valimock v1.6.0

• #77 (minor) Thanks @Saeris! - Add opt-in instrumentation hook (new Valimock({ instrument: true })) and a perf-regression spec asserting call-count and depth ceilings on realistic-shape schemas. Catches catastrophic regressions like the v1.5.0 wrapper-recursion bug — proven locally to detect the bug at ~2,100x explosion in call counts.
• #79 (patch) Thanks @Saeris! - Add Vitest bench() files for iteration-time performance evaluation, reusing the fixture library from the regression canary. Includes a scripts/profile.mjs convenience wrapper for CPU profiling. Pure tooling — no public-API or runtime changes.

Published to

• ✅ npm

valimock v1.5.4

• #74 (patch) Thanks @Saeris! - Dispatch and lookup optimizations: Map-based schema dispatch in the hot #mock path, precomputed lowercase keyNameGenerators lookup, WeakMap memoization of #getValidEnumValues, and a fast path through generateString for unconstrained strings. ~1.2-1.7x faster on string- and enum-heavy schemas.
• #75 (patch) Thanks @Saeris! - Cache findFakerForKeyName auto-discovery results in a module-level WeakMap keyed by Faker instance. Both positive and negative results are cached, so repeat lookups of the same keyName (cached hit) and repeat lookups of unknown keyNames (cached miss) both skip the full Object.keys(faker) walk. ~1.8-2.1x faster on string-heavy schemas.

Published to

• ✅ npm

valimock v1.5.3

• #72 (patch) Thanks @Saeris! - Lazify wrapper handlers (nullable / nullish / optional / undefinedable) so they descend into the wrapped schema only when the random roll picks the wrapped branch. Eliminates a ~4,700x slowdown on self-referencing schemas wrapped in nullish-lazy (the canonical referencedMessage: v.nullish(v.lazy(() => self)) pattern).

Published to

• ✅ npm

valimock v1.5.2

• #69 (patch) Thanks @Saeris! - Clean up package.json for ESM-only standards compliance: drop redundant main/module/types top-level fields and the non-standard module condition inside exports. The exports.import branch (with types listed first per Node's resolver requirement) is the canonical entrypoint for Node >=22 and modern bundlers, which is what this package already requires via engines.node. Normalizes engines.node from ">=22.x" to the canonical semver ">=22"; no behavior change (the two ranges are semver-equivalent), but the new form is idiomatic and what node-semver outputs after parsing.
• #70 (patch) Thanks @Saeris! - Fix intersect dropping variant discriminator when shared keys diverge, and let customMocks override built-in schema handlers (notably customMocks.any: () => undefined to restore pre-1.5 v.any behavior).

Published to

• ✅ npm

valimock v1.5.1

• (patch) - Fix broken export references in package.json

Published to

• ✅ npm

valimock v1.5.0

• #66 (minor) Thanks @Saeris! - Valibot 1.4 compatibility: warn on unhandled custom schemas instead of silent undefined, expand tested API surface (instance, custom, seed, throwOnUnknownType, lazyAsync), and finish the structural cleanup pass started in 1.4.0.

Published to

• ✅ npm

v1.4.0

Minor Changes

• eb2553c Thanks @Saeris! - Add basic support for required() and omit()

v1.3.1

Patch Changes

• 50dfba1 Thanks @Saeris! - Enhance test resiliancy for string() schemas, Retry by default for possible falsy values

v1.3.0

Minor Changes

• 4170123 Thanks @Saeris! - Add variant and exactOptional schema support, Improve test reliability

v1.2.0

Minor Changes

• 2f875f1 Thanks @Saeris! - Add support for new String validations and Keyword matchers

  New Validations:

  • base64
  • bic
  • creditCard
  • decimal
  • empty
  • hexadecimal
  • hexColor
  • mac
  • nanoid (unstable)
  • octal

  New keywords:

  • username
  • displayName
  • firstName
  • middleName
  • lastName
  • fullName
  • gender
  • sex
  • zodiacSign
  • isbn
  • iban
  • vin
  • vrm

  Also adds support for empty() for both Strings and Arrays