Re: Big Change

[Andy Clark <an...@apache.org>]

Rahul Srivastava wrote:
> Can somebody step forward and explain these and any other planned 
> changes not reflected in CVS yet.

The removing of the parallel arrays in favor of using a single
array of object structures to hold the data is being considered
for the following reasons:

1) Because of the way in which the data is used (it is semi-
   persistent vs. created and destroyed very quickly), no
   performance difference is seen between the two styles.
2) Management of the arrays is cumbersome and error-prone.
3) The code will become simpler and easier to both extend 
   and maintain.

If we consider the point where we currently are in writing the
new Schema code, it's probably better to make these types of
implementation changes earlier rather than later. XML Schema
is a big, hairy beast to implement so I'm all in favor of
design choices that make the code simpler and easier to 
understand and extend in the future.

-- 
Andy Clark * IBM, TRL - Japan * andyc@apache.org

---------------------------------------------------------------------
To unsubscribe, e-mail: xerces-j-dev-unsubscribe@xml.apache.org
For additional commands, e-mail: xerces-j-dev-help@xml.apache.org