Re: C++11 on master (towards LO 4.5)

2014-12-09 Thread Jan Holesovsky
Hi, Stephan Bergmann píše v Pá 28. 11. 2014 v 09:22 +0100: > Maintainers of active Windows tinderboxes (on cc, as of > ), > > please make sure until, say, mid of next week (Dec 3) that they are > using M

Re: C++11 on master (towards LO 4.5)

2014-12-05 Thread Andras Timar
On Fri, Dec 5, 2014 at 11:49 AM, Stephan Bergmann wrote: > On 12/04/2014 09:55 PM, Terrence Enger wrote: >> >> It's not a big deal for me (because I am just looking for trouble, >> not trying to get work done), but maybe somebody cares: my attempt to >> run LibreOffice installed from (I am retypin

Re: C++11 on master (towards LO 4.5)

2014-12-05 Thread Stephan Bergmann
On 12/04/2014 09:55 PM, Terrence Enger wrote: It's not a big deal for me (because I am just looking for trouble, not trying to get work done), but maybe somebody cares: my attempt to run LibreOffice installed from (I am retyping this with newlines added) http://dev-build.libreoffice.org/da

Re: C++11 on master (towards LO 4.5)

2014-12-04 Thread Terrence Enger
On Wed, 2014-12-03 at 15:35 +0100, Stephan Bergmann wrote: > On 11/28/2014 09:22 AM, Stephan Bergmann wrote: > > Maintainers of active Windows tinderboxes (on cc, as of > > ), > > please make sure until, say,

Re: C++11 on master (towards LO 4.5)

2014-12-04 Thread Thorsten Behrens
Stephan Bergmann wrote: > Maintainers of active Windows tinderboxes (on cc, as of > ), > please make sure until, say, mid of next week (Dec 3) that they are > using MSVC 2013 for master builds. > To make it ex

Re: C++11 on master (towards LO 4.5)

2014-12-03 Thread Stephan Bergmann
On 11/28/2014 09:22 AM, Stephan Bergmann wrote: Maintainers of active Windows tinderboxes (on cc, as of ), please make sure until, say, mid of next week (Dec 3) that they are using MSVC 2013 for master build

Re: C++11 on master (towards LO 4.5)

2014-12-03 Thread Stephan Bergmann
On 12/03/2014 07:39 AM, julien2412 wrote: Will we be able to consider that erase method of every container (not just some of them) will return an iterator so we won't need a temporary iterator when using erase in a loop block? (eg: http://www.cplusplus.com/reference/map/map/erase/) The only ans

Re: C++11 on master (towards LO 4.5)

2014-12-02 Thread julien2412
Hi, Will we be able to consider that erase method of every container (not just some of them) will return an iterator so we won't need a temporary iterator when using erase in a loop block? (eg: http://www.cplusplus.com/reference/map/map/erase/) Julien -- View this message in context: http://n

Re: C++11 on master (towards LO 4.5)

2014-11-28 Thread Stephan Bergmann
On 11/25/2014 09:13 AM, Stephan Bergmann wrote: So dropping MSVC 2012 for 2013 would give us six new features: * Variadic templates * Initializer lists * Default template argumetns for function templates * Explicit conversion operators * Raw string literals * Defaulted and deleted functions And

Re: C++11 on master (towards LO 4.5)

2014-11-26 Thread Michael Stahl
On 25.11.2014 09:13, Stephan Bergmann wrote: > Until now, our minimum C++ compiler requirements on master are: > >* Clang 3.1 >* GCC 4.6 (specifically for Munich; preferred 4.7) >* MSVC 2012 > > Branch-off of LO 4.4 from master seems like a good time to re-visit. > And natural candid

Re: C++11 on master (towards LO 4.5)

2014-11-26 Thread Jan-Marek Glogowski
Am 25.11.2014 um 09:13 schrieb Stephan Bergmann: > Until now, our minimum C++ compiler requirements on master are: > > * Clang 3.1 > * GCC 4.6 (specifically for Munich; preferred 4.7) > * MSVC 2012 > > Branch-off of LO 4.4 from master seems like a good time to re-visit. And > natural candi

Re: C++11 on master (towards LO 4.5)

2014-11-25 Thread Stephan Bergmann
On 11/25/2014 04:34 PM, Terrence Enger wrote: The range-based `for` offers clearer code, IMHO. When, I wonder, is it appropriate to use it? (*) New code? (*) When I am changing a for-statement anyway? (*) Throughout a function when I am changing a for-statement anyway? (*) Not at all until we ar

Re: C++11 on master (towards LO 4.5)

2014-11-25 Thread Terrence Enger
On Tue, 2014-11-25 at 09:13 +0100, Stephan Bergmann wrote: interesting stuff about features of C++ supported by various compilers. The range-based `for` offers clearer code, IMHO. When, I wonder, is it appropriate to use it? (*) New code? (*) When I am changing a for-statement anyway? (*) Through

Re: C++11 on master (towards LO 4.5)

2014-11-25 Thread Tor Lillqvist
> Thoughts? > I am all for it! I already tried using initializer lists (which IMHO seem quite useful and not "obscure" at all;) but was then saddened to notice that MSVS2012 did not support them. --tml ___ LibreOffice mailing list LibreOffice@lists.free

C++11 on master (towards LO 4.5)

2014-11-25 Thread Stephan Bergmann
Until now, our minimum C++ compiler requirements on master are: * Clang 3.1 * GCC 4.6 (specifically for Munich; preferred 4.7) * MSVC 2012 Branch-off of LO 4.4 from master seems like a good time to re-visit. And natural candidates for consideration appear to be dropping MSVC 2012 for 201