Cookbook

>>> import remagic as rm

ISO date

>>> day = rm.DIGIT.times(2)
>>> iso = rm.START + rm.DIGIT.times(4) + "-" + day + "-" + day + rm.END
>>> bool(iso.compile().match("2027-03-14"))
True

Hex colour

>>> hexdigit = rm.char_in("", ranges=[("0", "9"), ("a", "f"), ("A", "F")])
>>> colour = "#" + (hexdigit.times(6) | hexdigit.times(3))
>>> [bool(colour.compile().fullmatch(s)) for s in ("#fff", "#a1b2c3", "#ffff")]
[True, True, False]

Semantic version

>>> number = rm.DIGIT.one_or_more()
>>> core = number.group("major") + "." + number.group("minor") + "." + number.group("patch")
>>> m = core.compile().fullmatch("1.20.3")
>>> m["major"], m["minor"], m["patch"]
('1', '20', '3')

Repeated words

>>> repeat = (
...     rm.WORD_BOUNDARY + rm.WORD.one_or_more().group("w") + rm.WHITESPACE + rm.ref("w")
... )
>>> repeat.compile().search("this is is a test").group("w")
'is'

Words that are not followed by something

>>> rm.exactly("foo").not_followed_by("bar").compile().findall("foobar foobaz")
['foo']

Case-insensitive part of a pattern

>>> tag = "ID-" + rm.char_in("", ranges=[("a", "z")]).times(3).ignore_case()
>>> bool(tag.compile().fullmatch("ID-AbC"))
True
>>> bool(tag.compile().fullmatch("id-abc"))
False

Any word from a list

any_of escapes each item and removes duplicates.

>>> keywords = rm.any_of(["if", "else", "c++"])
>>> [m.group() for m in (rm.WORD_BOUNDARY + keywords).compile().finditer("if c++ else")]
['if', 'c++', 'else']

Letters in any script

Needs the regex extra.

>>> word = rm.unicode_property("L").one_or_more()
>>> word.compile().findall("naïve façade")
['naïve', 'façade']