Regular Expressions
Overview
Regular expressions are a cross-cutting skill used in every language, every editor, every CLI tool, and every CI pipeline. They are powerful for pattern matching but can be cryptic — this skill provides a readable reference.
Core Syntax
| Element |
Syntax |
Description |
Example |
| Literal |
abc |
Matches "abc" |
abc matches "abc" |
| Dot |
. |
Any single character |
a.c matches "abc", "a1c" |
| Start anchor |
^ |
Start of string/line |
^Hello matches "Hello world" |
| End anchor |
$ |
End of string/line |
world$ matches "Hello world" |
| Word boundary |
\b |
Word boundary |
\bcat\b matches "cat" not "scatter" |
| Character class |
[abc] |
Any of a, b, or c |
[aeiou] matches vowels |
| Range |
[a-z] |
Any char in range |
[0-9] matches digits |
| Negated class |
[^abc] |
Any char NOT in set |
[^0-9] matches non-digits |
| Shorthand |
\d \w \s |
Digit, word char, whitespace |
\d+ matches "123" |
| Quantifier |
* + ? {n} |
Repetition |
a{3} matches "aaa" |
| Alternation |
| |
OR |
cat|dog matches either |
| Group |
() |
Capturing group |
(ab)+ matches "abab" |
| Non-capturing |
(?:) |
Group without capture |
(?:ab)+ matches "abab" |
| Escape |
\ |
Literal special char |
\. matches "." |
Quantifiers
| Quantifier |
Meaning |
Greedy |
Lazy |
* |
0 or more |
.* |
.*? |
+ |
1 or more |
.+ |
.+? |
? |
0 or 1 |
.? |
.?? |
{n} |
Exactly n |
a{3} |
— |
{n,} |
n or more |
a{2,} |
a{2,}? |
{n,m} |
Between n and m |
a{2,4} |
a{2,4}? |
Greedy quantifiers match as much as possible. Lazy quantifiers (with ? suffix) match as little as possible. Use lazy quantifiers when you need the shortest match.
Character Classes
| Class |
Matches |
Equivalent |
\d |
Digit |
[0-9] |
\D |
Non-digit |
[^0-9] |
\w |
Word character |
[a-zA-Z0-9_] |
\W |
Non-word character |
[^a-zA-Z0-9_] |
\s |
Whitespace |
[ \t\n\r\f\v] |
\S |
Non-whitespace |
[^ \t\n\r\f\v] |
Common Patterns
| Purpose |
Pattern |
Notes |
| Email (simplified) |
[\w.+-]+@[\w-]+\.[\w.]+ |
Covers most common formats |
| URL |
https?://[\w\-.]+(:\d+)?(/[\w\-./?%&=]*)? |
HTTP and HTTPS |
| IPv4 |
\b\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}\b |
Does not validate range |
| ISO date |
\d{4}-\d{2}-\d{2} |
YYYY-MM-DD format |
| UUID |
[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12} |
Lowercase hex |
| US phone |
\(?\d{3}\)?[-.\s]?\d{3}[-.\s]?\d{4} |
With optional formatting |
| Semantic version |
\d+\.\d+\.\d+(-[\w.]+)?(\+[\w.]+)? |
Major.Minor.Patch with optional pre-release |
| Strong password |
^(?=.*[a-z])(?=.*[A-Z])(?=.*\d)(?=.*[@$!%*?&])[A-Za-z\d@$!%*?&]{8,}$ |
Min 8 chars, upper, lower, digit, special |
Lookahead and Lookbehind
| Type |
Syntax |
Description |
| Positive lookahead |
(?=...) |
Matches if followed by |
| Negative lookahead |
(?!...) |
Matches if NOT followed by |
| Positive lookbehind |
(?<=...) |
Matches if preceded by |
| Negative lookbehind |
(?<!...) |
Matches if NOT preceded by |
Practical Examples
Positive lookahead — match a number only if followed by "px":
\d+(?=px)
In "12px 5em 8px", matches "12" and "8".
Negative lookahead — match "http" only if NOT followed by "s":
http(?!s)
Matches "http://" but not "https://".
Positive lookbehind — match a number only if preceded by "$":
(?<=\$)\d+
In "Price: $42 and 10 items", matches "42".
Negative lookbehind — match "cat" only if NOT preceded by "scat":
(?<!s)cat
Matches "cat" in "the cat" but not in "scatter".
