W3C home > Mailing lists > Public > public-exi@w3.org > March 2006

Re: [xml-dev] Call for fast XML implementations

From: Phil Nash <phil.nash.lists@gmail.com>
Date: Fri, 24 Mar 2006 18:35:14 +0000
Message-ID: <44243BE2.9080306@gmail.com>
To: Robin Berjon <robin.berjon@expway.fr>
CC: XML Developers List <xml-dev@lists.xml.org>, W3C EXI Public <public-exi@w3.org>

Robin Berjon wrote:
>  * If you wish to be extra helpful, you can also include the small 
> amount of code and configuration required to run the parser within our 
> framework, which is built on top of Japex (https://japex.dev.java.net/). 
> If you're interested, we'd be delighted to help you get started.

Does this imply that only java parsers will be considered?
I've had a parser project in C++ on my shelf for a while now and am just 
turning back to look at it. It's heavily template based, using policies 
to turn on and off various types of support that may not be needed (and 
would otherwise inhibit performance) at compilation time. This means 
that, while it will be compliant (that is the aim), you can deliberately 
remove some features to squeeze more speed out if you know your use 
cases don't require it (e.g. various levels of validation, certain types 
of error handling, string conversions etc).
By using templates I also avoid the overhead of virtual functions, which 
may or may not be significant, but I suspect it could be, in view of the 
number of them used in a typical transaction.

As it is the implementation is a way off complete yet. What sort of time 
scales are we looking at?

Best regards,

[)o
IhIL..
Received on Saturday, 25 March 2006 05:18:36 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:52:42 UTC