If you look at the code, it's because for whatever reason, the emoji loader code creates an array that maps tags to emojis... but for some reason, they decided to implement this by storing an array of tags and then binary searching that array for EVERY TAG that must be updated -- consider that each of the 5,000 some emojis have multiple tags! Plus, they create multiple DOM elements for each emoji! This code has O(m * n * log(n)) complexity, where m is the average number of tags per emoji and n is the number of emojis. This code could very simply be sped up by using a hashmap (which is builtin with JS's Map) or by doing this work on the server and saving the result, so you would send back a JSON file like
{
"tags": {
"marsey": ["marseylove", "marseyfemboy", ...],
"othertag": ["emoji1", "emoji2"....]
}
}
Jump in the discussion.
No email address required.
Would it have to be rebuilt everytime an emoji is added?
Jump in the discussion.
No email address required.
Yes. But right now its being rebuilt every time you start typing an emoji...
Jump in the discussion.
No email address required.
compute time on the client isn't actual compute time. Anyone who cares about browser performance is a little soyboy cuckold
Jump in the discussion.
No email address required.
More options
Context
More options
Context
More options
Context