Re: Error definition

Dave Pawson writes:

> On Sat, 5 Feb 2022 at 15:44, C. M. Sperberg-McQueen
> <cmsmcq@blackmesatech.com> wrote:

>> Maybe we just need to say that any reference to 'error' in connection
>> with input strings which are not sentences w.r.t. the input grammar
>> denotes a failure of the input to match the grammar but not repeat not a
>> failure of the grammar or the processor to conform to this spec.
>
> (for me), far too many words?

It's not a proposal for spec prose.  It's an attempt to consider a
possible common frame of reference.

> A user view.
> 1.input.ixml
> 2.input.txt
> (hoped for)
> 3. output.xml
>
> Assumption: The processor is not buggy.
>
> I process 1 and 2.
>   Options:
> I get 3. No messages. We're all happy. Success.
> I get no 3. Something has gone wrong (my defn, an error has occurred).
>   I look to the processor to tell me where that error is?
> <ul>
>  <li> 1 not valid to ixml spec</li>
>  <li> (some part of )2 not a sentence in grammar of 1</li>
>   <li> ???</li>
> </ul>

> My view, the 'transformation' has failed, with an error reported.
> What it is, where to look for it, I turn to the error / warnings /
> output of the processor.

> Basically I'm uninterested in nuances. It worked | it failed to produce 3

What I don't understand in this scenario is:  why do you want to require
the other members of the CG to join you in using the terms "failure" and
"error" in ways that do not distinguish bugs in the processor from bugs
in the grammar from bugs in the input?

And why do you want to impose your way of seeing this case of every user
of ixml?


> AFAICT - it's an error that needs some debugging.

Nothing in your scenario actually specifies that the input in question
is or should be described successfully by the grammr in question.  Did
someone tell you that input.ixml describes the structure of input.txt?
Did someone contruct input.txt in an attempt to follow the rules of
input.ixml?

If someone lied to you, or someone failed to follow the rules of a
grammar, why should that be a violation of the rules of ixml?

Michael



-- 
C. M. Sperberg-McQueen
Black Mesa Technologies LLC
http://blackmesatech.com

Received on Saturday, 5 February 2022 17:26:57 UTC