Use Aho–Corasick or Teddy for literals (via aho_corasick).
Use a regular expression set for when patterns are needed (via regex-automata).
Gorispace is two mappings, one in
Japanese and one in English. The patterns ウホ+ is space, ウッホ+ is tab, and
ウホ+ーイ is LF. In English, the patterns are hoo+, hoo+s, and wraa+gh,
respectively. Any characters that don’t appear in those tokens are removed
before processing, so the patterns really have [^ウッホーイ]* or [^hoswragh]*,
respectively, between every character.
In practice, programs will be automatically generated, so a shortcut could check
only the fixed strings with 2–4 os and as and 1–4 ホs.
fact.gs excerpt (wrapped):
ウホホウホホホウホホホ、ウホ。ウホホホホーイウホホホ、ウホウホホホウッホホウホウホホホホウホホ、
ウッホホウホホホウッホホウホーイ。ウッホホホ。ウッホホホホウホホ。ウホホ、ウホホホホ。ウホホホホ
ウッホホホウホーイ。ウホホホ、ウホホホホ。ウホウッホホ。ウッホホ、ウホウッホホ。ウッホ、ウッホホホホ
ウホウホホホホーイウッホホホ。ウッホホ、ウホホホホウホホホウホホホウホホホ。ウッホホホウホホホホ。
ウホホホホーイウホホホホ。ウホホホホウホウッホホホホ。ウッホホホホ、ウッホホホウホウッホ。…
en/fact.gs excerpt (wrapped):
hoooo, hoooo hoo hooo. wraagh. hoooo, hoo hooo hooos. hooo. hoo, hoooo hoooos,
hoo hooos. wraaaagh hooos hooos, hooo, hoo hooo hooo hooos wraaagh, hooo hoooo,
hooo hooos hoooos hoooo hoooos hooos. hooos hoooo wraaagh. hooos, hooos, hoooo
hoo hoo, hoo hooos, hoooo. wraaaagh, hooo hooo hooo, hooos hoos hoooos hoooo
hoooos, hoo. hoooo, wraaaagh, hooos hooos hoooo hoo, hoo hoooo. hooos hoooos, …
hohoho uses snow for space, hohoho
for tab, and LF as itself. When converting from Whitespace to hohoho, spaces are
added after snow and hohoho, but when parsing, the spaces are ignored. It is
important that hohoho be parsed non-overlapping, otherwise, for example,
hohohohohoho would be parsed as TTTT, instead of TT.
hello.ho excerpt:
snow snow snow snow
snow snow snow hohoho snow snow hohoho snow snow snow
hohoho hohoho snow snow snow snow hohoho
snow snow snow hohoho hohoho snow snow hohoho snow hohoho
hohoho hohoho snow snow snow snow hohoho snow
…