Re: [sage-devel] Consider not upgrading to Xcode 9.0

2017-09-20 Thread François Bissey 
gfortran 7.2.0 builds out of the box. Don’t know about full blown gcc.

François

> On 21/09/2017, at 10:44, John H Palmieri  wrote:
> 
> Trying to build gfortran yields the same error as reported at 
> https://trac.sagemath.org/ticket/23898.
> 
>   John
> 
> 
> On Wednesday, September 20, 2017 at 3:08:05 PM UTC-7, John H Palmieri wrote:
> I successfully built with the current version (not the updated one) of eclib 
> plus the patch you mentioned. I will also try to see if gfortran builds.
> 
> On Wednesday, September 20, 2017 at 2:56:59 PM UTC-7, François Bissey wrote:
> So I have a build of sage 8.1.beta5 on OS X with Xcode 9 using 
> 
> * #12426 - building with clang - you need autotools installed or generate 
> configure 
> somewhere that does and copy it in place. 
> 
> * an external fortran compiler, given the problems with gcc it may 
> very well be that gfortran won’t build either - I need to further check that. 
> 
> * an updated eclib 
> https://github.com/JohnCremona/eclib/archive/v20170815.tar.gz 
> probably not necessary but defaulting to C++11 make logs much nicer since 
> it cuts down on the warnings. Beware this is not a sdist tarball. 
> 
> * the patch at 
> https://github.com/JohnCremona/eclib/issues/19#issuecomment-314952542 
> this is absolutely needed. 
> 
> François 
> 
> > On 21/09/2017, at 07:10, François Bissey  wrote: 
> > 
> > We in fact have a patch on github for a similar problem that was spotted 
> > before. It proved effective in this instance. So I will proceed with trying 
> > to build the rest of sage. 
> > 
> > François 
> > 
> >> On 21/09/2017, at 04:05, John H Palmieri  wrote: 
> >> 
> >> I put the log for that failure at #12426, and then François created an 
> >> upstream ticket at 
> >> 
> >>   https://github.com/JohnCremona/eclib/issues/28 
> >> 
> >> in case you want to take a look. 
> >> 
> >> 
> >> On Wednesday, September 20, 2017 at 3:08:13 AM UTC-7, Dima Pasechnik 
> >> wrote: 
> >> The eclib does not look like a big problem, some rather standard C++ 
> >> template issue to be tweaked... 
> >> 
> >> On Wednesday, September 20, 2017 at 12:50:32 AM UTC+1, John H Palmieri 
> >> wrote: 
> >> I just upgraded an OS X box to Xcode 9.0, and now Sage doesn't build: 
> >> 
> >> - with a fresh Sage 8.1.beta5 tarball, gcc doesn't build: 
> >> 
> >> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/functional:1398:2:
> >>  error: no member named 'fancy_abort' in namespace 'std::__1'; did you 
> >> mean simply 'fancy_abort'? 
> >>_VSTD::abort(); 
> >>^~~ 
> >> 
> >> I'll open a ticket and post the full gcc log. 
> >> 
> >> - instead using trac 12426 (compile with clang instead of gcc), eclib 
> >> doesn't build: 
> >> 
> >> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:572:12:
> >>  error: call to function 'isnan' that is neither visible in the template 
> >> definition nor found by argument-dependent lookup 
> >>return isnan(__lcpp_x); 
> >> 
> >>  
> >> 
> >> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:592:12:
> >>  error: call to function 'isinf' that is neither visible in the template 
> >> definition nor found by argument-dependent lookup 
> >>return isinf(__lcpp_x); 
> >> 
> >> -- 
> >> John 
> >> 
> >> 
> >> -- 
> >> You received this message because you are subscribed to the Google Groups 
> >> "sage-devel" group. 
> >> To unsubscribe from this group and stop receiving emails from it, send an 
> >> email to sage-devel+...@googlegroups.com. 
> >> To post to this group, send email to sage-...@googlegroups.com. 
> >> Visit this group at https://groups.google.com/group/sage-devel. 
> >> For more options, visit https://groups.google.com/d/optout. 
> > 
> 
> 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Consider not upgrading to Xcode 9.0

2017-09-20 Thread François Bissey 
Trying non gfortran fortran compiler is on my ToDo list. Not necessarily
intel though. You can get community PG compilers for free as well.
Not tried pgcc yet though.

