Re: SAX and whitespace (was Re: Problems with whitespace and

Tim Bray (tbray@textuality.com)
Fri, 02 Jan 1998 17:35:58 -0800


At 03:42 PM 02/01/98 -0500, David Megginson wrote:
>With that in mind, would it be conformant behaviour for SAX to report
>ignorable whitespace as regular character data when the parser
>underneath is using the DTD (and, say, supplying defaulted attribute
>values)?

Lark, BTW, does *not* catch ignorable white space unless it is
validating. Since it is perfectly OK to build SAX with such a
processor, *if* we want to build ignorable white-space notification
into SAX, it has to be out-of-band; i.e. white space is passed in
the same way as all other content; with perhaps another boolean argument
to the text() method (that what it's called now?) that if true, means
this is ignorable white space.

But I would oppose doing this in SAX; let's keep it simple for now. -T.