[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Mon, 20 Jan 2003 00:25:38 +0900
Gopal, sorry to say, I changed the decision again to go ahead. Miguel de
Icaza said that it would be better to go ahead, and I hope he is right.
I understand, however, that sometimes, it'd be better to discuss to
determine what is better of the specification. Anyway, I hope, all of
.NET implementors would go better, compatible with each other.
> Hmm... MS compat ... Adam's Read() thingy is supposed work like MS .NET
> because the ECMA spec is equivocal... It's open for interpretation as to
> what the word ``node'' means .. MS has apparently ignored attributes as
Sure. IMHO, Read() had better ignore attribute nodes and attributes had
better parsed already in parsing element. It is like startElement of SAX.
XmlReader's processing model of Read() is different from that of
MoveTo*Attribute() and ReadAttributeValue(). Taking those existence
into consideration, I think the ECMA's Read() specification "Moves the
position of the current instance to the next node in the stream,
exposing its properties" is nonsense. At least it is "linked node"
rather than "node."
Not always, but .NET implementation is more stable than ECMA
specification which is based on their buggy documentation.
I think, *current* ECMA specification is not worth respecting.
-- Atsushi Eno