re 0.1.0 on Edge Python

re

re works like Python’s. match anchors at the start, search looks anywhere and fullmatch takes the whole string, and each hands back a Match or None.

import re

m = re.search(r"(\d+)-(\d+)", "tel 12-34")
print(m)
print(m.group(), m[1], m[2])
print(m.groups(), m.span(), m.start(2))
print(re.match(r"\d+", "abc123"))

{
  "imports": {
    "re": "https://cdn.edgepython.com/pkg/re/0.1.0/app.edge#sha256-828e7c2ddf520235ffda8c833fff0fce579abe53ac15b0522afd8ef78da47aa4"
  }
}

<re.Match object; span=(4, 9), match='12-34'>
12-34 12 34
('12', '34') (4, 9) 7
None

Output

Named groups answer by name too, and a group that took no part reads as None.

import re

m = re.match(r"(?P<user>\w+)@(?P<host>\w+)", "ana@mail")
print(m["user"], m.groupdict())
print(re.match(r"(a)|(b)", "b").groups())

{
  "imports": {
    "re": "https://cdn.edgepython.com/pkg/re/0.1.0/app.edge#sha256-828e7c2ddf520235ffda8c833fff0fce579abe53ac15b0522afd8ef78da47aa4"
  }
}

ana {'user': 'ana', 'host': 'mail'}
(None, 'b')

Output

findall gives the text of every match, or its groups when the pattern has them, and finditer gives each Match with its position.

import re

print(re.findall(r"\d+", "a1 b22 c333"))
print(re.findall(r"(\w)=(\d)", "a=1 b=2"))
for m in re.finditer(r"\d+", "a1 b22"):
    print(m[0], m.start())

{
  "imports": {
    "re": "https://cdn.edgepython.com/pkg/re/0.1.0/app.edge#sha256-828e7c2ddf520235ffda8c833fff0fce579abe53ac15b0522afd8ef78da47aa4"
  }
}

['1', '22', '333']
[('a', '1'), ('b', '2')]
1 1
22 4

Output

sub replaces with a template, where \1 and \g<name> bring back a group, or with a function that gets each Match. count caps how many, subn also says how many it replaced, and split cuts at every match.

import re

print(re.sub(r"(\w+)@(\w+)", r"\2.\1", "user@host"))
print(re.sub(r"\d+", lambda m: str(int(m[0]) * 2), "a1 b22"))
print(re.subn(r"\d", "#", "a1b2c3", count=2))
print(re.split(r"[,;]\s*", "a, b;c"))

{
  "imports": {
    "re": "https://cdn.edgepython.com/pkg/re/0.1.0/app.edge#sha256-828e7c2ddf520235ffda8c833fff0fce579abe53ac15b0522afd8ef78da47aa4"
  }
}

host.user
a2 b44
('a#b#c3', 2)
['a', 'b', 'c']

Output

compile checks a pattern once and hands back a Pattern with the same methods. The flags re.I, re.M and re.S ignore case, anchor at every line and let the dot match a newline, and escape makes any text match as written.

import re

p = re.compile(r"hola", re.I)
print(p, p.search("¡HOLA!"))
print(re.findall(r"^\w", "a\nb", re.M))
print(re.escape("1+1=2?"))

{
  "imports": {
    "re": "https://cdn.edgepython.com/pkg/re/0.1.0/app.edge#sha256-828e7c2ddf520235ffda8c833fff0fce579abe53ac15b0522afd8ef78da47aa4"
  }
}

re.compile('hola', re.IGNORECASE) <re.Match object; span=(1, 5), match='HOLA'>
['a', 'b']
1\+1=2\?

Output

A bad pattern raises re.error, which is ValueError here, and a pattern that backtracks without end raises RuntimeError instead of hanging. bytes patterns are not supported.

import re

try:
    re.compile("(unbalanced")
except re.error as e:
    print("bad pattern:", e)

try:
    re.search("(a+)+$", "a" * 40 + "!")
except RuntimeError as e:
    print("too slow:", e)

{
  "imports": {
    "re": "https://cdn.edgepython.com/pkg/re/0.1.0/app.edge#sha256-828e7c2ddf520235ffda8c833fff0fce579abe53ac15b0522afd8ef78da47aa4"
  }
}

bad pattern: missing ), unterminated subpattern at position 0
too slow: catastrophic backtracking: O(n^2) time or worse on this input, simplify nested quantifiers

Output

re

re

Versions