Altinity / Altinity/altinity-sql-browser

Live ClickHouse compatibility test matrix for declared-parameter types

Abierto
#243 0 comentarios 0 reacciones 0 asignados Ver en GitHub

Nadie ha tomado este issue todavía.

Lenguaje dominante
TypeScript
Estrellas
8
Forks
2
Merge medio
1 h 34 min
PR fusionados (30 d)
6

Descripción

Problem

#238 (unify the two ClickHouse type parsers, make LowCardinality(T) transparent for parameter validation/serialization) explicitly requires live compatibility tests against real ClickHouse servers:

Run against every supported ClickHouse version using the real HTTP parameter path... Record whether each declaration is accepted by supported server versions. The compatibility test is authoritative for parameter behavior.

This was deferred out of the #238 implementing PR because there's no existing pattern in this repo for hitting a real ClickHouse server from the unit test suite (Playwright e2e is browser-only, against static fixtures/happy-dom).

What's already known (live-verified manually during #238/#241, not automated)

Verified against otel (ClickHouse 26.3.13.20001, Altinity build) via the deployed SPA:

  • LowCardinality(String), LowCardinality(UInt32), Array(LowCardinality(UInt64)) — accepted, serialize/execute correctly end-to-end (including a UInt64 value near max, preserved exactly).
  • LowCardinality(Enum8(...))rejected by the server: Code: 43. DB::Exception: DataTypeLowCardinality is supported only for numbers, strings, Date or DateTime, but got Enum8(...). (ILLEGAL_TYPE_OF_ARGUMENT). This is now correctly reflected client-side (#241): the parser treats LowCardinality wrapping an Enum8/Enum16, in any nesting order, as invalid — no Enum dropdown/membership behavior, degrades to opaque passthrough.

Goal

Build the actual live-compatibility-test scaffolding: a way to run a small matrix of declared-parameter type strings (the ones listed in #238's "Live compatibility tests" section, plus whatever else the LowCardinality-transparency work depends on) against every supported ClickHouse version via the real HTTP param_* path, and record accepted/rejected per version. Use the result to:

  • confirm (or correct) which type/wrapper combinations the client should treat as supported vs. permissively-invalid across the whole support matrix, not just one server version;
  • catch any other combination (beyond the now-known LowCardinality(Enum...)) where client-side "looks supported" diverges from server-side "actually accepted."

Non-goals

  • Building a general live-ClickHouse CI harness for the whole test suite — this is scoped to the parameter-type compatibility matrix only.
  • Changing any already-shipped #238/#241 client behavior unless this work finds a real divergence.

Files

Likely a new tests/live/ (or similar, TBD) location — no existing pattern to follow, per #238's PR discussion; this issue owns designing where it lives.

Guía de contribución

Abrir la guía de contribución

Primeros pasos

  1. Lee el issue completo y luego la guía de contribución del proyecto.
  2. Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
  3. Haz un fork del repositorio y trabaja en una rama.
  4. Abre un pull request que haga referencia al número del issue.

Línea de trabajo

Lee primero la sección “Live compatibility tests” de #238 y el comportamiento relacionado de #241; después, inspecciona la configuración de pruebas existente para determinar cómo se puede ejercitar la ruta HTTP param_* real. Diseña la ubicación tests/live/ u otra ubicación adecuada y ejecuta la matriz de tipos declarados contra cada versión de ClickHouse compatible. La tarea estará completada cuando se registren los resultados aceptados o rechazados por versión y se identifiquen las divergencias entre cliente y servidor.

Escrito por el modelo de indexación a partir del texto del issue.

Evaluación

Stack tecnológico
typescript
Área
databases, testing-qa
Tipo de issue
Nueva funcionalidad
Dificultad
5/5
Tiempo estimado
Más de una semana
Estado de actividad
Tranquilo
Claridad
Necesita aclaración
Aptitud para principiantes
35/100

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.