[R6RS] Fancy symbol literals [was: How to make -> a valid identifier]

[Michael Sperber]

Marc Feeley <feeley at iro.umontreal.ca> writes:

> While we're on the subject of symbol syntax, we should discuss a  
> notation for write/read invariance of symbols.  For example, what is  
> the external representation of (string->symbol "Will \"does\" |not|  
> (like) this")?  I propose:
>
>        |Will "does" \|not\| (like) this|

There's also the suggestion Will made in Snowbird, which is to
disallow the creation of symbols that do not match <symbol>, and
either return #f or signal an error in this case.  This would avoid
having to extend the literal syntax, which I think would be a good
thing:  The bar notation also drags in the notion of escapes, and
possibly creates a conflict with the mantissa-width notation for
floating-point literals.

Could someone say what the rationale for the full-scale
`string->symbol' is?  I always assumed it was a leftover from the Lisp
days when there weren't any real strings.

-- 
Cheers =8-} Mike
Friede, V?lkerverst?ndigung und ?berhaupt blabla