbits 2024 - Europe's largest expert led data conference. This year SQLBits was hosted at Farnborough IECC, Hampshire.

Explore a step-by-step guide on designing a side nav in Power BI, covering form, icons, states, actions, with a view to enhancing report design & UI.

Learn how to access multi-select choice column choice labels from Azure Synapse Link for Dataverse using PySpark or SQL.

The first step on the road to implementing observability in your Python notebooks is basic logging. In this post, we look at how you can use Python's built in logging inside a Synapse notebook.

No-code/Low-code democratizes software development with little to no coding skills needed. But how do you evaluate if software DIY is the right choice for you?

.NET's System.Text.Json offers very high performance JSON parsing. However, JsonElement's text parsing methods offer a trap for the unwary

This post walks through the process of using PerfView to analyse ETW data to get to the bottom of a performance issue in a WPF application.