François

> On 21/09/2017, at 11:15, Dima Pasechnik  wrote:
> 
> I wonder if anyone tried lately using Intel's Fortran on OSX, I understand 
> that with a university affiliation one can get it for free.
> https://software.intel.com/en-us/parallel-studio-xe/choose-download/educator-mac-c
> 
> On Wednesday, September 20, 2017 at 11:44:50 PM UTC+1, John H Palmieri wrote:
> Trying to build gfortran yields the same error as reported at 
> https://trac.sagemath.org/ticket/23898.
> 
>   John
> 
> 
> On Wednesday, September 20, 2017 at 3:08:05 PM UTC-7, John H Palmieri wrote:
> I successfully built with the current version (not the updated one) of eclib 
> plus the patch you mentioned. I will also try to see if gfortran builds.
> 
> On Wednesday, September 20, 2017 at 2:56:59 PM UTC-7, François Bissey wrote:
> So I have a build of sage 8.1.beta5 on OS X with Xcode 9 using 
> 
> * #12426 - building with clang - you need autotools installed or generate 
> configure 
> somewhere that does and copy it in place. 
> 
> * an external fortran compiler, given the problems with gcc it may 
> very well be that gfortran won’t build either - I need to further check that. 
> 
> * an updated eclib 
> https://github.com/JohnCremona/eclib/archive/v20170815.tar.gz 
> probably not necessary but defaulting to C++11 make logs much nicer since 
> it cuts down on the warnings. Beware this is not a sdist tarball. 
> 
> * the patch at 
> https://github.com/JohnCremona/eclib/issues/19#issuecomment-314952542 
> this is absolutely needed. 
> 
> François 
> 
> > On 21/09/2017, at 07:10, François Bissey  wrote: 
> > 
> > We in fact have a patch on github for a similar problem that was spotted 
> > before. It proved effective in this instance. So I will proceed with trying 
> > to build the rest of sage. 
> > 
> > François 
> > 
> >> On 21/09/2017, at 04:05, John H Palmieri  wrote: 
> >> 
> >> I put the log for that failure at #12426, and then François created an 
> >> upstream ticket at 
> >> 
> >>   https://github.com/JohnCremona/eclib/issues/28 
> >> 
> >> in case you want to take a look. 
> >> 
> >> 
> >> On Wednesday, September 20, 2017 at 3:08:13 AM UTC-7, Dima Pasechnik 
> >> wrote: 
> >> The eclib does not look like a big problem, some rather standard C++ 
> >> template issue to be tweaked... 
> >> 
> >> On Wednesday, September 20, 2017 at 12:50:32 AM UTC+1, John H Palmieri 
> >> wrote: 
> >> I just upgraded an OS X box to Xcode 9.0, and now Sage doesn't build: 
> >> 
> >> - with a fresh Sage 8.1.beta5 tarball, gcc doesn't build: 
> >> 
> >> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/functional:1398:2:
> >>  error: no member named 'fancy_abort' in namespace 'std::__1'; did you 
> >> mean simply 'fancy_abort'? 
> >>_VSTD::abort(); 
> >>^~~ 
> >> 
> >> I'll open a ticket and post the full gcc log. 
> >> 
> >> - instead using trac 12426 (compile with clang instead of gcc), eclib 
> >> doesn't build: 
> >> 
> >> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:572:12:
> >>  error: call to function 'isnan' that is neither visible in the template 
> >> definition nor found by argument-dependent lookup 
> >>return isnan(__lcpp_x); 
> >> 
> >>  
> >> 
> >> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:592:12:
> >>  error: call to function 'isinf' that is neither visible in the template 
> >> definition nor found by argument-dependent lookup 
> >>return isinf(__lcpp_x); 
> >> 
> >> -- 
> >> John 
> >> 
> >> 
> >> -- 
> >> You received this message because you are subscribed to the Google Groups 
> >> "sage-devel" group. 
> >> To unsubscribe from this group and stop receiving emails from it, send an 
> >> email to sage-devel+...@googlegroups.com. 
> >> To post to this group, send email to sage-...@googlegroups.com. 
> >> Visit this group at https://groups.google.com/group/sage-devel. 
> >> For more options, visit https://groups.google.com/d/optout. 
> > 
> 
> 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group,

Re: [sage-devel] Consider not upgrading to Xcode 9.0

2017-09-20 Thread Dima Pasechnik 
I wonder if anyone tried lately using Intel's Fortran on OSX, I understand 
that with a university affiliation one can get it for free.
https://software.intel.com/en-us/parallel-studio-xe/choose-download/educator-mac-c

