[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Patches to enable MSVC compilation
From: |
John W. Eaton |
Subject: |
Re: Patches to enable MSVC compilation |
Date: |
Tue, 8 Dec 2009 16:18:09 -0500 |
On 8-Dec-2009, Jaroslav Hajek wrote:
| The problem is that an #ifdef would be needed at each call, which
| means a workaround spread across a number of sources, and I think
| that's not good.
| Ummm. This problem would probably not exist if the Array class
| hierarchy was more linear, so that the appliers could be parametrized
| by value types rather than array types.
What about doing something like
#if defined (AVOID_MSVC_FOO_BUG)
#define DO_MX_UMARY_MAP(T1, T2, F) \
do_mx_unary_map<T1, T2> (*this, F)
#else
#define DO_MX_UMARY_MAP(T1, T2, F) \
do_mx_unary_map<T1, T2, F> (*this)
#fi
? Would that work? Doesn't that allow the difference to be in one
header file and also preserve the possibility for better inlining on
non-MSVC systems?
| Here's an idea: what about including the compatibility patches
| directly in Octave's sources?
That would be OK with me.
jwe
- Re: Patches to enable MSVC compilation, (continued)
- Re: Patches to enable MSVC compilation, Jaroslav Hajek, 2009/12/08
- Re: Patches to enable MSVC compilation, Michael Goffioul, 2009/12/08
- Re: Patches to enable MSVC compilation, John W. Eaton, 2009/12/08
- Re: Patches to enable MSVC compilation, Jaroslav Hajek, 2009/12/10
- Re: Patches to enable MSVC compilation, John W. Eaton, 2009/12/10
- Re: Patches to enable MSVC compilation, Jaroslav Hajek, 2009/12/10
- Re: Patches to enable MSVC compilation, John W. Eaton, 2009/12/10
- Re: Patches to enable MSVC compilation, Jaroslav Hajek, 2009/12/10
- Re: Patches to enable MSVC compilation, John W. Eaton, 2009/12/08
- Re: Patches to enable MSVC compilation, Jaroslav Hajek, 2009/12/08
- Re: Patches to enable MSVC compilation,
John W. Eaton <=
- Re: Patches to enable MSVC compilation, Michael Goffioul, 2009/12/08
- Re: Patches to enable MSVC compilation, Benjamin Lindner, 2009/12/09