[Bf-committers] Blender XML

Jonathan Merritt j.merritt at pgrad.unimelb.edu.au
Mon Feb 14 02:44:03 CET 2005


joeedh wrote:

> Jeremy Wall wrote:
>
>> I am beginning testing of the XML parse. I plan to run a series of
>> tests on tree creation and text parsing. Once that is done I will be
>>
> >> <SNIP>
>
> So, do you plan to make an XML parser based on the current DNA 
> system?  Or, in other words, will your xml files simply contain xml 
> versions of DNA structs?


... and will it be smart enough to adaptively change when the DNA system 
changes (like the current binary system :-) ?  It just concerns me that 
the description sounded like a component-wise approach was being taken, 
whereas an adaptive system should work in a "one size fits all" kind of way.

What's wrong with simply "textualizing" the current binary system.  An 
XML DTD or schema could be generated from the header files (heck, this 
could even be done using Python's powerful text processing capabilities 
these days), as could the C code to parse or generate the XML itself.  
It should be possible to name various bits of the XML (I guess using 
attributes) after parts of the C structs, thereby making the XML file 
format as readable as the original code.  Is something like this 
planned, or will the XML file format be an additional tight dependency 
that needs to be changed whenever something is added or changed?

-- 
Jonathan Merritt BE(Mech)/BSc
PhD Student - Equine Biomechanics
The University of Melbourne
Veterinary Clinical Centre, Werribee



More information about the Bf-committers mailing list