[developers] "nested unification context"

[Ned Letcher]

I'm one of the ones experiencing the problem. Unfortunately I haven't been
able to reliably reproduce it. All I can say is that it only happens after
loading a grammar that has already been loaded. Which makes me wonder if
it's something some kind of cruft floating around that isn't being cleaned
up somehow. It also seems to happen in concentrated amounts, then disappear
for a while, so maybe something about certain grammar states. If I
encounter concentration of errors I will try to see if I can work out how
to replicate it.

Ned


On Fri, May 10, 2013 at 12:11 AM, Stephan Oepen <oe at ifi.uio.no> wrote:

> well, it shouldn't happen.
>
> (pretty much) each unification attempt establishes a new context, which
> determines the scope of which quasi-destructive modifications will be
> undone, when leaving the context.
>
> it sounds as if the problem is not trivially reproduced?  to debug, we
> would need a recipe (and grammar) that (preferably reliably) triggers the
> error.  any chance you could try to put that together?  i would then be
> interested in debugging this issue.
>
> best, oe
>
>
> On Friday, May 10, 2013, Emily M. Bender wrote:
>
>> Dear all,
>>
>> Several students in 567 this quarter are intermittently getting
>> the following error when loading their grammars:
>>
>> Error: Entered a nested unification context - should not happen
>>
>> Any idea what this is?
>>
>> Thanks,
>> Emily
>>
>> --
>> Emily M. Bender
>> Associate Professor
>> Department of Linguistics
>> Check out CLMS on facebook! http://www.facebook.com/uwclma
>>
>
>
> --
>
> +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> +++ Universitetet i Oslo (IFI); Boks 1080 Blindern; 0316 Oslo; (+47) 2284
> 0125
> +++    --- oe at ifi.uio.no; stephan at oepen.net; http://www.emmtee.net/oe/ ---
>
> +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
>


-- 
nedned.net
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.delph-in.net/archives/developers/attachments/20130510/3f4f66dc/attachment.html>