On Tue, May 19, 2015 at 8:50 AM, Matthew Flatt <[email protected]> wrote:
> At Sun, 17 May 2015 10:32:15 +0100, Laurent wrote:
>> The following recent change breaks rwind/x11 [1, 2]:
>> https://github.com/plt/racket/commit/ad899173b97ce2ca51612a496d17ea32db45ecdc#di
>> ff-4f20ebf5b6abb6ab4c0b80b41efa1614R819
>>
>> Now when running rwind, it fails with:
>> _enum: key is not a symbol
>>   symbols: '(2)
>>   key: 2
>>   value: 3
>>   context...:
>>    /usr/share/racket-6.2.0.3/collects/ffi/unsafe.rkt:832:2: loop
>>    /usr/share/racket-6.2.0.3/collects/ffi/unsafe.rkt:821:0: unpack63
>>    /home/orseau/Unison/Prog/Racket/x11/x11/x11.rkt: [running body]
>>    /home/orseau/Unison/Prog/Racket/rwind/color.rkt: [traversing imports]
>>    /home/orseau/Unison/Prog/Racket/rwind/rwind.rkt: [traversing imports]
>>
>> Unfortunately, I couldn't get a better error stacktrace, even with `racket
>> --no-jit -l errortrace`.
>>
>> I'm having trouble debugging this, as it's hidden under piles of FFI calls,
>> but would allowing for integers in place of symbols be really a bad idea
>> here?
>
> The old treatment of integers in `_enum` was not intentional, but I
> think it would count as a bad idea if it had been intentional:

This behavior seems clearly like a bad idea starting from scratch, but
is fixing it worth the backwards incompatibility?

An alternative might be a dynamic error instead of converting 2 to 3.

Sam

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-dev/CAK%3DHD%2BaQ9zGUUejWjSNMbFr1h4Z8Hyq1WeB%3D2QStO9KF3Tkpwg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to