Regexp escape error with Regexp option 'u' and dash in string

[mdesantis]

Description

Encountered an error while escaping a string containing a - and passing it to Regexp.new() with u flag enabled

Related Issues

No response

Questions

No response

Demo

https://codesandbox.io/s/node-playground-forked-edd4wv?file=/src/index.js

Reproduction

import rescape from "@stdlib/utils-escape-regexp-string";

console.log(new RegExp(rescape("en-US"))); //=> /en\-US/

console.log(new RegExp(rescape("en-US"), "u")); //=> SyntaxError: invalid identity escape in regular expression

Expected Results

console.log(new RegExp(rescape("en-US"), "u")) should print /en-US/

Actual Results

console.log(new RegExp(rescape("en-US"), "u")) raises an error with a message similar to the following:

SyntaxError: Invalid regular expression: /en\-US/: Invalid escape

Version

@stdlib/utils-escape-regexp-string 0.0.9

Environments

Node.js, Firefox, Chrome

Browser Version

No response

Node.js / npm Version

16

Platform

No response

Checklist

• Read and understood the Code of Conduct.
• Searched for existing issues and pull requests.

[github-actions]

🎉 Welcome! 🎉

And thank you for opening your first issue! We will get back to you shortly. 🏃 💨

[kgryte]

@mdesantis Thanks for opening this issue. Unfortunately, this does not appear to be a bug on our side, but simply a feature of how "Unicode" regular expressions work. Quoting from the MDN docs:

The - character is interpreted differently within character classes. In particular, for Unicode regular expressions, - is interpreted as a literal - (and not as part of a range) only if it appears at the start or end of the character class. For example, /[\w-:]/ is a valid regular expression matching a word character, a -, or :, but /[\w-:]/u is an invalid regular expression, as \w to : is not a well-defined range of characters.

and further

Unicode regular expressions do not support so-called "identity escapes"; that is, patterns where an escaping backslash is not needed and effectively ignored. For example, /\a/ is a valid regular expression matching the letter 'a', but /\a/u is not.

As the purpose of the utils-escape-regexp-string package is to escape regular expression strings for non-unicode regular expressions, the package currently works as intended. However, we should add a note to the package README.

@Planeshifter would you mind taking this on?

[mdesantis]

Makes sense, thanks

[kgryte]

Of course! Thanks for using stdlib!