[racket] datum<->syntax with same context, yet "unbound identifier"

From: John Clements (clements at brinckerhoff.org)
Date: Wed Nov 21 13:00:57 EST 2012

On Nov 21, 2012, at 7:50 AM, Greg Hendershott wrote:

> Thank you.
> 
>> The lexical information on the outer list structure can
>> be different than the lexical information on those objects inside.
> 
> Ah.
> 
> By the way, I've read descriptions of syntax objects that list the
> accessors, but when it comes to lexical context say "but you don't
> really need that".
> 
> But how come there isn't an accessor named something like
> `syntax-lexical-context'?  To explore, visualize, debug?
> 
> (Even if it returned a value that's just some opaque ID, but unique so
> it could be compared to other such values.)

I have, in the past, shared your occasional frustration with the lack of transparency in the syntax system; over time, though, I've come to realize that limiting the observable aspects of the system makes it possible to update, optimize, and reorganize it later. Every additional observation you can make is another operation that has to be supported in the future.

Stupid Java Stack traces, grr. (sorry. :)  )

John

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4370 bytes
Desc: not available
URL: <http://lists.racket-lang.org/users/archive/attachments/20121121/d7bce2c8/attachment.p7s>

Posted on the users mailing list.