Re: running code vs. formal testing methods

[Jacob Palme <jpalme at DSV.SU.SE>]

At 14.27 +0100 99-08-02, Graham Klyne wrote:
>I think it's worth noting that there is SOME track record of formal
>specification in RFCs -- the one that first comes to mind is ABNF.  On the
>semantic front, RFC 821 uses state diagrams to describe key aspects of
>processing semantics.

XML has a validator service. If you specify your protocols using XML, 
then you can validate test examples against the specified syntax in 
the XML DTD. This has some similarities to validating ABNF. However, 
both ABNF and XML are limited in what they can check. In XML, I can 
specify that a certain attribute must have one of a list of allowed 
values (so-called enumerated) which is very useful, but I can not 
specify that a certain attribute must be numerical, which is possible 
with ABNF. And they will certainly only check some of your syntax, 
not your semantics.

A problem with ABNF is that there are different dialects of it, you 
can only use a validator if you are using the same dialect of ABNF as 
the validator.

I have tried XML (http://www.stg.brown.edu/service/xmlvalid/) and 
HTML validators (http://validator.w3.org/) available on the net, they 
are very useful. Are there ABNF validators on the net?
------------------------------------------------------------------------
Jacob Palme <jpalme at dsv.su.se> (Stockholm University and KTH)
for more info see URL: http://www.dsv.su.se/~jpalme