On Wednesday, September 20, 2017 at 11:44:50 PM UTC+1, John H Palmieri 
wrote:
>
> Trying to build gfortran yields the same error as reported at 
> https://trac.sagemath.org/ticket/23898.
>
>   John
>
>
> On Wednesday, September 20, 2017 at 3:08:05 PM UTC-7, John H Palmieri 
> wrote:
>>
>> I successfully built with the current version (not the updated one) of 
>> eclib plus the patch you mentioned. I will also try to see if gfortran 
>> builds.
>>
>> On Wednesday, September 20, 2017 at 2:56:59 PM UTC-7, François Bissey 
>> wrote:
>>>
>>> So I have a build of sage 8.1.beta5 on OS X with Xcode 9 using 
>>>
>>> * #12426 - building with clang - you need autotools installed or 
>>> generate configure 
>>> somewhere that does and copy it in place. 
>>>
>>> * an external fortran compiler, given the problems with gcc it may 
>>> very well be that gfortran won’t build either - I need to further check 
>>> that. 
>>>
>>> * an updated eclib 
>>> https://github.com/JohnCremona/eclib/archive/v20170815.tar.gz 
>>> probably not necessary but defaulting to C++11 make logs much nicer 
>>> since 
>>> it cuts down on the warnings. Beware this is not a sdist tarball. 
>>>
>>> * the patch at 
>>> https://github.com/JohnCremona/eclib/issues/19#issuecomment-314952542 
>>> this is absolutely needed. 
>>>
>>> François 
>>>
>>> > On 21/09/2017, at 07:10, François Bissey  wrote: 
>>> > 
>>> > We in fact have a patch on github for a similar problem that was 
>>> spotted 
>>> > before. It proved effective in this instance. So I will proceed with 
>>> trying 
>>> > to build the rest of sage. 
>>> > 
>>> > François 
>>> > 
>>> >> On 21/09/2017, at 04:05, John H Palmieri  
>>> wrote: 
>>> >> 
>>> >> I put the log for that failure at #12426, and then François created 
>>> an upstream ticket at 
>>> >> 
>>> >>   https://github.com/JohnCremona/eclib/issues/28 
>>> >> 
>>> >> in case you want to take a look. 
>>> >> 
>>> >> 
>>> >> On Wednesday, September 20, 2017 at 3:08:13 AM UTC-7, Dima Pasechnik 
>>> wrote: 
>>> >> The eclib does not look like a big problem, some rather standard C++ 
>>> template issue to be tweaked... 
>>> >> 
>>> >> On Wednesday, September 20, 2017 at 12:50:32 AM UTC+1, John H 
>>> Palmieri wrote: 
>>> >> I just upgraded an OS X box to Xcode 9.0, and now Sage doesn't build: 
>>> >> 
>>> >> - with a fresh Sage 8.1.beta5 tarball, gcc doesn't build: 
>>> >> 
>>> >> 
>>> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/functional:1398:2:
>>>  
>>> error: no member named 'fancy_abort' in namespace 'std::__1'; did you mean 
>>> simply 'fancy_abort'? 
>>> >>_VSTD::abort(); 
>>> >>^~~ 
>>> >> 
>>> >> I'll open a ticket and post the full gcc log. 
>>> >> 
>>> >> - instead using trac 12426 (compile with clang instead of gcc), eclib 
>>> doesn't build: 
>>> >> 
>>> >> 
>>> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:572:12:
>>>  
>>> error: call to function 'isnan' that is neither visible in the template 
>>> definition nor found by argument-dependent lookup 
>>> >>return isnan(__lcpp_x); 
>>> >> 
>>> >>  
>>> >> 
>>> >> 
>>> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:592:12:
>>>  
>>> error: call to function 'isinf' that is neither visible in the template 
>>> definition nor found by argument-dependent lookup 
>>> >>return isinf(__lcpp_x); 
>>> >> 
>>> >> -- 
>>> >> John 
>>> >> 
>>> >> 
>>> >> -- 
>>> >> You received this message because you are subscribed to the Google 
>>> Groups "sage-devel" group. 
>>> >> To unsubscribe from this group and stop receiving emails from it, 
>>> send an email to sage-devel+...@googlegroups.com. 
>>> >> To post to this group, send email to sage-...@googlegroups.com. 
>>> >> Visit this group at https://groups.google.com/group/sage-devel. 
>>> >> For more options, visit https://groups.google.com/d/optout. 
>>> > 
>>>
>>>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Consider not upgrading to Xcode 9.0

2017-09-20 Thread John H Palmieri 
Trying to build gfortran yields the same error as reported at 
https://trac.sagemath.org/ticket/23898.

