[XML-SIG] SAX for 1.6

Greg Stein gstein@lyra.org
Mon, 17 Apr 2000 17:17:11 -0700 (PDT)

On Mon, 17 Apr 2000, Fred L. Drake, Jr. wrote:
> Lars Marius Garshol writes:
>  > This is more or less the same response I got from David Megginson.
>  > Also, the most controversial parts (DTD declarations and lexical
>  > events) have been moved out into an extension package.
>  > 
>  > So I would say that we should put the core SAX2 stuff in the Python
>  > 1.6 distribution, and then put the extension stuff with more drivers
>  > in the XML-SIG package.
> Lars,
>   This is great news, especicially considering Paul's comments
> regarding the expected lifetimes of the two APIs.
>   +1 on using SAX2's final definition.
>   -0 on using a "convenient" subset of SAX2.

How do you plan to use the final definition if it isn't complete before
Python's ship date?!?

This is why I suggested sticking to SAX1 since we know its final form.

Using a partial SAX2 could simply be confusing "gee, what parts were left
out?" And the argument about SAX1 being dead by the end of the year is
bogus. It is useful technology, and there is still the XML-SIG
distribution to provide additional XML capabilities. An extra ship
vehicle, if you will.

+1 on SAX1
-0 on SAX2 "light"
-1 on SAX2 "final" (impossible)


Greg Stein, http://www.lyra.org/