> Hi Bela,
> 
> sorry, for the moment you can not improve this. The problem is
> in resolving symbols. Xref actually needs to parse the whole
> file with all headers in order to catch the symbol. We have 
> several ideas how to improve those things in future versions, 
> however for the moment, we are still overloaded by fixing bugs. 
> Also the problem seems to appear especially under Windows. I do
> not know where it resides, but Xrefactory is considerably slower
> under Windows than under any Unix like system on the same hardware.
> Anyway, I am sure, that this problem will be improved, but probably
> not in the first commercial version.
> 
> Marian
> 

Hi Marian,

Thanks for the quick reply.
Your guess was right... it was a Windows system I tested on. Sorry
not mentioning it.
If I'll have time I will try to run the code under a profiler
to try to nail down the performance bottleneck. If anything turns
out I'll get back to you with the results.

However this also could be considered as a bug ;)

Regards, 
 Bela