String methods

Joins two strings. A null operand stringifies as "null", matching JavaScript — including a `number | null` held in a parameter, a local, an object field, or returned from a function.

console.log('x' + null)   // xnull   (matches JS)

The number of bytes in the string. This compiler's strings are byte sequences, so for ASCII/Latin-1 text this equals the character count.

console.log('Hello'.length)   // 5

Returns a section of the string without mutating the original. Negative indices count from the end.

startnumber

Zero-based start index; negative counts from the end.

end?number

Exclusive end index; defaults to the string length.

string — A new string spanning [start, end).

console.log('Hello, World!'.slice(7, 12))   // World
console.log('Hello, World!'.slice(-6))      // orld!

Like slice, but negative arguments clamp to 0 and a start greater than end is swapped, matching JavaScript.

startnumber

Start index; clamped to 0 if negative.

end?number

Exclusive end index; defaults to the string length.

string — A new string between the two indices.

console.log('Hello, World!'.substring(0, 5))   // Hello
console.log('Hello, World!'.substring(5, 0))   // Hello   (swapped)

Deprecated but common: extracts length characters from start. A negative start counts from the end; length defaults to the rest of the string.

startnumber

Start index; negative counts from the end.

length?number

Number of characters to take; defaults to the rest of the string.

string — A new string of at most length characters from start.

console.log('Hello, World!'.substr(7, 5))   // World
console.log('Hello, World!'.substr(-6))     // orld!

Returns the byte index of the first occurrence of the substring at or after the optional fromIndex, or -1 if absent. Binary-safe past an embedded NUL.

number — First match index, or -1.

console.log('the quick brown fox'.indexOf('brown'))   // 10

Returns the byte index of the LAST occurrence of the substring at or before the optional fromIndex, or -1. Binary-safe (a descending memcmp scan); an empty needle returns the string length.

number — Last match index, or -1.

console.log('path/to/file'.lastIndexOf('/'))   // 7

Whether the string contains the substring, searching from the optional position. Binary-safe (searches with memmem over a length header).

boolean — true if the substring is present.

console.log('report.pdf'.includes('.pdf'))   // true

Whether the string begins with the prefix at the optional position (default 0). Returns false when the string is shorter than the prefix.

boolean — true if the string starts with the prefix.

console.log('https://example.com'.startsWith('https'))   // true

Whether the string ends with the suffix, treating the string as if it ended at the optional endPosition. Returns false when the string is shorter than the suffix.

boolean — true if the string ends with the suffix.

console.log('report.pdf'.endsWith('.pdf'))   // true

Replaces the first occurrence of `from` with `to`, returning the original string when there is no match. See RegExp for regex-pattern edges.

string — A new string with the first match replaced.

console.log('Hello World'.replace('World', 'TypeGo'))   // Hello TypeGo

Splits the string on a string or RegExp separator into an array. An empty separator splits into individual characters; the optional limit caps the result to the first limit segments — all matching JavaScript.

string[] — The array of substrings (at most limit of them).

console.log('one,two,three'.split(','))      // ['one','two','three']
console.log('a-b-c-d'.split('-', 2))        // ['a','b']

Strips leading and trailing whitespace. Uses the full JS WhiteSpace/LineTerminator set (U+00A0, U+1680, U+2000–200A, U+2028/29, U+202F, U+205F, U+3000, U+FEFF — UTF-8-aware), not just ASCII.

string — The trimmed string.

console.log('  hello world  '.trim())   // hello world

Strips whitespace from one side only, using the same full whitespace set as trim().

string — The one-sided-trimmed string.

console.log('[' + '  hi  '.trimStart() + ']')   // [hi  ]

Identity on a string, matching JavaScript. Kept because Node code habitually calls it on values that are Buffers there but strings here (spawnSync results, stream chunks).

string — The same string.

