[ruby-core:30913] String#rindex is faster with Regexps than with Strings?

hi,

I noticed that, for a long utf-8 string, rindex(/\n/) is much faster
than rindex("\n"), depending on how far from the end of the string the :

ruby19 -rbenchmark -e 'str = "test\n" * 5_000_000 + "端nicode" * 50; \
p Benchmark.realtime { p str.rindex(/\n/) }'
24999999
0.02609705924987793

ruby19 -rbenchmark -e 'str = "test\n" * 5_000_000 + "端nicode" * 50; \
p Benchmark.realtime { p str.rindex("\n") }'
24999999
2.0839710235595703

Wouldn't it be possible to make rindex(string) faster then, using the
method Regexp uses?

[murphy]