Vanillas
JavaScript utility functions. Comparable to Lodash or Ramda, but faster. In some cases it even outperforms native implementations (or leverages native implementations that are messy to type out).
Why Another Utils Lib?
Lodash/Underscore created an API many years ago which is unfavorable to Currying. For currying to be possible you need the final argument to your function to be the data on which you operate (rather than the first arg). And although Lodash/Underscore has an "fp" (functional programming) subset, the fundamental flaw in the argument order can't be corrected.
A similar utils library Ramda was created as a classic functional programming library, and the most notable change was to reorder the arguments to be favorable to currying. Ramda took it a step further than than (perhaps too far) of making every function in the whole library curried. This isn't always desired. But the more critical flaw with Ramda is it is slow. Very slow.
Vanillas was created to correct those perceived flaws in both Lodash/Underscore and Ramda. It outperforms those libraries, makes currying possible, yet allows you to opt-in to currying.
Docs
View the full docs. These are generated from - and always in-sync with - the JsDoc code annotations.
Curried vs Non-Curried
Importing from the root vanillas namespace yields all the non-curried utility functions (including the curry() function itself if you want to do your own). To use the curried versions of the Vanillas utility functions though, just import from the vanillas/curried directory rather than from the root.
Installation
npm install vanillas
Benchmarks
Run any of the files in the benchmark/test/ directory.
Examples (run one or more benchmark tests, space-delimited):
npm run benchmark compose
npm run benchmark propAt
npm run benchmark mapObject
npm run benchmark compose mapObject propAt
Or you can run all of them (grab a cup of coffee while you wait):
npm run benchmark