K-user mailing list

["Seyed H. HAERI (Hossein)" <hossein.haeri AT gmail.com>]

Hi Traian,

> Have you modified ../exp.k If so, could you please provide that, too?

I was apparently using the wrong exp.k here. See below:

> Apparently, the key to the error is this:
> Advisory: <automatic>: operator __ has been imported from both
> "exp-lambda.maude", line 281 (mod EXP-SYNTAX) and "exp-lambda.maude",
> line 471 (mod EXP-LAMBDA-SYNTAX) with no common ancestor.
>
> This is the Maude parser complaining that you basically have two 
> "concatenation" productions;

Having had a look into exp-lambda.maude on lines 281 and 471, it looks
like Maude was unhappy about the following two lines:

http://code.google.com/p/k-framework/source/browse/branches/papers/k-primer/exp/exp-imp/exp.k#22

and

http://code.google.com/p/k-framework/source/browse/branches/papers/k-primer/exp/exp-lambda/exp-lambda.k#8

which made me understand that wasn't quite the intended mixture of the
Primer. However, I wonder why should Maude anyhow be unhappy with
defining two concat operators for different syntactic categories (one
for Exp and another for Stmt)?

> There were some plans of beautifying the errors from the Maude parser, but 
> they haven't been implemented yet.

Oh, I cordially think you'd better do that ASAP, or K will never get
spread outside Urbana-Champagne. So far, a sizeable fraction of my K
learning time has been spent on demystifying error messages on trivial
cases like this. I wonder how that scales in real-world examples for
non-experts...

Thanks,
--Hossein

--------------------------------------------------------------------------------------------------------------

Seyed H. HAERI (Hossein)

Research Assistant
Institute for Software Systems (STS)
Technical University of Hamburg (TUHH)
Hamburg, Germany

ACCU - Professionalism in programming - http://www.accu.org/
--------------------------------------------------------------------------------------------------------------