Re: [Boost-testing] Too many results?

From: Sebastian Redl (sebastian.redl_at_[hidden])
Date: 2007-09-28 05:13:08

Rene Rivera wrote:
> How likely is it you can improve on the performance? And with how much 
> effort?
>   
Well, as I have never measured the performance of XSLT before, I can't
reliably optimize it. Though I have a pretty good idea of at least one
part of the style sheet that has to be very wasteful on performance.
However, changing that would essentially mean rewriting the style sheet
from scratch. So, that would be very much effort with dubious results -
I wouldn't for sure in advance if my version would be faster.

Would almost be smarter then to rewrite the thing in C++ instead. But
you know how it is - it's a new tool, it needs to be written, tested,
verified, etc. The effort would be just as great if not greater. And all
that for a tool we want to get rid of anyway.

Sebastian