[Larceny-users] Char - Printed Representation

[William D Clinger]

Ray Racine wrote:
> Yes, I just glanced at R6RS 4.2.1 and see '@' is only an allowed
> subsequent and not an initial.
>
> Though it is not immediately apparent why that is the case...

'@' is not an initial in IEEE/ANSI Scheme or R5RS
Scheme either, but Larceny used to allow it as a
matter of course, as did many other implementations
of Scheme.  What's new about the R6RS is that it
forbids lexical extensions outside of code that
contains an implementation-specific flag such as
#!larceny.

The net effect, in this case, is that Larceny's
support for the R6RS has reduced the portability
of your programs.

Larceny will accept @ as a symbol when the reader
is in #!larceny mode, but Larceny's output routines
currently try to print all symbols with enough
escapes so they can be read back in no matter what
the reader mode.  We probably should add a switch
that tells the output routines they don't have to
do that, but that switch isn't likely to be added
until v0.96.

Will