Re: [YANG] 7.1.5 import statement

Juergen Schoenwaelder wrote:
> On Wed, Jan 09, 2008 at 08:34:19AM -0800, Andy Bierman wrote:
> 
>> This section does not say anything about whether it is an
>> error to import a non-existent module, if it is not used.
>> The pyang program reports this as a fatal error.
>>
>> This gets into implementation details a little.
>> A well-formed import that never gets used should be a warning.
>> Consider implementations that do not actually find and compile
>> the module for an import clause until the first time the prefix is used.
> 
> I expect a compiler to complain if an import can not be found and I
> expect a compiler to generate a warning if there is an import that is
> not used. In general, we don't talk much about warnings so we might
> choose to leave the generation of a warning for unused imports also an
> implementation detail.
> 

Fine.  One compiler complains and continues, and another complains
and exits.

It would be nice if DM writers who invest time and money in YANG
know what 'valid YANG' is, so they know what tools should accept or not.
But we should not over-specify the DML, such as defining compiler warnings.

> /js
> 

Andy

_______________________________________________
YANG mailing list
YANG@ietf.org
https://www1.ietf.org/mailman/listinfo/yang