  John


On Wednesday, September 20, 2017 at 3:08:05 PM UTC-7, John H Palmieri wrote:
>
> I successfully built with the current version (not the updated one) of 
> eclib plus the patch you mentioned. I will also try to see if gfortran 
> builds.
>
> On Wednesday, September 20, 2017 at 2:56:59 PM UTC-7, François Bissey 
> wrote:
>>
>> So I have a build of sage 8.1.beta5 on OS X with Xcode 9 using 
>>
>> * #12426 - building with clang - you need autotools installed or generate 
>> configure 
>> somewhere that does and copy it in place. 
>>
>> * an external fortran compiler, given the problems with gcc it may 
>> very well be that gfortran won’t build either - I need to further check 
>> that. 
>>
>> * an updated eclib 
>> https://github.com/JohnCremona/eclib/archive/v20170815.tar.gz 
>> probably not necessary but defaulting to C++11 make logs much nicer since 
>> it cuts down on the warnings. Beware this is not a sdist tarball. 
>>
>> * the patch at 
>> https://github.com/JohnCremona/eclib/issues/19#issuecomment-314952542 
>> this is absolutely needed. 
>>
>> François 
>>
>> > On 21/09/2017, at 07:10, François Bissey  wrote: 
>> > 
>> > We in fact have a patch on github for a similar problem that was 
>> spotted 
>> > before. It proved effective in this instance. So I will proceed with 
>> trying 
>> > to build the rest of sage. 
>> > 
>> > François 
>> > 
>> >> On 21/09/2017, at 04:05, John H Palmieri  wrote: 
>> >> 
>> >> I put the log for that failure at #12426, and then François created an 
>> upstream ticket at 
>> >> 
>> >>   https://github.com/JohnCremona/eclib/issues/28 
>> >> 
>> >> in case you want to take a look. 
>> >> 
>> >> 
>> >> On Wednesday, September 20, 2017 at 3:08:13 AM UTC-7, Dima Pasechnik 
>> wrote: 
>> >> The eclib does not look like a big problem, some rather standard C++ 
>> template issue to be tweaked... 
>> >> 
>> >> On Wednesday, September 20, 2017 at 12:50:32 AM UTC+1, John H Palmieri 
>> wrote: 
>> >> I just upgraded an OS X box to Xcode 9.0, and now Sage doesn't build: 
>> >> 
>> >> - with a fresh Sage 8.1.beta5 tarball, gcc doesn't build: 
>> >> 
>> >> 
>> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/functional:1398:2:
>>  
>> error: no member named 'fancy_abort' in namespace 'std::__1'; did you mean 
>> simply 'fancy_abort'? 
>> >>_VSTD::abort(); 
>> >>^~~ 
>> >> 
>> >> I'll open a ticket and post the full gcc log. 
>> >> 
>> >> - instead using trac 12426 (compile with clang instead of gcc), eclib 
>> doesn't build: 
>> >> 
>> >> 
>> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:572:12:
>>  
>> error: call to function 'isnan' that is neither visible in the template 
>> definition nor found by argument-dependent lookup 
>> >>return isnan(__lcpp_x); 
>> >> 
>> >>  
>> >> 
>> >> 
>> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:592:12:
>>  
>> error: call to function 'isinf' that is neither visible in the template 
>> definition nor found by argument-dependent lookup 
>> >>return isinf(__lcpp_x); 
>> >> 
>> >> -- 
>> >> John 
>> >> 
>> >> 
>> >> -- 
>> >> You received this message because you are subscribed to the Google 
>> Groups "sage-devel" group. 
>> >> To unsubscribe from this group and stop receiving emails from it, send 
>> an email to sage-devel+...@googlegroups.com. 
>> >> To post to this group, send email to sage-...@googlegroups.com. 
>> >> Visit this group at https://groups.google.com/group/sage-devel. 
>> >> For more options, visit https://groups.google.com/d/optout. 
>> > 
>>
>>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Consider not upgrading to Xcode 9.0

2017-09-20 Thread John H Palmieri 
I successfully built with the current version (not the updated one) of 
eclib plus the patch you mentioned. I will also try to see if gfortran 
builds.

