[racket] Exception Stack Trace Troubles

No, that's not the way Racket works. :)

There is something wrong and our messages are designed to figure out
what that is. Thanks for your patience. I think Matthew's message is
next on the list to answer.

Robby

On Wed, Jun 27, 2012 at 10:15 PM, Chad Albers <calbers at neomantic.com> wrote:
> Yes, intel/amd 64 bit.
>
> I don't have a .racketrc...I need to find out why I need that in the docs.
>
> I did you -j and success!   I got a stack trace.
>
> So, that brings up a question.  Is there a way to support stack traces
> without disabling the jit compiler? I understand that stack traces add
> a certain overhead to executing a program.  I'm just not used to a
> language, not supporting stack traces by default.  Is that just the
> way that Racket works?
>
> --
> Chad Albers
>
>
> On Wed, Jun 27, 2012 at 11:20 PM, Robby Findler
> <robby at eecs.northwestern.edu> wrote:
>> By 64 bit architecture, I assume it is an intel/amd chip?
>>
>> What happens if you pass -j on the command-line to racket when you try
>> this? (You don't have a .racketrc, right?)
>>
>> Does the test suite pass? You'd run that with
>>
>>   racket -qr plt/collects/tests/racket/quiet.rktl
>>
>> (where "plt" is the path to the place where you've installed Racket).
>>
>> Robby
>>
>> On Wed, Jun 27, 2012 at 6:34 AM, Chad Albers <calbers at neomantic.com> wrote:
>>> Hi Eli,
>>>
>>> Here's the file that I'm running in a gist: https://gist.github.com/3003496
>>>
>>> In other words, it's a cut-and-paste of code that you posted earlier
>>> in this chain:
>>>
>>> #lang racket
>>> (with-handlers ([void (λ (e)
>>>                         (continuation-mark-set->context
>>>                          (exn-continuation-marks e)))])
>>>   (+ 1 "two"))
>>>
>>>
>>> Let's say the file is called "example.rkt".  I execute this file using
>>> the CLI as follows:  racket example.rkt.
>>>
>>> That's it.  I'm running Debian linux on 64 bit architecture.  I've
>>> checked the Debian bug reports on racket, and nothing similar has been
>>> reported.
>>>
>>> Thanks again for your help,
>>>
>>> Chad
>>>
>>>
>>> --
>>> Chad Albers
>>>
>>>
>>>
>>> On Wed, Jun 27, 2012 at 2:00 AM, Eli Barzilay <eli at barzilay.org> wrote:
>>>>
>>>> Two days ago, Michael Wilber wrote:
>>>> > If I understand correctly, by default, Racket doesn't provide forms
>>>> > with stack trace information when running from the CLI by
>>>> > default.
>>>>
>>>> It does (and I tried my examples on both racket and drracket).
>>>>
>>>>
>>>> > Does it work from within DrRacket? If so, look into the errortrace
>>>> > module, or add (require errortrace) to the top of the .rkt, or run
>>>> > it like this: racket -l errortrace test.rkt
>>>>
>>>> The purpose of errortrace is to provide a more accurate stacktrace
>>>> (and originally, to provide a trace when mzscheme didn't have one).
>>>>
>>>>
>>>> 9 hours ago, Chad Albers wrote:
>>>> > I'm using Debian Linux.
>>>>
>>>> I tried it now with our debian build (which is an x86_64 build), and
>>>> it worked.
>>>>
>>>>
>>>> 9 hours ago, Chad Albers wrote:
>>>> > It doesn't produce a stack trace of DrRacket.
>>>> >
>>>> > Also, it doesn't produce a stack trace if I include the errortrace
>>>> > on the command line.
>>>>
>>>> Can you post the exact file that you tried and how you ran it?
>>>>
>>>> --
>>>>          ((lambda (x) (x x)) (lambda (x) (x x)))          Eli Barzilay:
>>>>                    http://barzilay.org/                   Maze is Life!
>>>
>>> ____________________
>>>  Racket Users list:
>>>  http://lists.racket-lang.org/users