regexpx v0.1.5

This module exports a single function which turns template strings into RegExp objects, with the following benefits:

New flags

It adds support for the additional regular expression flags n, s, U, x, and X known from other languages:

• n/noCapture: Turns all unnamed capturing groups into non-capturing groups.
• s/singleLine: Makes the . match newline characters.
• U/ungreedy: Makes all greedy quantifiers ungreedy, and all ungreedy quantifiers greedy.
• x/extended: Whitespaces and comments after # are ignored unless escaped. Is active by default and (currently) can't be turned off.
• X/extra: Forbid unnecessary escapes. A \ followed by a character (sequence) without special meaning is a SyntaxError.

New syntax:

It adds support for the additional regular expression features:

• Named capturing groups: ...
• Forbid octal escapes in the input: Octal escapes are generally confusing and even more so in regular expressions: The \10 in /(.)(.)(.)(.)(.)(.)(.)(.)(.)(x)\10/ is a reference to the capturing group (x). The \10 in /(.)(.)(.)(.)(.)(.)(.)(.)(.)[x]\10/ it is the char with code 8. Therefore, and because octal escapes in template strings are (currently) SyntaxErrors themselves, thy are not allowed by RegExpX. To reference unnamed capturing groups, use $1 or \k<0>

Substitutions:

...

• string: ...
• RegExp: ...
• array: ...
• object: ...

API:

const RegExpX = require('regexpx') || window.RegExpX; // CommonJS
/**
 * Transforms a template string into a RegExp object. May be called in two different ways:
 *
 * - to specify flags / options:
 *     @param  {string|object}  options  Either a string of flag characters or an object of flags properties.
 *         Examples:
 *             'i' adds the native case insensitivity flag.
 *             'gimuy' adds all native flags.
 *             'g-im-uy' adds the flags 'g', 'm' and 'y', and removes the flags 'i' and 'u'.
 *             { global: true, } adds the native 'g' flag.
 *             { ungreedy: false, } removes the 'U' flag (if previously added).
 *             'o' would add 'o' as a (unrecognised) native flag (which probably throws in the RegExp constructor).
 *     @return {function}                The RegExpX function with the new options bound.
 *                                       Can now be used to create RexExp objects with these options or can have more options bound.
 *
 * - to create a RegExp object call as a template string tag function:
 *     Since it uses the template strings '.raw' property, no additional escaping compared to regular RegExp literal is required,
 *     except that whitespaces and '#' characters that are not to be removed need to be escaped.
 *     Note: The sequence '${' needs to be escaped to '\${' or '$\{' and will then appear as '\${' or '$\{' in the resulting RegExp.
 *     @param  {object}  hasRaw          An object with a .raw property which is an array of strings.
 *     @param  {...any}  substitutions   Additional values to substitute into the template string.
 *     @invariant                        substitutions.length must be exactly hasRaw.raw.length - 1.
 *     @return {RegExp}                  Enhanced RexExp object.
 */

Examples:

...