[Ltru] Re: ABNFs in draft-ietf-ltru-matching-05

Kent Karlsson wrote:

> alphanum      = (ALPHA / DIGIT)       ; letters and numbers

> Now that spaces are so important, I'm not sure how important
> the parentheses are... 

It's equivalent, the parentheses are unnecessary in this case.
4234 "strongly advises" to use parentheses  for something like
x = a b / c

It's not obvious that this is the same as x = ( a  b ) / c.

But for y = d / e or for z = f g parentheses are unnecessary.

> So please use non-terminal names that aren't used in
> -registry, unless the ABNF for them is exactly the same
> (in which case we can just refer to figure 1 of -registry
> for those).

I'd think that readers would appreciate it if used -registry
stuff is copied verbatim.  References are "cleaner" (impossible
to introduce subtle differences), but a reader interested in
47 KB -matching might be unwilling to read 134 KB -registry in
parallel only to understand the -matching ABNF...

> But it would suffice to say:
> +++++++++++++++++++++alternative++++++++++++++++++++++
> variant-pattern
>               = variant
>               / "*"                    ; ... or wildcard
> ++++++++++++++++++++++++++++++++++++++++++++++++++
> relying on 'variant' in figure 1 of -registry.

...okay, if we adopt that idea (dito for some other terms),
then a reader simply must get <variant> etc. from -registry.
Maybe "range" is a better name resulting in...

       script-range  = "*" / script
       region-range  = "*" / region
       variant-range = "*" / variant

...we discussed this already, but I forgot what the result was.

                              Bye, Frank

_______________________________________________
Ltru mailing list
Ltru@ietf.org
https://www1.ietf.org/mailman/listinfo/ltru