Awesome, installing python 2.7.5.p2  got me past the readline error and it 
continued building from there. 

It ran for about 30 minutes(running a ton of sanity tests, a good sign?), 
then froze here for an hour or two:

/cygdrive/c/cygwin//home/evan.oman/sage-6.1/local/var/tmp/sage/build/atlas-3.10.1.p7/src/ATLAS-build/../ATLAS//src/blas/level2//ATL_trsv.c:
 
In function ‘ATL_trsvUN_k’:
/cygdrive/c/cygwin//home/evan.oman/sage-6.1/local/var/tmp/sage/build/atlas-3.10.1.p7/src/ATLAS-build/../ATLAS//src/blas/level2//ATL_trsv.c:105:6:
 
warning: assignment discards ‘const’ qualifier from pointer target type 
[enabled by default]
    a = A + ((size_t)lda2)*(N-nb);
      ^
ar r 
/cygdrive/c/cygwin//home/evan.oman/sage-6.1/local/var/tmp/sage/build/atlas-3.10.1.p7/src/ATLAS-build/lib/libatlas.a
 
ATL_L2AIsOverlapped.o ATL_sgbmv.o ATL_sgpmv.o ATL_sgpr.o ATL_ssbmv.o 
ATL_sspmv.o ATL_sspr.o ATL_sspr2.o ATL_ssymv.o ATL_ssyr.o ATL_ssyr2.o 
ATL_stbmv.o ATL_stbsv.o ATL_stpmv.o ATL_stpsv.o ATL_strmv.o ATL_strsv.o
echo 
/cygdrive/c/cygwin//home/evan.oman/sage-6.1/local/var/tmp/sage/build/atlas-3.10.1.p7/src/ATLAS-build/lib/libatlas.a
/cygdrive/c/cygwin//home/evan.oman/sage-6.1/local/var/tmp/sage/build/atlas-3.10.1.p7/src/ATLAS-build/lib/libatlas.a
touch sblas2.grd
make[7]: Leaving directory 
'/cygdrive/c/cygwin/home/evan.oman/sage-6.1/local/var/tmp/sage/build/atlas-3.10.1.p7/src/ATLAS-build/src/blas/level2'
make[6]: Leaving directory 
'/cygdrive/c/cygwin/home/evan.oman/sage-6.1/local/var/tmp/sage/build/atlas-3.10.1.p7/src/ATLAS-build/tune/blas/ger'
make[5]: Leaving directory 
'/home/evan.oman/sage-6.1/local/var/tmp/sage/build/atlas-3.10.1.p7/src/ATLAS-build/bin'

I decided to cancel the build and I am rerunning it now, wish me luck!


On Tuesday, March 11, 2014 5:33:25 PM UTC-5, Jean-Pierre Flori wrote:
>
> I did not look at the logs yet, but you definitely need python 2.7.5.p2 
> which is not in the 6.1 series.
> See http://trac.sagemath.org/ticket/15317
>
> On Tuesday, March 11, 2014 7:32:28 PM UTC+1, Evan Oman wrote:
>>
>> Alright so I have attached the entire Python log so hopefully that will 
>> contain what you were looking for. 
>>
>> Starting a sage shell, running its python, and typing import readline 
>> yields:
>> >>> import readline
>> Traceback (most recent call last):
>>   File "<stdin>", line 1, in <module>
>> ImportError: No module named readline
>>
>>
>> Also I started fresh from a new tarball and that fixed the multiple 
>> versions of python issue but the build stopped at the same point(trying to 
>> import readline).
>>
>> Thank you all for your willingness to help!
>>
>> On Tuesday, March 11, 2014 5:34:59 AM UTC-5, mmarco wrote:
>>>
>>> That is what singular does, right?
>>>
>>> El martes, 11 de marzo de 2014 11:23:57 UTC+1, Dima Pasechnik escribió:
>>>>
>>>> On 2014-03-11, Jean-Pierre Flori <jpf...@gmail.com> wrote: 
>>>> > 
>>>> > 
>>>> > On Tuesday, March 11, 2014 10:20:08 AM UTC+1, mmarco wrote: 
>>>> >> 
>>>> >> Is it possible now to build sage on cygwin and obtain something that 
>>>> can 
>>>> >> be distributed? In that case, maybe we should include that in the 
>>>> download 
>>>> >> section. 
>>>> >> 
>>>> > I'd say mostly yes, just like on linux. 
>>>> > But some people said we should have something even more easy to 
>>>> install, 
>>>> > i.e. include Sage, but all parts of cygwin needed. 
>>>> probably the most natural thing to do would be to make a Sage Cygwin 
>>>> package hosted on a custom Cygwin setup server: 
>>>> https://sourceware.org/cygwin-apps/package-server.html 
>>>>
>>>> Dima 
>>>>   
>>>> > Occasionally I do package some binary things with the sage part (so 
>>>> you 
>>>> > have to install cygwin, the needed dependencies, and potentially 
>>>> rebase and 
>>>> > update hardcoded paths), without any promise it's working. 
>>>> > At least the initial install is working in the VMs I use. 
>>>> > 
>>>> > IIRC last time Harald put some of these tarballs somewhere on the net 
>>>> (or 
>>>> > maybe that was solaris trarballs), that's all I know. 
>>>> > 
>>>> > Without a VM running somewhere I or someone else can access all the 
>>>> time, 
>>>> > it won't be easy to automate such a task. 
>>>> > 
>>>>
>>>>

-- 
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 http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Reply via email to