Controversial opinion time: The advantages of Thompson NFA regexes are just theoretical and not worth it. PCRE is faster in the real world.
Note that the context behind that post is from experience with SpiderMonkey’s JS regex implementation and in particular how it gets benchmarked. For JS regex benchmarks it is all about how quickly you can JIT the regex to something reasonable.
-
-
If I was silly enough to return to the world of regex (having been - ahem - granted my freedom by Intel) I think working with something where speed of starting to match an entirely unknown regex is paramount would be interesting. The world of Hyperscan, particularly, assumes ...
-
JS compilers are somewhat frustrating to work on because most of the clever optimizations you want to implement end up not being worth it, because speed of compilation >>> everything else :(
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.