- From: Uche Ogbuji <uche@ogbuji.net>
- Date: Tue, 14 Aug 2012 08:50:51 -0600
- To: public-microxml@w3.org
- Message-ID: <CAPJCua2nhAFRokfsveYSVUHtp6Y7BGbFG81oGbs11tNi6_yN2A@mail.gmail.com>
On Tue, Aug 14, 2012 at 8:39 AM, Murray Maloney <murray@muzmo.com> wrote: > I noticed that attribute values are defined thus... > > attributeValue ::= '"' ((dataChar - '"') | charRef)* '"' > | "'" ((dataChar - "'") | charRef)* "'" > > So, attributes may take the form foo="bar" or foo='bar' but not foo=bar > > I further note that this definition does not distinguish between strings > and tokens, by which I mean those sequences which contain spaces and those > which don't. > > It seems that data types are not in scope for uXML, but I think that it > could still be helpful to distinguish among strings and tokens, so as to > allow foo=bar as a legitimate form of attribute value assignation. > Unfortunately foo=bar syntax would break backward syntactical compatibility pretty badly. FWIW I do also wish XML 1.0 had supported that syntax, rather than just giving lip service to NMTOKENS. > Frankly, I would like to see data types baked into XML, by way of a > corrigendum if necessary, but that is another discussion. > Hmm. Remember that we are likely to develop a separate, schematic layer for MicroXML, perhaps based on Examplotron or RELAX NG. Don't you think it would be better to leave (very lightweight and generic) data typing to that layer? -- Uche Ogbuji http://uche.ogbuji.net Founding Partner, Zepheira http://zepheira.com http://wearekin.org http://www.thenervousbreakdown.com/author/uogbuji/ http://copia.ogbuji.net http://www.linkedin.com/in/ucheogbuji http://twitter.com/uogbuji
Received on Tuesday, 14 August 2012 14:51:20 UTC