Escape -?

[benjamingr]

Currently the proposal escapes ] and } which isn't really required since we escape [ and {. The only reason we might need to escape ] is because we need to support the case that the pattern is inserted inside the [.

A preliminary GH code search shows that this pattern is actually used. In particular, underscore's string model does "[" + escapedRegExp +"].

So, in my opinion we should support escaping inside matched sets to be on the safe side, this is somewhat wasteful (because there are other characters that do not need to be escaped in sets but the pattern is very common.

Opinions?

[benjamingr]

Also related, discussion in #17

[ljharb]

I don't think there's any concern with over-escaping here. The important thing RegExp.escape provides is a safety guarantee - not a performance guarantee.

Escaping every single character might be a bit extreme, but certainly escaping the ones that might be problematic in some cases doesn't seem to be.

[domenic]

Maybe what we want is a table in the readme with each escaped character and a reason why it's escaped (preferably in the form new RegExp(something + RegExp.escape("the character") + something) or some subset thereof)

[benjamingr]

@domenic good idea, I'll make one soon.

@ljharb well, the most obvious reason is the readability impact, but I agree this is a use case we want to address.

[nikic]

Context:

• rtrim
• calls defaultToWhitespace
• calls escapeRegExp
• which escapes .*+?^=!:${}()|[\]\/\\

Note that this does not include -, as underscore presumably wants to allow you to write something like trim("a-zA-Z0-9") to trim all alphanumerics.

I suspect that this might be the cases in many places where something is inserted into [] and as such having - escaped there might not really be what you want.

[benjamingr]

Yes, but if you want - to work as part of a group then escape is not
what you're after anyway. You want to be able to use a user input string or
a regular string literally. You can always just concat the string as a
regular expression and not call .escape

On Sat, Jun 20, 2015 at 1:56 PM, Nikita Popov [email protected]
wrote:

Context:

• rtrim
  https://github.com/epeli/underscore.string/blob/71c49d6b9e44567f997b9e11d7a1d748a71040d6/rtrim.js
• calls defaultToWhitespace
  https://github.com/epeli/underscore.string/blob/71c49d6b9e44567f997b9e11d7a1d748a71040d6/helper/defaultToWhiteSpace.js
• calls escapeRegExp
  https://github.com/epeli/underscore.string/blob/71c49d6b9e44567f997b9e11d7a1d748a71040d6/helper/escapeRegExp.js
• which escapes .*+?^=!:${}()|[]/\

Note that this does not include -, as underscore presumably wants to
allow you to write something like trim("a-zA-Z0-9") to trim all
alphanumerics.

I suspect that this might be the cases in many places where something is
inserted into [] and as such having - escaped there might not really be
what you want.

—
Reply to this email directly or view it on GitHub
#28 (comment)
.

[benjamingr]

Superseding this with #29