bug#74385: [PATCH 3/4] srfi-64: Export define-equality-test.

[Tomas Volf <~@wolfsden.cz>]

[-- Attachment #1: Type: text/plain, Size: 2098 bytes --]

Ludovic Courtès <ludo@gnu.org> writes:

> Hi Tomas,
>
> Tomas Volf <~@wolfsden.cz> skribis:
>
>> +(define-syntax define-equality-test
>>    (syntax-rules ()
>>      ((_ name test-proc)
>>       (define-syntax name
>>         (λ (x)
>>           (syntax-case x ()
>>             ((_ test-name expected test-expr)
>> -            #`(%%test-2 #,x test-proc test-name expected test-expr))
>> +            #`(%test-2 #,x test-proc test-name expected test-expr))
>>             ((_ expected test-expr)
>> -            #`(%%test-2 #,x test-proc #f        expected test-expr))))))))
>> +            #`(%test-2 #,x test-proc #f        expected test-expr))))))))
>> +(set-documentation! 'define-equality-test
>> +  "@defspec define-equality-test identifier proc
>> +Define a new test form named @var{identifier} with same signature and usage as
>> +@code{test-eq} but using @var{proc} instead of @code{eq?}.
>
> I didn’t notice earlier, but you can add docstrings like this:
>
>   (define-syntax define-equality-test
>     (syntax-rules ()
>       "This is the docstring."
>       …))

Ah, good catch.  Since that works just for syntax-rules and not
syntax-case, I probably did not moved the documentation string around
when switching between them.  Will update.

>
> As for exporting ‘define-equality-test’, I would clearly mark it as a
> “GNU extension”.
>
> The way this was done before is by having more exports in a separate
> module, like (srfi srfi-9 gnu).
>
> Whether or not you pick this approach, please make sure to document it
> in the manual and to prominently mark it as a GNU extension.

I personally think that leaving it in one module is fine, and I agree it
should be documented.  There are other GNU extensions already in the new
SRFI-64, and I plan to document all once #71300 is merged.  I will wait
for v2 until that happens and will include the documentation changes
required.

Have a nice day,
Tomas

-- 
There are only two hard things in Computer Science:
cache invalidation, naming things and off-by-one errors.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 853 bytes --]