Re: sming wg agenda proposal
Andy Bierman <[email protected]> Thu, 14 Nov 2002 04:05:35 -0800
| Newsgroups | gmane.ietf.sming |
|---|---|
| Message-ID | <[email protected]> |
At 10:10 AM 11/14/2002 +0100, Frank Strauss wrote: >Hi! > >Andy> [...] I would also like to add clauses that make it easier for >Andy> tools to generate XML Schema documents from SMIv3 MIBs: > >Andy> XML-NAMESPACE "<URI>" would be a clause added to the MODULE-IDENTITY >Andy> macro to associate a namespace with the MIB module. > >I think I'd like this. ;-) > >Andy> XML-NAME "<string>" would be a clause added to the ATTRIBUTE clause >Andy> (or higher level) of a TYPE or VAR macro to associate an XML element >Andy> name with an object. If absent, then the object descriptor would >Andy> define the element name. > >Could you please explain or give an example why/where this XML-NAME >deviation from the descriptor would be useful? In general, I would prefer >to have "less chances for exceptions". MIB descriptors tend to be hard-to-read, almost random abbreviations, in an attempt to create globally unique identifiers with <= 32 chars. I want to allow for simpler XML names that are only unique within the scope of their parent element. XML element names can have hyphens and SMI descriptors can't. >Andy> There are probably other details that need to added to >Andy> unambiguously convert an SMIv3 MIB to an XSD, to be determined >Andy> later. > > -frank Andy