[Xenocide-programming] XML Parser issues

[<red...@pr...>]

Hi Mamutas,
 
I had been researching how big is the memory footprint of Xerces and I hadnt 
found much about it. I think that will be the only thing that would flip the 
coin to or against its adoption. The only thing that I had read was that the 
whole library takes like 8Mb, the bad part is that it didnt said if it was a 8Mb 
.DLL file or 8 Mb of main memory just on initialization... So dont take that as 
a fact.
 
On the other size, there is an important issue that we will have to discuss at 
deep. XML data for the game will be write once, use the rest of the times. If we 
follow the common sense, there is no need to revalidate non modified data each 
time you load an XML file from the sources. Because of that XML validation is 
mostly important only for developers/modders, not for the game users. In that 
case, even if for us that means that we have to run a validation script once in 
a while, just to be sure that we are not messing the files (that can be 
programmed in Java or C++ with Xerces and we can run it everywhere), it is not 
that traumatic.
 
TinyXML is though to be small and have a very small memory footprint, so that 
would be an interesting alternative to a full blown XML Parser like Xerces. And 
we offload the validation into a development only activity.
 
BTW as we talked about SAX Style Parsers are out of the question, we will only 
look for DOM Parsers.
 
So that are our alternatives. If someone else has a comment dont hesitate to 
drop an email to the list too.
 
Greetings
Red Knight