Flavor Differences
| Feature |
JavaScript |
Python |
PCRE (.NET/PHP/grep -P) |
POSIX (grep/sed) |
| Named groups |
(?<name>) |
(?P<name>) |
(?P<name>) or (?<name>) |
Not supported |
| Lookbehind |
Fixed-length only |
Variable-length |
Variable-length |
Not supported |
| Unicode |
\p{L} |
\p{L} with re.UNICODE |
\p{L} |
Limited |
| Backreferences |
\1 |
\1 |
\1 |
\1 |
When writing cross-platform regex, stick to the common subset: basic character classes, quantifiers, alternation, and fixed-length lookbehind. Test in multiple flavors if portability matters.
Testing and Debugging
- regex101.com — The gold standard for regex testing. Supports PCRE, JavaScript, Python, and Go flavors. Provides real-time matching, group highlighting, and a detailed explanation of each token.
- regexr.com — Interactive regex editor with a community pattern library. Good for learning and exploring.
Always test your regex against both matching and non-matching inputs, including edge cases like empty strings, special characters, and Unicode text.
Best Practices
- Always test regex with edge cases — empty strings, special characters, very long inputs, and Unicode
- Prefer named groups for readability —
(?<year>\d{4}) is clearer than (\d{4}) when extracting data
- Use non-capturing groups
(?:) when you don't need the match — it avoids cluttering group numbering
- Avoid catastrophic backtracking — nested quantifiers like
(a+)+ can cause exponential time on certain inputs
- Use character classes over alternation where possible —
[aeiou] is faster and clearer than a|e|i|o|u
- Add comments with verbose/extended mode (
(?x) in most flavors, re.VERBOSE in Python) for complex patterns
1---2name: regex3description: |4 Use when writing or debugging regular expressions for pattern matching, validation, search, and text transformation. Covers regex syntax, common patterns, lookahead/lookbehind, character classes, quantifiers, and differences between regex flavors (PCRE, JavaScript, Python, .NET, POSIX).5 USE FOR: regular expressions, regex, pattern matching, text validation, search and replace, PCRE, regex lookahead, regex lookbehind, character classes, quantifiers, capturing groups, non-capturing groups, regex anchors6 DO NOT USE FOR: HTML/XML parsing (use a proper parser), complex data extraction (use jq or a programming language), natural language processing7license: MIT8metadata:9 displayName: "Regular Expressions"10 author: "Tyler-R-Kendrick"11compatibility: claude, copilot, cursor12references:13 - title: "Regular-Expressions.info"14 url: "https://www.regular-expressions.info"15 - title: "PCRE2 Specification"16 url: "https://www.pcre.org/current/doc/html/"17---1819# Regular Expressions2021## Overview2223Regular expressions are a cross-cutting skill used in every language, every editor, every CLI tool, and every CI pipeline. They are powerful for pattern matching but can be cryptic — this skill provides a readable reference.2425## Core Syntax2627| Element | Syntax | Description | Example |28|---------|--------|-------------|---------|29| Literal | `abc` | Matches "abc" | `abc` matches "abc" |30| Dot | `.` | Any single character | `a.c` matches "abc", "a1c" |31| Start anchor | `^` | Start of string/line | `^Hello` matches "Hello world" |32| End anchor | `$` | End of string/line | `world$` matches "Hello world" |33| Word boundary | `\b` | Word boundary | `\bcat\b` matches "cat" not "scatter" |34| Character class | `[abc]` | Any of a, b, or c | `[aeiou]` matches vowels |35| Range | `[a-z]` | Any char in range | `[0-9]` matches digits |36| Negated class | `[^abc]` | Any char NOT in set | `[^0-9]` matches non-digits |37| Shorthand | `\d \w \s` | Digit, word char, whitespace | `\d+` matches "123" |38| Quantifier | `* + ? {n}` | Repetition | `a{3}` matches "aaa" |39| Alternation | `\|` | OR | `cat\|dog` matches either |40| Group | `()` | Capturing group | `(ab)+` matches "abab" |41| Non-capturing | `(?:)` | Group without capture | `(?:ab)+` matches "abab" |42| Escape | `\` | Literal special char | `\.` matches "." |4344## Quantifiers4546| Quantifier | Meaning | Greedy | Lazy |47|------------|---------|--------|------|48| `*` | 0 or more | `.*` | `.*?` |49| `+` | 1 or more | `.+` | `.+?` |50| `?` | 0 or 1 | `.?` | `.??