Marc Nieper-Wißkirchen <xxxxxx@gmail.com> writes:

> Am Mo., 16. Aug. 2021 um 14:55 Uhr schrieb Jeronimo Pellegrini - j_p at
> aleph0.info (via srfi-discuss list) <xxxxxx@srfi.schemers.org>:
>
>> Shiro Kawai <xxxxxx@gmail.com> writes:
>>
>> > But I don't think we need to accelerate the shift by srfi, either.   I
>> > expect the shift will happen spontaneously, just like Scheme has adopted
>> > the case-sensitive reader, after many existing implementations supported
>> > it.
>>
>> Currently only Guile, Racket and STklos support unicode lambda (and
>> Gauche has an ASCII alternative for a quick lambda).
>> I just uploaded a survey here:
>>
>> https://github.com/schemedoc/surveys/blob/master/surveys/UnicodeLambda.md
>
>
> Thanks!
>
> What does "support Unicode lambda" mean in detail? That the reader treats
> the Greek lambda as the plain-old symbol lambda or that in whatever global
> lexical environment the Greek lambda is bound to the same denotation as the
> plain-old symbol lambda?

I only did a quick test, so it means that, in the default interaction
environment, evaluating (lambda () 'whatever) works.

Either:

some-scheme> ((λ () "ok"))
"ok"

Or

some-scheme> ((λ () "ok"))
--> some error

For example, scheme9 calls λ complains that it found a
funny character:

*** error: funny input character, code: 206
*** trace:
> *** error: funny input character, code: 187
*** trace:

It may be interesting to extend the survey to include more iformation,
as I see it may make a difference wether theunicode lambda is a symbol
alias, a second name for the core form, hardcoded in the compiler,
or whatever else the implementor decided to do.

In the case of STklos, which has no interpreter (it's compiler-only),
it is hardcoded in the compiler (because it was a trivial change -
literally a one-liner).

J.