Returns the string uppercased by the Unicode Default Case Conversion, as Node: every plane's simple mappings plus the SpecialCasing expansions ('ß' becomes 'SS', 'fi' becomes 'FI'); an embedded NUL is an ordinary character.

string — The uppercased string.

console.log('straße'.toUpperCase())   // STRASSE

Returns the string lowercased by the Unicode Default Case Conversion, as Node — including the context-sensitive final sigma ('ΟΔΥΣΣΕΥΣ' becomes 'οδυσσευς') and the 'İ' expansion.

string — The lowercased string.

console.log('ΟΔΥΣΣΕΥΣ'.toLowerCase())   // οδυσσευς

Returns the string repeated n times.

string — The repeated string.

console.log('ab'.repeat(3))   // ababab

Pads the start of the string to a target length. An empty pad string is a no-op, matching JavaScript.

lennumber

Target total length.

pad?string

Pad string; defaults to a space.

string — The padded string.

console.log('7'.padStart(3, '0'))   // 007

Pads the end of the string to a target length. Same empty-pad rule as padStart.

lennumber

Target total length.

pad?string

Pad string; defaults to a space.

string — The padded string.

console.log('7'.padEnd(3, '.'))   // 7..

Returns the string followed by each argument, left to right. Arguments are converted with ToString, as in a template literal: an object's toString() is asked before its valueOf() (the + operator asks valueOf() first). A spread array contributes each of its elements.

...valuesstring[]

Values to append; none returns the string itself.

string — The concatenated string.

console.log('a'.concat('b', 'c'))          // abc
console.log('s:'.concat(...['p', 'q']))   // s:pq

Returns the byte-space code unit at index i. Bounds-checked: an out-of-range index (negative or >= length) returns NaN, as real JS — the result is a double for exactly that reason.

number — The code unit, or NaN if out of range.

console.log('Hello'.charCodeAt(0))   // 72

Returns the character at index i, with negative indices counting from the end; an out-of-range index is a real undefined, as in Node.

string | undefined — The character at i, or undefined if i is out of range.

Returns the character at index i. Never wraps a negative index from the end — always "" for any out-of-range i, matching real JS's distinction from .at().

string — The character at i, or "" if out of range.

Returns the code point at index i, or undefined if out of range.

number | undefined — The code point, or undefined if the index is out of range (as in Node).

console.log('Hello'.codePointAt(0))   // 72
console.log('Hello'.codePointAt(99) ?? -1)   // -1

Unicode normalization (NFC/NFD/NFKC/NFKD). Deliberately deferred, not attempted — it needs real Unicode normalization tables this compiler has no infrastructure for; a fake identity-only implementation would silently mis-normalize any non-ASCII composed/decomposed text.

PCRE2-backed regex matching. .match() is real-JS-shaped, and .matchAll() is a RegExp String Iterator of exec results. See RegExp for the pattern-language caveats.

RegExpMatchArray | null | RegExpStringIterator<RegExpExecArray> — The match, or an iterator of matches.

Searches for a pattern and returns its index, or -1. A plain-string pattern is coerced to a RegExp exactly as in real JS (metacharacters are interpreted — `'a.b'.search('.')` is 0), and a RegExp pattern runs a real PCRE2 search.

number — First match index, or -1.

console.log('the quick brown fox'.search('brown'))   // 10

Replaces every non-overlapping occurrence of `from` with `to`. An empty search is JS-faithful — `'abc'.replaceAll('', '-')` is `'-a-b-c-'`.

string — A new string with all matches replaced.

console.log('a-b-a-b'.replaceAll('-', ' '))   // a b a b

Compares two strings, returning exactly -1, 0, or 1. Binary-safe past an embedded NUL.

number — -1, 0, or 1.

console.log('apple'.localeCompare('banana'))   // -1

Builds a string from UTF-16 code units (each argument's ToUint16), stored as UTF-8; a high/low surrogate pair among the arguments is one character. Static method on String.

string — The constructed string.

console.log(String.fromCharCode(72, 105))   // Hi

Builds a string from code points, encoded as UTF-8; a value that is not an integer in [0, 0x10FFFF] throws RangeError: Invalid code point.

string — The constructed string.

console.log(String.fromCodePoint(84, 121, 112, 101, 71, 111))   // TypeGo

Interleaves the raw (undecoded) quasi text with the string-coerced interpolations — escape sequences appear verbatim (String.raw`a\nb` is a\nb), byte-for-byte the same as Node. Called as a function it takes any object whose raw is array-like, which is how a tag forwards its strings (String.raw(strings, ...values)).

console.log(String.raw`C:\path\to\file`)  // C:\path\to\file