@montegoulding, thanks for your PR! By analyzing the history of the files in this pull request, we identified @livecodefraser and @livecodeali to be potential reviewers.

Probably should add support for \u and \U to this...

Added \u and \U, however, I still need to work out how to handle parser errors... eg \U without 8 hex digits after etc. Ideally it would throw a parser error but at the moment it's just setting the token to empty.

The implementation of \U is wrong as it should use codepoint rather than unichar...

Tis good fun ;-)

OK - I gotta admit that's impresive. I do echo Mike's concern, though, that things like Windows UNC formats like "C:\user\docs\textLetter.txt" not get get mangled in transit.

Well "C:\user\docs\textLetter.txt" would remain untouched of course and if you needed to hard code such a string in single quotes then you would need to double escape the backslash like in every other language ('C:\\user\\docs\\textLetter.txt').

Like this

Ah... I missed that. Escaping is within single quotes, not double quotes. I expect there may be some user confusion ahead since the two aren't completely interchangeable, but it does very neatly get around any backwards compatibility issues and allow embedding at the same time. Nice.

@mwieder as it's not likely this PR will be accepted then there really won't be any chance for user confusion... having said that the whole point of single quotes is to create strings that aren't the same as can be in double quotes even if that is just the double quote char permitted and not escapes. I don't really see the point of offering a solution only for quotes though. I'd rather leave things as they are until I can have a proper solution allowing me to embed return & tab & quote and so on in a constant declaration.

A bit more fun because it came up in discussion ;-)

Can we revive this PR and discussion now (three years later)?

I'd be surprised if we did ;-)