` |51| `{n}` | Exactly n | `a{3}` | — |52| `{n,}` | n or more | `a{2,}` | `a{2,}?` |53| `{n,m}` | Between n and m | `a{2,4}` | `a{2,4}?` |5455Greedy quantifiers match as much as possible. Lazy quantifiers (with `?` suffix) match as little as possible. Use lazy quantifiers when you need the shortest match.5657## Character Classes5859| Class | Matches | Equivalent |60|-------|---------|------------|61| `\d` | Digit | `[0-9]` |62| `\D` | Non-digit | `[^0-9]` |63| `\w` | Word character | `[a-zA-Z0-9_]` |64| `\W` | Non-word character | `[^a-zA-Z0-9_]` |65| `\s` | Whitespace | `[ \t\n\r\f\v]` |66| `\S` | Non-whitespace | `[^ \t\n\r\f\v]` |6768## Common Patterns6970| Purpose | Pattern | Notes |71|---------|---------|-------|72| Email (simplified) | `[\w.+-]+@[\w-]+\.[\w.]+` | Covers most common formats |73| URL | `https?://[\w\-.]+(:\d+)?(/[\w\-./?%&=]*)?` | HTTP and HTTPS |74| IPv4 | `\b\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}\b` | Does not validate range |75| ISO date | `\d{4}-\d{2}-\d{2}` | YYYY-MM-DD format |76| UUID | `[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}` | Lowercase hex |77| US phone | `\(?\d{3}\)?[-.\s]?\d{3}[-.\s]?\d{4}` | With optional formatting |78| Semantic version | `\d+\.\d+\.\d+(-[\w.]+)?(\+[\w.]+)?` | Major.Minor.Patch with optional pre-release |79| Strong password | `^(?=.*[a-z])(?=.*[A-Z])(?=.*\d)(?=.*[@$!%*?&])[A-Za-z\d@$!%*?&]{8,}$` | Min 8 chars, upper, lower, digit, special |8081## Lookahead and Lookbehind8283| Type | Syntax | Description |84|------|--------|-------------|85| Positive lookahead | `(?=...)` | Matches if followed by |86| Negative lookahead | `(?!...)` | Matches if NOT followed by |87| Positive lookbehind | `(?<=...)` | Matches if preceded by |88| Negative lookbehind | `(?<!...)` | Matches if NOT preceded by |8990### Practical Examples9192**Positive lookahead** — match a number only if followed by "px":93```94\d+(?=px)95```96In "12px 5em 8px", matches "12" and "8".9798**Negative lookahead** — match "http" only if NOT followed by "s":99```100http(?!s)101```102Matches "http://" but not "https://".103104**Positive lookbehind** — match a number only if preceded by "$":105```106(?<=\$)\d+107```108In "Price: $42 and 10 items", matches "42".109110**Negative lookbehind** — match "cat" only if NOT preceded by "scat":111```112(?<!s)cat113```114Matches "cat" in "the cat" but not in "scatter".115116## Flavor Differences117118| Feature | JavaScript | Python | PCRE (.NET/PHP/grep -P) | POSIX (grep/sed) |119|---------|-----------|--------|------------------------|-----------------|120| Named groups | `(?<name>)` | `(?P<name>)` | `(?P<name>)` or `(?<name>)` | Not supported |121| Lookbehind | Fixed-length only | Variable-length | Variable-length | Not supported |122| Unicode | `\p{L}` | `\p{L}` with `re.UNICODE` | `\p{L}` | Limited |123| Backreferences | `\1` | `\1` | `\1` | `\1` |124125When writing cross-platform regex, stick to the common subset: basic character classes, quantifiers, alternation, and fixed-length lookbehind. Test in multiple flavors if portability matters.126127## Testing and Debugging128129- **[regex101.com](https://regex101.com)** — The gold standard for regex testing. Supports PCRE, JavaScript, Python, and Go flavors. Provides real-time matching, group highlighting, and a detailed explanation of each token.130- **[regexr.com](https://regexr.com)** — Interactive regex editor with a community pattern library. Good for learning and exploring.131132Always test your regex against both matching and non-matching inputs, including edge cases like empty strings, special characters, and Unicode text.133134## Best Practices135136- Always test regex with edge cases — empty strings, special characters, very long inputs, and Unicode137- Prefer named groups for readability — `(?<year>\d{4})` is clearer than `(\d{4})` when extracting data138- Use non-capturing groups `(?:)` when you don't need the match — it avoids cluttering group numbering139- Avoid catastrophic backtracking — nested quantifiers like `(a+)+` can cause exponential time on certain inputs140- Use character classes over alternation where possible — `[aeiou]` is faster and clearer than `a|e|i|o|u`141- Add comments with verbose/extended mode (`(?x)` in most flavors, `re.VERBOSE` in Python) for complex patterns
Tyler-R-Kendrick