[Jsdsi-devel] SimpleTag -- ALL_TAG

[Michael J. <mij...@in...>]

Hi,

there has been a lot of silence on the lists the last months.  Today I found 
the time to work on jsdsi and the first thing i wanted to fix was the 
SimpleTag:  As i wrote previously, the combination with the ALL_TAG is 
currently not correct.  As a reminder here is what i wrote:

> I've got a suggestion for JSDSI:  I don't know if I'm right, but maybe we 
> should overthink the implementation of the ALL_TAG.  Currently it's a 
> Tag and not an ExprTag, which means that you can't use it in a SimpleTag.  
In 
> RFC 2693 Ellison et al. quote an example in section 6.5.7 which looks like 
> this:
> 
> (tag (* set (ssl) (dns (*))))
> 
> This examples looks like the ALL_TAG is allowed in a SimpleTag.  What do you 
> think?  In my opinion it makes sense to allow the ALL_TAG in SimpleTags.

As Carl Ellison replied to my question on the SPKI Mailinglist this behaviour 
is wrong and not covered in the BNF.

For my diplomathesis I just changed the implies()-method in the following way, 
which is quite a dirty fix:

	boolean implies(Tag that) {
		// By mic...@in...
		 if( (value.equals("*") && tags.length==0) ) {
			 return true;
		 }

		 if( that instanceof SimpleTag ) {
			 SimpleTag sThat = (SimpleTag) that;
			 if( sThat.getValue().equals(value) && tags.length==sThat.getTags().length) 
{
				 for(int i=0; i<tags.length; i++ ) {
					 if( !tags[i].implies(sThat.tags[i]) ) {
						 return false;
					 }
				 }
				 return true;
			 }
		 }
		 return false;
	}

My suggestion is to solve the problem by refactoring the ALL_TAG.  This could 
otherwise be problematic as this would mean altering the BNF-scheme defined 
by Carl Ellison.  What do you think?

All the best,
Michael.