Python — RegEx

The re module

import re

pattern = r"\d{3}-\d{4}"          # raw string — backslashes stay literal

re.search(pattern, "Call 555-1234")     # Match or None
re.match(pattern, "555-1234")           # anchored at START only
re.findall(pattern, "555-1234, 555-5678")   # all matches as list
re.sub(pattern, "XXX", text)            # search & replace

Raw strings: always write patterns as r"..." so \d reaches the regex engine instead of Python interpreting the backslash.

Character building blocks

Pattern	Matches
\d / \w / \s	digit / word char / whitespace
.	any character
[aeiou]	one listed char; [a-z0-9] ranges
[^0-9]	NOT a digit
^…$	start / end anchors
colou?r	optional (0–1)
\d{2,4}	two to four
a+ / a*	one+ / zero+

Groups extract pieces

Parentheses capture:

m = re.search(r"(\d{4})-(\d{2})-(\d{2})", "Released 2026-08-23")

m.group(0)    # "2026-08-23"  whole match
m.group(1)    # "2026"
m.group(2, 3) # ("08", "23")

# named groups — self-documenting:
m = re.search(r"(?P<year>\d{4})-(?P<month>\d{2})", "2026-08")
m["year"]     # "2026"

split & sub with patterns

re.split(r"[;,]", "a,b;c")        # ['a', 'b', 'c']
re.sub(r"\s+", " ", "too   many   spaces")
re.sub(r"(\w+)@(\w+)", r"<\1 at \2>", "ada@py")   # groups reused in replacement

Practical pattern collection

email_ok = re.fullmatch(r"\S+@\S+\.\S+", email)
phone = re.search(r"\b\d{10}\b", text)
words = re.findall(r"\b\w+\b", sentence)         # tokenizer
slug = re.sub(r"[^\w]+", "-", title.lower()).strip("-")

Flags

re.search(r"hello", text, re.IGNORECASE)   # case-insensitive
re.findall(r"^.", text, re.MULTILINE)      # ^ per line
re.search(r"a  b", text, re.VERBOSE)       # ignore spaces in pattern + comments

match vs search vs fullmatch

Function	Succeeds when pattern matches…
match	starts at position 0
search	anywhere in string
fullmatch	the ENTIRE string

Validation wants fullmatch; extraction wants search/findall.

Honest limits

Regex can't parse nested HTML/JSON reliably, and mega-patterns become write-only. Use it for shapes; use real parsers for structured data.

Mini Practice

1. Extract every word ≥5 letters from a paragraph.
2. Validate five phone shapes with fullmatch.
3. Named-group parse "name=Alice;age=36".
4. Slugify three titles via sub.
5. Mask all but last 4 digits of card numbers.

Next: PIP →