Skip to content

TASKS.md

Latest commit

 

History

History
51 lines (39 loc) · 3.35 KB

TASKS.md

File metadata and controls

51 lines (39 loc) · 3.35 KB
Raw

Tasks

Active work only. Historical completed work lives in notes/AGENT_NOTES.md and AGENTS.md.

  • Recently completed: Task 42 landed the boxed contract runtime and Schema<'T> surface. Keep the historical rationale, benchmark notes, and follow-up findings in notes/AGENT_NOTES.md and docs/PROFILE-REPORT-ANALYSIS-INSTRUCTIONAL.md.

  • Task 50: Run a narrow post-Task-42 performance follow-up

    • Treat the current numbers as two separate problems: typed record decode overhead and handwritten parser hot loops.
    • Stay disciplined: no broad runtime rewrite until a narrower experiment wins clearly on the published scenario set.
    • First pass:
      • generalize the typed JSON record-decode lane beyond the benchmark-only hand-written shapes
      • profile string-heavy decode and serialize hotspots again after each step
      • identify whether the next worthwhile work is parser scanning, string handling, unknown-field skipping, or record assembly
    • Output:
      • refresh the benchmark notes with before/after numbers
      • either promote one proven optimization direction into production work or explicitly close the line of investigation

  • Task 49: Review and improve the new DSL for DX

    • Review the new Schema.* surface for compactness, clarity, and maintainability after the immediate perf follow-up is settled.
    • Capture improvements in PLAN-TO-IMPROVE-DSL.
    • Do not fold speculative API cleanup into performance work unless it materially simplifies a measured hot path.

  • Task 37: Add structured decode error outputs for app boundaries

    • Provide a structured error model that callers can use for REST responses, startup config failures, and message rejection logs.
    • Preserve the existing path-aware detail across JSON, XML, YAML, and KeyValue.
    • Keep the fast path cheap when callers still want exception-based failures.

  • Task 40: Add explicit unknown-field policy controls

    • Support contract-level decisions for rejecting, allowing, or collecting unknown fields during decode.
    • Keep the default behavior conservative and aligned with explicit authored contracts.

  • Task 18: Add code generation modes for contracts

    • Generate checked-in F# or C# contract code from CLR models, deterministic JSON examples, or imported contract inputs.
    • Keep generators in separate .NET-only projects so reflection-heavy tooling does not leak into the AOT/Fable-safe core.

  • Task 43: Improve the C# authoring story

    • Expand the C# contract authoring DSL.
    • Add opt-in source generation layered on top of the stable runtime DSL.

  • Task 44: Add OpenAPI and schema-ecosystem guidance

    • Document how exported JSON Schema can feed external OpenAPI or validation toolchains.
    • Evaluate whether a narrow OpenAPI bridge belongs in an opt-in package.

  • Task 45: Strengthen external-schema import and lowering

    • Improve guidance and support for contracts that start from JSON Schema rather than authored Schema<'T>.

  • Task 46: Publish a JSON-first onboarding path

    • Add a short path for developers who only care about JSON first.

  • Task 47: Expand contract-pattern coverage

    • Add polished patterns for string enums, message envelopes, PATCH/update DTOs, event payloads, and layered config sections.