Yes. However since you can precompile the DB (with the provided flag), this works fine when you have a lot of regexps which don't change and new packs of data

Yes. However since you can precompile the DB (with the provided flag), this works fine when you have a lot of regexps which don't change and new packs of data to parse.. which was (basically) the original use case of hyperscan. For a single use, unless you have hundreds of GB of data to parse, the default ripgrep engine might just be better.