Re: Error definition

We had a similar issue in Saxon-JS with the doc() function running in a certain browser (IIRC now superseded but in the style of HAL, was termed ‘HD’). 

When presented with a file which wasn’t valid XML, the internal parser invoked would in most browsers, fail with a trappable error. 

But not in this particular and very very common one… No - what you got back was a perfectly valid XML structure, in an HTML vocabulary, which contained within verbiage to the effect ‘The input is not valid XML and fails to parse at……’. Helpful to a human looking at it perhaps, but somewhat irksome when one had to examine a ‘returned XML’ to find out whether it contained such error-reporting information. 

Of course it also meant that there was a certain class of entirely legitimate XML files (those that contained the ‘doom-laden’ phrase) that could never be read as distinguishable from absolute failure.

Sent from my iPad

> On 6 Feb 2022, at 14:15, C. M. Sperberg-McQueen <cmsmcq@blackmesatech.com> wrote:
> 
> 
> Dave Pawson writes:
> 
>>> On Sat, 5 Feb 2022 at 17:26, C. M. Sperberg-McQueen
>>> <cmsmcq@blackmesatech.com> wrote:
>> 
>> IMHO a bug in the processor does not give me 3, hence it is an error.
> 
> The dictionary tells us that "error" describes a situation in which
> someone makes a mistake by violating a rule.
> 
> On your view, in the situation you describe, who made the mistake? Who
> committed the error? What rule did they violate?
> 
>> btw, I'm not requiring anything of anyone. I'm putting forward my
>> view, as you do.
> 
> And I'm trying to understand it.
> 
> If a processor fails to produce an XML parse tree for the input and
> instead produces diagnostic information saying something like "this
> input does not match the grammar; further details below ...", does that
> suffice for your purposes?  Or is it necessary that the word "error" be
> used in the message?
> 
> Is it necessary for your goals with respect to ixml that the spec use
> the word "error" to describe the situation in which you do not get your
> expected output?
> 
> Do problems arise if the word "error" is not used in the spec when
> describing that situation?
> 
>> I am a (potential) user, I think a user view as important as any.
>> And (again) I'm not imposing my view, simply presenting it, as I said
>> in my original post.
> 
> My apologies for misunderstanding.  
> 
>> Do you wish to build a playground for devs only?
> 
> Not particularly.  I would like a playground that is open to all and not
> marked as closed off to me.
> 
> 
>>>> AFAICT - it's an error that needs some debugging.
> 
> Again - who committed the error?
> 
>>> 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?
> 
>> Obfuscation Michael?
> 
> Thank you, no more for me right now. Thank you for offering.
> 
> 
> -- 
> C. M. Sperberg-McQueen
> Black Mesa Technologies LLC
> http://blackmesatech.com
>

Received on Sunday, 6 February 2022 15:38:52 UTC