You are viewing a plain text version of this content. The canonical link for it is here.
Posted to c-users@xalan.apache.org by John-Mark Newton <jo...@commedica.com> on 2002/09/05 17:38:31 UTC

Building and transforming DOM in memory

Hi,
I'm interested in building a DOM document in memory, element by element and then transforming the result with the XalanTransformer. The output after transformation is immediately dumped to output.

I'm new to Xalan C++ but after looking at the examples and conducting a few experiments I have succeeded in doing this using a Xerces DOM. However, this leaves me with a few questions:

(1) 
I have code already using the Xerces 2.0+ DOM model. Are there any plans to support this in the near future in Xalan?

(2)
The documentation suggests that I'll be taking a performance hit using the Xerces DOM instead of the native Xalan DOM. Is the performance hit big enough to consider serializing the Xerces DOM in memory and then recreating it as a Xalan DOM?

(3)
Does anyone know if the performance hit is due to the smart handle approach taken by Xerces 1.x or are there other factors? I could justify the time to write a Xerces 2 bridge if it would perform well when finished.

(4)
Has anyone got any experience using the Xalan DOM to build a document in memory before passing it to the XalanTransformer? How restrictive is the DOM subset that is implemented? I ran into problems with unimplemented member exceptions pretty quickly but didn't persevere hard to see what was available,

JM
-^-^-^-^-^-^-^-^-^-^-^-^-^-^-^-^-^-^
john-mark.newton@commedica.com