node-unicode/node-unicode-data

Use `gzipInline` more often

Closed this issue · 0 comments

Should we?

We could use a simple heuristic, e.g. when the codePoints array has a length greater than 10000, gzip it.

This would be problematic for Browserified builds of mathiasbynens/regexpu-core#2, though. Not anymore now that https://github.com/mathiasbynens/regenerate-unicode-properties is used.