Why I built this: I have noticed in several codebases the tendency to keep writing the same or similar filter functions over and over again. Often inlined, taxing readability. And realized that most of them do pretty much the same N things. Therefore they could be isolated, curated, and maintained separately.
fp-filters provides 130+ composable, point-free filtering functions that make data transformation pipelines more readable and maintainable. All functions are predicates and can be used as such, but they shine when used as filters.
Key features:
1. Pure functional approach with no side effects 2. Composable predicates (and, or, not combinators through fp-booleans) 3. Common filter patterns as reusable functions 4. Type safety with TypeScript definitions 5. Exported as both esm and cjs 6. Zero dependencies, tiny footprint 7. Individual exports. Tree shaking is not even needed. 8. 100% test coverage
Example usage:
const input = [[1, 2, 3], [2, 4], [0, 4, 8, 16]]; // JS input.filter((array) => array.every((element) => element % 2 === 0)); // fp-filters input.filter(everyElement(isEven))
// JS dates.filter((date) => { const day = date.getDay(); return day === 0 || day === 6; }); // fp-filters dates.filter(isWeekend);
// JS array.filter((arg) => arg.length > 0); // fp-filters array.filter(isNotEmpty);
What makes this different: Unlike lodash/fp or ramda, fp-filters focuses specifically on array filtering patterns with an emphasis on readability in complex data transformations. It's designed to be intuitive for developers already familiar with JavaScript's array methods while providing the benefits of functional composition. The project is MIT licensed and open to contributions. I'd love to hear feedback from the HN community. Links:
npm: https://www.npmjs.com/package/fp-filters GitHub: https://github.com/Oaxoa/fp-filters Documentation: https://oaxoa.github.io/fp-filters
Thanks for checking it out
spacemankiller•9mo ago