Regular expression syntax as implemented by JavaScript (ECMAScript), which is also what the Regex Tester uses. Nearly all of it is shared with PCRE, Python, Java, .NET and Ruby; the differences that matter are listed at the end.
Characters and escapes
| Syntax | Matches |
|---|---|
. | Any character except line terminators (\n, \r, \u2028, \u2029). With the s flag, any character at all. |
\d | ASCII digit [0-9]. Unicode digits need \p{Nd} with the u flag. |
\w | Word character [A-Za-z0-9_]. ASCII only, even with u. |
\s | Whitespace: space, tab, line breaks, form feed and Unicode spaces such as U+00A0. |
\D \W \S | The negations: not a digit, not a word character, not whitespace. |
\t \n \r \f \v \0 | Tab, line feed, carriage return, form feed, vertical tab, NUL. |
\xHH \uHHHH \u{H…} | Character by hex code. The brace form needs the u flag. |
\p{…} \P{…} | Unicode property, e.g. \p{L} letter, \p{Lu} uppercase, \p{Script=Greek}, \p{Emoji}. Requires u or v. |
\ | Escape a metacharacter: \. \* \? \( \[ \{ \\ \/ \| \^ \$. |
Character classes
| Syntax | Matches |
|---|---|
[abc] | One of a, b or c. |
[^abc] | Any character except a, b, c. |
[a-z] | Range. [a-zA-Z0-9_] equals \w. |
[\w.-] | Class with an escape and literal dot and hyphen. A hyphen is literal when first, last, or escaped. |
[\]\\] | A literal ] or backslash inside a class must be escaped. |
[\p{L}&&[^\p{Lu}]] | Set operations (intersection &&, subtraction --) with the v flag: letters that are not uppercase. |
Anchors and boundaries
| Syntax | Matches |
|---|---|
^ | Start of input, or start of a line with the m flag. |
$ | End of input, or end of a line with m. Does not match before a trailing newline the way Perl’s does. |
\b | Word boundary: between a \w and a \W (or the edge). \bcat\b matches “cat” but not “concatenate”. |
\B | Not a word boundary. |
(?<=^|\s) | There is no \A, \Z or \G in JavaScript; use anchors, boundaries or lookarounds instead. |
Quantifiers
| Syntax | Matches |
|---|---|
* | Zero or more. |
+ | One or more. |
? | Zero or one. |
{3} | Exactly three. |
{2,5} | Two to five. |
{2,} | Two or more. |
*? +? ?? {2,5}? | Lazy versions: match as little as possible. <.+?> stops at the first >. |
Quantifiers are greedy by default: they take as much as possible and give back only if the rest of the pattern cannot match. Lazy quantifiers take as little as possible. JavaScript has no possessive quantifiers (*+) or atomic groups ((?>…)); rewrite with a negated character class to avoid backtracking.
Groups, alternation and backreferences
| Syntax | Matches |
|---|---|
(abc) | Capturing group. Referenced as $1 in replacements and m[1] in match results. |
(?<name>abc) | Named capturing group. $<name> in replacements, m.groups.name in results. |
(?:abc) | Non-capturing group: grouping without a capture slot. Prefer it when you do not need the text. |
a|b | Alternation, lowest precedence. ^(cat|dog)$ not ^cat|dog$. |
\1 \k<name> | Backreference to what a group captured. (["'])(.*?)\1 matches a quoted string with matching quotes. |
(?i:abc) | Inline modifier group (ES2025): case-insensitive inside the group only. Supported in current Chrome, Firefox and Safari; check older targets. |
Lookahead and lookbehind
| Syntax | Matches |
|---|---|
x(?=y) | Positive lookahead: x only if followed by y. y is not consumed. |
x(?!y) | Negative lookahead: x only if not followed by y. |
(?<=y)x | Positive lookbehind: x only if preceded by y. |
(?<!y)x | Negative lookbehind: x only if not preceded by y. |
Lookarounds are zero-width: they test without moving the position. JavaScript lookbehind may contain variable-length patterns, unlike PCRE and Python, which require fixed length.
\d+(?=%) // "50" in "50%"
(?<=\$)\d+(\.\d\d)? // "12.50" in "$12.50"
^(?=.*[A-Z])(?=.*\d).{8,}$ // at least 8 chars with an uppercase letter and a digitFlags
| Flag | Effect |
|---|---|
g | Global: find all matches; enables matchAll and repeated exec. |
i | Case-insensitive. |
m | Multiline: ^ and $ match at line breaks. |
s | dotAll: . matches line breaks. |
u | Unicode: correct surrogate pair handling, \p{…}, \u{…}. Also makes invalid escapes an error. |
v | Unicode sets (ES2024): everything u does plus set operations and string properties like \p{RGI_Emoji}. |
y | Sticky: match only at lastIndex. |
d | Indices: m.indices gives [start, end] for each group. |
Recipes
| Pattern | Purpose |
|---|---|
^[\w.+-]+@[\w-]+\.[\w.-]+$ | Pragmatic email check. Full RFC 5322 validation is impractical in a regex; verify by sending mail. |
^https?:\/\/[^\s/$.?#][^\s]*$ | URL with http or https scheme. |
^\d{4}-(0[1-9]|1[0-2])-(0[1-9]|[12]\d|3[01])$ | ISO date YYYY-MM-DD with plausible month and day (does not check month length). |
^(?:(?:25[0-5]|2[0-4]\d|1?\d?\d)\.){3}(?:25[0-5]|2[0-4]\d|1?\d?\d)$ | IPv4 address with each octet 0–255. |
^#(?:[0-9a-f]{3}){1,2}$ | Hex colour, 3 or 6 digits (add i). |
^[0-9a-f]{8}-[0-9a-f]{4}-[1-8][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ | UUID (add i). |
[ \t]+$ | Trailing whitespace (use gm). |
(\r?\n){3,} | Three or more consecutive line breaks; replace with \n\n to collapse blank lines. |
<[^>]+> | HTML tag (rough). Do not parse HTML with regex beyond quick cleanups. |
\B(?=(\d{3})+(?!\d)) | Positions for thousands separators: "1234567".replace(re, ",") → 1,234,567. |
Differences between engines
| Feature | JavaScript | PCRE / PHP | Python re | Go RE2 |
|---|---|---|---|---|
| Named groups | (?<n>…) | (?<n>…) or (?P<n>…) | (?P<n>…) | (?P<n>…) |
| Lookbehind | Yes, variable length | Fixed length | Fixed length | No |
| Possessive / atomic | No | Yes | 3.11+ | No |
Inline flags (?i) | Group form only, recent engines | Yes | Yes | Yes |
| Unicode classes | \p{L} with u | \p{L} with u | Via regex module | \p{L} |
| Backreferences | Yes | Yes | Yes | No (linear-time guarantee) |
\A \Z \z | No | Yes | \A \Z | \A \z |
The full ECMAScript grammar is in the ECMA-262 specification; MDN’s regular expression guide is the practical reference.