Python regex cheat sheet fast easy only examples

README.md

Official python re examples are so bad

Read this small note before you go

The python regex structure is mostly this
re.<method> is the go to interface with common signature as (pattern, string, flags=0)
and most of the time there is an equivalent Pattern.<method> with the first argument being the string instead of pattern
But the <method> has [, pos[, endpos]] and no flags=0 which might be useful
So I will omit all the equivalent methods' signatures

# Example
re.search(pattern, string, flags=0)
re.match(pattern, string, flags=0)
re.fullmatch(pattern, string, flags=0)
# this is how the signatures are changed
Pattern.search(string[, pos[, endpos]])
Pattern.match(string[, pos[, endpos]])
Pattern.fullmatch(string[, pos[, endpos]])

re.Match object

>>> m = re.search(r"c", "abcde")
>>> a
<re.Match object; span=(2, 3), match='c'>
>>> m.group(0) # this gets the whole match
'c'
>>> m = re.search(r"(c)", "abcde") # this one has a group
>>> m.group(1) # first group
'c'
>>> m.group(2) # if out of index
IndexError: no such group

Match

re.search("c",  "abcdef")  # <re.Match object; span=(2, 3), match='c'>
r"c".search("abcdef") # <re.Match object; span=(2, 3), match='c'>

# Below two are equivalent
re.search("^c", "abcdef")  # None
re.match("c", "abcdef")  # None
# re.match implies ^
re.fullmatch("c", "abcdef") # None
# re.fullmatch implies ^$

Signature

re.search(pattern, string, flags=0)
re.match(pattern, string, flags=0)
re.fullmatch(pattern, string, flags=0)
# this is how the signatures are changed
Pattern.search(string[, pos[, endpos]])
Pattern.match(string[, pos[, endpos]])
Pattern.fullmatch(string[, pos[, endpos]])

Match multiple

re.findall("c",  "cc")       # ["c", "c"]
re.findall("(c).", "cdce")      # ["c", "c"] - when only 1 group, return group
re.findall("(c)(.)", "cdce") # [("c", "d"), ("c", "e")] - when 2 or more, return tuple
re.finditer("c", "cc")  # Same as findall but an iterable
# If you want to get multiple match objects go to the Bonus section

Signature

re.findall(pattern, string, flags=0)
re.finditer(pattern, string, flags=0)

Substitution

re.sub("c", "5", "abcde")  # "ab5de"
re.sub("c(.)", "\\1k", "abdke")  # "ab5de"
# a Match object is always passed
re.sub("c", lambda x: x.group(0).upper(), "abcde")  # "abCde"
re.subn("c", "5", "abcde") # ("ab5de", 1) # because it replaced once

Signature

re.sub(pattern, repl, string, count=0, flags=0)
re.subn(pattern, repl, string, count=0, flags=0)

Split - implies //g

# ['space', 'tab', 'newline', 'allworks']
re.split(r"\s+", "space tab	newline\nallworks")

Signature

re.split(pattern, string, maxsplit=0, flags=0)

Flags

# Only the useful ones. There is no g flag
re.IGNORECASE -> re.I
re.MULTILINE -> re.M
re.DOTALL -> re.S # single line

Misc

r = re.compile("my\spattern", re.I | re.M)
r.match("whatever") # None

Bonus

# Match multiple Match objects
def match_multi(pattern, string, flags=0):
    i = 0
    while val := pattern.search(string, i):
        i = val.end()
        yield val

# Example code
regex = re.compile(r"hi (\w+)", flags=re.I)
for m in match_multi(regex, "Hi Stewie, hi Joe, hi Peter"):
    print(f"{m.group(1):10}| {repr(m)}")