Re: [QuantLib-svn] SF.net SVN: quantlib:[16709] branches/R000909-branch/QuantLib/ql/models/ marketmodels

[Luigi Ballabio]

On Nov 6, 2009, at 8:11 PM, [hidden email] wrote:

> Revision: 16709
>          http://quantlib.svn.sourceforge.net/quantlib/?rev=16709&view=rev
> Author:   nando
> Date:     2009-11-06 19:11:18 +0000 (Fri, 06 Nov 2009)
>
> Log Message:
> -----------
> scoped loop variable declaration to avoid warnings if one wants NOT  
> to "disable language extension" (in order to compile with boost 1.40)

Sorry, I'm reverting this one on the 0.9.9 branch---at this point, I'd  
rather not make new tarballs and start another round of tests for a  
couple of warnings in a non-standard configuration.
I applied your changes on the 1.0 branch instead.

Later,
        Luigi



------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
QuantLib-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-dev

[Ferdinando M. Ametrano-3]

On Fri, Nov 6, 2009 at 9:24 PM, Luigi Ballabio <[hidden email]> wrote:
>> scoped loop variable declaration to avoid warnings if one wants NOT to
>> "disable language extension" (in order to compile with boost 1.40)
>
> Sorry, I'm reverting this one on the 0.9.9 branch---at this point, I'd
> rather not make new tarballs and start another round of tests for a couple
> of warnings in a non-standard configuration.
> I applied your changes on the 1.0 branch instead.

mmm... to release 0.9.9 which does not compile with VC 7/8/9 and boost
1.40 is asking for endless questions on the mailing lists.

Besides don't you have to patch the code anyway for VC7 since it does
not compile with boost a1.39 and probably even some earlier version?

I'm trying to compile with 1.40 over the week-end switching "disable
language extensions" to NO.

Of course I'll leave any final judgment to you

ciao -- Nando

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
QuantLib-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-dev

[Plamen Neykov]

Guys,

just as a side info - I managed to compile QuantLib with vs9 and boost 1.40 by just defining BOOST_HAS_LONG_LONG and leaving the language extensions switched on.
Had no side effects so far ...

cheers,
Plamen


On Saturday 07 Nov 2009 09:53:20 Ferdinando Ametrano wrote:

> On Fri, Nov 6, 2009 at 9:24 PM, Luigi Ballabio <[hidden email]> wrote:
> >> scoped loop variable declaration to avoid warnings if one wants NOT to
> >> "disable language extension" (in order to compile with boost 1.40)
> >
> > Sorry, I'm reverting this one on the 0.9.9 branch---at this point, I'd
> > rather not make new tarballs and start another round of tests for a couple
> > of warnings in a non-standard configuration.
> > I applied your changes on the 1.0 branch instead.
>
> mmm... to release 0.9.9 which does not compile with VC 7/8/9 and boost
> 1.40 is asking for endless questions on the mailing lists.
>
> Besides don't you have to patch the code anyway for VC7 since it does
> not compile with boost a1.39 and probably even some earlier version?
>
> I'm trying to compile with 1.40 over the week-end switching "disable
> language extensions" to NO.
>
> Of course I'll leave any final judgment to you
>
> ciao -- Nando
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> QuantLib-dev mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
QuantLib-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-dev

[Luigi Ballabio]

On Nov 7, 2009, at 10:53 AM, Ferdinando Ametrano wrote:
> mmm... to release 0.9.9 which does not compile with VC 7/8/9 and boost
> 1.40 is asking for endless questions on the mailing lists.

Wait a mnute now.  The latest changeset was just avoiding a couple of  
warnings, so it wouldn't solve this anyway.  A change that made the  
library compile would be another matter, but given the choice, I'd  
rather take Plamen's solution than having the compiler change its  
behavior.  If you confirm that the library compiles by adding  
BOOST_HAS_LONG_LONG to the preprocessor definitions, I'd be happy to  
take that change.

Luigi


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
QuantLib-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-dev

[Ferdinando M. Ametrano-3]

On Sat, Nov 7, 2009 at 1:06 PM, Luigi Ballabio <[hidden email]> wrote:
> A change that made the library
> compile would be another matter, but given the choice, I'd rather take
> Plamen's solution than having the compiler change its behavior.  If you
> confirm that the library compiles by adding BOOST_HAS_LONG_LONG to the
> preprocessor definitions, I'd be happy to take that change.

I cannot confirm it until Monday, but I trust Plamen :-) so you might
just make up your mind if you want the define in some header file (my
preference) or the project settings.

Just for the record, it seems to me we changed compiler native
behavior when we switched "disable language extensions" to Yes; anyway
I would stick to that choice.

Finally on Monday I might report if everything compiles switching
"disable language extensions" back to No, and commit a few more code
tweaks to avoid warnings in the compiler native behavior ;-)

BTW what about the patches needed by VC7 ? Don't we include them in 0.9.9 ?

ciao -- Nando

PS C'mon Luigi... a third release candidate is looming...

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
QuantLib-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-dev

[Luigi Ballabio]

On Sat, 2009-11-07 at 20:53 +0100, Ferdinando Ametrano wrote:
> On Sat, Nov 7, 2009 at 1:06 PM, Luigi Ballabio <[hidden email]> wrote:
> I cannot confirm it until Monday, but I trust Plamen :-) so you might
> just make up your mind if you want the define in some header file (my
> preference) or the project settings.

Inside config.msvc.hpp would probably be easiest.


> Just for the record, it seems to me we changed compiler native
> behavior when we switched "disable language extensions" to Yes; anyway
> I would stick to that choice.

We changed the behavior back and forth a few times (the native behavior
being... let me check... disable language extensions = NO for a
newly-created project.)  What made me cringe was changing it right
before release.  I'd be happier if it worked as is.


> Finally on Monday I might report if everything compiles switching
> "disable language extensions" back to No, and commit a few more code
> tweaks to avoid warnings in the compiler native behavior ;-)

Let's touch base first.  If we find out that we can keep them disabled
with Plamen's trick, further tweaks are not necessary and can go in the
1.0 branch.


> BTW what about the patches needed by VC7 ? Don't we include them in 0.9.9 ?

I'm trying to find out what Boost versions are affected.  It worked on
my box with 1.35.  I'm looking at more recent versions.

Later,
        Luigi


--

Academic: a term of opprobrium applied to those that do their job well
by those who cannot.
-- Sir Ernest Gowers



------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
QuantLib-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-dev