[erlang-questions] unicode in string literals

Richard Carlsson <>
Tue Jul 31 09:39:55 CEST 2012

On 07/31/2012 12:57 AM, Richard O'Keefe wrote:
> On 31/07/2012, at 3:17 AM, Richard Carlsson wrote:
>> Even leading or trailing spaces are prone to being edited out by
>> mistake. (Erlang supports the \s escape as a synonym for space, for
>> example $\s, but I don't know any other language that has this
>> pretty obvious feature.)
> Some Prolog systems do (the tokeniser in my book offered it, IIRC).
> One of the people on the Prolog standard mailing list has been
> jumping up and down in anger because SWI Prolog has it.  But when you
> have characters like 0'x, 0'y, 0' , it's really nice to have 0\s.
> (Don't laugh.  Erlang's $x, $y, $ , is just as bad.)

Yes, I think it was some horrible occurrences of $ , in erl_scan that 
originally got me to suggest that \s should be added to Erlang, around 
the time when Barklund was working on the draft Standard.

> I actually have a keystroke in my emacs-ish editor to remove
> trailing spaces because their presence breaks things more often than
> it helps.

I wasn't terribly clear - what I meant was leading or trailing spaces 
within strings, as in "Hello ". It's always slightly worrying that 
someone might not see that they just removed something important.


More information about the erlang-questions mailing list