On Wednesday, September 20, 2017 at 2:56:59 PM UTC-7, François Bissey wrote:
>
> So I have a build of sage 8.1.beta5 on OS X with Xcode 9 using 
>
> * #12426 - building with clang - you need autotools installed or generate 
> configure 
> somewhere that does and copy it in place. 
>
> * an external fortran compiler, given the problems with gcc it may 
> very well be that gfortran won’t build either - I need to further check 
> that. 
>
> * an updated eclib 
> https://github.com/JohnCremona/eclib/archive/v20170815.tar.gz 
> probably not necessary but defaulting to C++11 make logs much nicer since 
> it cuts down on the warnings. Beware this is not a sdist tarball. 
>
> * the patch at 
> https://github.com/JohnCremona/eclib/issues/19#issuecomment-314952542 
> this is absolutely needed. 
>
> François 
>
> > On 21/09/2017, at 07:10, François Bissey  > wrote: 
> > 
> > We in fact have a patch on github for a similar problem that was spotted 
> > before. It proved effective in this instance. So I will proceed with 
> trying 
> > to build the rest of sage. 
> > 
> > François 
> > 
> >> On 21/09/2017, at 04:05, John H Palmieri  > wrote: 
> >> 
> >> I put the log for that failure at #12426, and then François created an 
> upstream ticket at 
> >> 
> >>   https://github.com/JohnCremona/eclib/issues/28 
> >> 
> >> in case you want to take a look. 
> >> 
> >> 
> >> On Wednesday, September 20, 2017 at 3:08:13 AM UTC-7, Dima Pasechnik 
> wrote: 
> >> The eclib does not look like a big problem, some rather standard C++ 
> template issue to be tweaked... 
> >> 
> >> On Wednesday, September 20, 2017 at 12:50:32 AM UTC+1, John H Palmieri 
> wrote: 
> >> I just upgraded an OS X box to Xcode 9.0, and now Sage doesn't build: 
> >> 
> >> - with a fresh Sage 8.1.beta5 tarball, gcc doesn't build: 
> >> 
> >> 
> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/functional:1398:2:
>  
> error: no member named 'fancy_abort' in namespace 'std::__1'; did you mean 
> simply 'fancy_abort'? 
> >>_VSTD::abort(); 
> >>^~~ 
> >> 
> >> I'll open a ticket and post the full gcc log. 
> >> 
> >> - instead using trac 12426 (compile with clang instead of gcc), eclib 
> doesn't build: 
> >> 
> >> 
> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:572:12:
>  
> error: call to function 'isnan' that is neither visible in the template 
> definition nor found by argument-dependent lookup 
> >>return isnan(__lcpp_x); 
> >> 
> >>  
> >> 
> >> 
> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:592:12:
>  
> error: call to function 'isinf' that is neither visible in the template 
> definition nor found by argument-dependent lookup 
> >>return isinf(__lcpp_x); 
> >> 
> >> -- 
> >> John 
> >> 
> >> 
> >> -- 
> >> You received this message because you are subscribed to the Google 
> Groups "sage-devel" group. 
> >> To unsubscribe from this group and stop receiving emails from it, send 
> an email to sage-devel+...@googlegroups.com . 
> >> To post to this group, send email to sage-...@googlegroups.com 
> . 
> >> Visit this group at https://groups.google.com/group/sage-devel. 
> >> For more options, visit https://groups.google.com/d/optout. 
> > 
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Consider not upgrading to Xcode 9.0

2017-09-20 Thread François Bissey 
So I have a build of sage 8.1.beta5 on OS X with Xcode 9 using

* #12426 - building with clang - you need autotools installed or generate 
configure 
somewhere that does and copy it in place.

* an external fortran compiler, given the problems with gcc it may 
very well be that gfortran won’t build either - I need to further check that.

* an updated eclib https://github.com/JohnCremona/eclib/archive/v20170815.tar.gz
probably not necessary but defaulting to C++11 make logs much nicer since
it cuts down on the warnings. Beware this is not a sdist tarball.

* the patch at 
https://github.com/JohnCremona/eclib/issues/19#issuecomment-314952542
this is absolutely needed.

