Domeika, Max J wrote:
> I'm going to have to delay trying to get Intel Compilers on the Blitz++
> supported list due to two reasons:
> 1. We won't have a Linux product for ~6 months.
> 2. The Blitz build scripts are not NT friendly and thus not easily usable
> by our NT compiler. I don't have the time to scratch and claw through it.
>
> I will get back in touch once our Linux product gets closer to being
> available. It works with it, but no customer can buy it yet.
>
> In the mean time, I would like to request a change made to the file
> random/mt.h.
>
> The problem is that iterators are not implicitly convertible to pointers.
>
> Any chance you can get the change in?
>
> Thanks,
>
> Max
>
Done, thanks for the fix.
Cheers,
Todd
[tveldhui@ulthing random]$ bcvs diff mt.h
--------------------- blitz-dev list --------------------------------
* To subscribe/unsubscribe: use the handy web form at
http://oonumerics.org/blitz/lists.html
This archive was generated by hypermail 2b29
: Wed Feb 20 2002 - 04:30:12 EST
Index: mt.h
===================================================================
RCS file: /cvsroot/Blitz++/random/mt.h,v
retrieving revision 1.1.1.1
diff -r1.1.1.1 mt.h
40a41,46
> /*
> * $Id: 0450.html,v 1.1.1.1 2005/06/15 14:30:51 tveldhui Exp $
> *
> * $Log: 0450.html,v $
> * Revision 1.1.1.1 2005/06/15 14:30:51 tveldhui
> *
> *
> */
>
98c104
< I = &S[N];
---
> I = S.end();
135c141
< Iter s = &S[0];
---
> Iter s = S.begin();
153c159
< Iter p0 = &S[0];
---
> Iter p0 = S.begin();
157c163
< for (Iter pf_end = &S[N-PF]; p0 != pf_end; ++p0, ++pM)
---
> for (Iter pf_end = (Iter)&S[N-PF]; p0 != pf_end; ++p0, ++pM)
160c166
< for (Iter s_end = &S[N-1]; p0 != s_end; ++p0, ++pM)
---
> for (Iter s_end = (Iter)&S[N-1]; p0 != s_end; ++p0, ++pM)
164c170
< I = &S[0];
---
> I = S.begin();