Optimize the duplicated-items column to be ~O(n) instead of O(n^2)

glideapps / glide-code-columns

Custom computed columns for Glide

https://glideapps.com/plugins

Apache License 2.0

22 stars 22 forks source link

Closed djsweet closed 2 years ago

vercel[bot] commented 2 years ago

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Updated
glide-code-columns	✅ Ready (Inspect)	Visit Preview	Oct 22, 2022 at 1:08PM (UTC)

djsweet commented 2 years ago

It seems that push is definitely faster than unshift. So I'll use that instead.
The thought was that this was a pretty cheap way to preserve the previous behavior of keeping the order based on first element, but you're right, that's not always true. It worked for ["a", "b", "a", "b", "c", "b"] but not ["a", "b", "a", "b", "c", "b", "a"].

So I'll keep indexes in a separate map, and use those indexes to sort the output according to the first-detected order.