[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: genx - consensus items

• To: "David Tolpin" <dvd@d...>,<xml-dev@l...>
• Subject: Re: genx - consensus items
• From: "Karl Waclawek" <karl@w...>
• Date: Fri, 23 Jan 2004 09:15:50 -0500
• References: <200401230635.i0N6ZOHM025816@a...>

----- Original Message ----- 
From: "David Tolpin" <dvd@d...>
To: <xml-dev@l...>
Sent: Friday, January 23, 2004 1:35 AM
Subject: Re:  genx - consensus items


> David Tolpin:
> > Rich Salz:
> > > > I'm inclined to agree.  A teeny bit more complexity in the
> > > > implementation but probably worth it.  Anyone want to push back? -Tim
> > > 
> > > asymnmetry is bad.  open-close, begin-end, etc.
> > 
> > The idea was to improve asymmetry, not to defeat it, I think.
> > 
> > startTag
> >   attribute*
> >   text
> > endTag
> > 
> > But in this case startTag is not start 'Tag', it is startElement/endElement.
> > 
> > Not that I like this approach, just explaining. Simple logic is better than
> > complex.
> 
> That is, startElement/endElement are good for building a tree, but inadequate
> for serializing,
> 
> startTagOpen
>   attribute
> startTagClose
> 
>   Text
> 
> endTag
> 
> are simpler and more natural for serialization tasks. I'd keep the startTagClose

Simpler than

startTag
  attribute
  text
endTag

?

Generally one should make an API as small as possible for a given set of features,
otherwise logical redundancy can creep in, with the need for additional state checking 
and preventing inconsistent/contradictory calls.

Karl

• References:
  • Re: genx - consensus items
    • From: David Tolpin <dvd@d...>

• Prev by Date: RE: XLink and mixed vocabulary design
• Next by Date: Re: genx - consensus items
• Previous by thread: implicit StartTagClose [Re: genx - consensus items]
• Next by thread: Re: genx - consensus items
• Index(es):
  • Date
  • Thread