56 suggests some new atoms that perform regex operations, but perhaps it'd be better to introduce a new kind of "string start" that syntactically starts a regex, with its own set of terminators that each corresponds to a regex operation.
This would be a good use for (, which doesn't have meaning yet.
For example, #56 proposes
œF: Regex Findall: Find all instances of regex x in string y; return a list of matches
And “[a-z]”œF is 9 bytes, but something like ([a-z]Ḟ is only 7. (Here Ḟ would be the regex terminator for “findall”).
Instead of “[a-z]+”UÐR, we'd write U([a-z]+Ṙ. So Ṙ is a regex terminator that makes the result act as a quick.
It also makes sense to introduce all kinds of “macros” within the regex syntax. For example, ạ could be short for [a-z], and ¹ for \1. And of course, Ɗ makes a (...) group out of the previous three regex tokens.
56 suggests some new atoms that perform regex operations, but perhaps it'd be better to introduce a new kind of "string start" that syntactically starts a regex, with its own set of terminators that each corresponds to a regex operation.
This would be a good use for
(, which doesn't have meaning yet.For example, #56 proposes
And
“[a-z]”œFis 9 bytes, but something like([a-z]Ḟis only 7. (HereḞwould be the regex terminator for “findall”).Instead of
“[a-z]+”UÐR, we'd writeU([a-z]+Ṙ. SoṘis a regex terminator that makes the result act as a quick.It also makes sense to introduce all kinds of “macros” within the regex syntax. For example,
ạcould be short for[a-z], and¹for\1. And of course,Ɗmakes a(...)group out of the previous three regex tokens.