François

> On 21/09/2017, at 07:10, François Bissey  wrote:
> 
> We in fact have a patch on github for a similar problem that was spotted 
> before. It proved effective in this instance. So I will proceed with trying
> to build the rest of sage.
> 
> François
> 
>> On 21/09/2017, at 04:05, John H Palmieri  wrote:
>> 
>> I put the log for that failure at #12426, and then François created an 
>> upstream ticket at
>> 
>>   https://github.com/JohnCremona/eclib/issues/28
>> 
>> in case you want to take a look.
>> 
>> 
>> On Wednesday, September 20, 2017 at 3:08:13 AM UTC-7, Dima Pasechnik wrote:
>> The eclib does not look like a big problem, some rather standard C++ 
>> template issue to be tweaked...
>> 
>> On Wednesday, September 20, 2017 at 12:50:32 AM UTC+1, John H Palmieri wrote:
>> I just upgraded an OS X box to Xcode 9.0, and now Sage doesn't build:
>> 
>> - with a fresh Sage 8.1.beta5 tarball, gcc doesn't build:
>> 
>> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/functional:1398:2:
>>  error: no member named 'fancy_abort' in namespace 'std::__1'; did you mean 
>> simply 'fancy_abort'?
>>_VSTD::abort();
>>^~~
>> 
>> I'll open a ticket and post the full gcc log.
>> 
>> - instead using trac 12426 (compile with clang instead of gcc), eclib 
>> doesn't build:
>> 
>> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:572:12:
>>  error: call to function 'isnan' that is neither visible in the template 
>> definition nor found by argument-dependent lookup
>>return isnan(__lcpp_x);
>> 
>> 
>> 
>> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:592:12:
>>  error: call to function 'isinf' that is neither visible in the template 
>> definition nor found by argument-dependent lookup
>>return isinf(__lcpp_x);
>> 
>> -- 
>> John
>> 
>> 
>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "sage-devel" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to sage-devel+unsubscr...@googlegroups.com.
>> To post to this group, send email to sage-devel@googlegroups.com.
>> Visit this group at https://groups.google.com/group/sage-devel.
>> For more options, visit https://groups.google.com/d/optout.
> 

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Consider not upgrading to Xcode 9.0

2017-09-20 Thread François Bissey 
We in fact have a patch on github for a similar problem that was spotted 
before. It proved effective in this instance. So I will proceed with trying
to build the rest of sage.

François

> On 21/09/2017, at 04:05, John H Palmieri  wrote:
> 
> I put the log for that failure at #12426, and then François created an 
> upstream ticket at
> 
>https://github.com/JohnCremona/eclib/issues/28
> 
> in case you want to take a look.
> 
> 
> On Wednesday, September 20, 2017 at 3:08:13 AM UTC-7, Dima Pasechnik wrote:
> The eclib does not look like a big problem, some rather standard C++ template 
> issue to be tweaked...
> 
> On Wednesday, September 20, 2017 at 12:50:32 AM UTC+1, John H Palmieri wrote:
> I just upgraded an OS X box to Xcode 9.0, and now Sage doesn't build:
> 
> - with a fresh Sage 8.1.beta5 tarball, gcc doesn't build:
> 
> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/functional:1398:2:
>  error: no member named 'fancy_abort' in namespace 'std::__1'; did you mean 
> simply 'fancy_abort'?
> _VSTD::abort();
> ^~~
> 
> I'll open a ticket and post the full gcc log.
> 
> - instead using trac 12426 (compile with clang instead of gcc), eclib doesn't 
> build:
> 
> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:572:12:
>  error: call to function 'isnan' that is neither visible in the template 
> definition nor found by argument-dependent lookup
> return isnan(__lcpp_x);
> 
> 
> 
> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:592:12:
>  error: call to function 'isinf' that is neither visible in the template 
> definition nor found by argument-dependent lookup
> return isinf(__lcpp_x);
> 
> -- 
> John
> 
> 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Consider not upgrading to Xcode 9.0

2017-09-19 Thread John H Palmieri 
I just upgraded an OS X box to Xcode 9.0, and now Sage doesn't build:

- with a fresh Sage 8.1.beta5 tarball, gcc doesn't build:

/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/functional:1398:2:
 
error: no member named 'fancy_abort' in namespace 'std::__1'; did you mean 
simply 'fancy_abort'?
_VSTD::abort();
^~~

I'll open a ticket and post the full gcc log.

- instead using trac 12426 (compile with clang instead of gcc), eclib 
doesn't build:

/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:572:12:
 
error: call to function 'isnan' that is neither visible in the template 
definition nor found by argument-dependent lookup
return isnan(__lcpp_x);



/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/cmath:592:12:
 
error: call to function 'isinf' that is neither visible in the template 
definition nor found by argument-dependent lookup
return isinf(__lcpp_x);

-- 
John

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.