HiveMind does actually not rely on XML's ID type for the various "id"
attributes in module descriptors. The currently enforced pattern for
IDs is "[a-zA-Z0-9_]+". A module's ID may consist of multiple such
segments separated by a dot.

The only non-alphanumeric character currently allowed is underscore.
(So it's effectively equal to the regex word character class "\w".)  I
suppose you'd like to see the colon as a further legal character...
Howard, what's your opinion on this?

--knut

On Wed, 16 Feb 2005 12:55:25 -0700, Dave Smith <[EMAIL PROTECTED]> wrote:
> My reading of the XML spec is that IDs must be a "Name" production.
> Specifically see section 3.3.1, validity constrains on ID:
> 
> "Values of type ID must match the Name production."
> 
> And from section 2.3 of the spec:
> 
> "[Definition:] A Name is a token beginning with a letter or one of a
> few punctuation characters, and continuing with letters, digits,
> hyphens, underscores, colons, or full stops, together known as name
> characters."
> 
> All of the above would imply that hivemind is being unnecessarily
> strict on what IDs may be. Or is there some piece of functionality in
> Hivemind that requires a particular format on those ids?
> 
> D.
> 
> 
> On Wed, 16 Feb 2005 14:45:09 -0300, Marcus Brito <[EMAIL PROTECTED]> wrote:
> > > So, why does Hivemind require that IDs only have alphanumeric
> > > characters? Is there something I'm missing here?
> >
> > Not hivemind per se, but the XML specification. "id" attributes are
> > declared in the DTD as ID types, which, according to the XML
> > specification, should be a string composed only from letters and
> > numbers.
> >
> > -- Marcus Brito
> >
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
> 
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to