On Debian testing running on core i7 + 16 GB RAM, incremental build (Python
2.7) starting from 8.7.beta5 builds and passes ptestlong with no failure.
HTH,
Le dimanche 3 mars 2019 01:30:18 UTC+1, Volker Braun a écrit :
>
> As always, you can get the latest beta version from the "develop" git
> b
Testing with --optional=sage,optional,external
with
Using
--optional=bliss,cbc,ccache,cmake,cryptominisat,dot2tex,external,glucose,lrslib,memlimit,mpir,normaliz,notedown,pandoc_attributes,pycosat,pynormaliz,python2,rst2ipynb,sage
External software detected for doctesting:
ffmpeg,graphviz,gurobi
Ditto on two other similar achines; however, I got a transient failure on
tesing an example coming form the "Mathematical Computihg with Sagemath"
book :
--
sage -t --long --warn-long 53.4
src/sage/tests/books/computational-mat
upgrading from some early 8.6 beta, I get something weird with file
permissions on gap. Everything seems to go well, but:
pkg/laguna-3.9.0 ->
/Users/.../sage/local/var/tmp/sage/build/gap-4.10.0.p0/inst/Users/.../sage/local/share/gap/pkg
pkg/polenta-1.3.8 ->
/Users/.../sage/local/var/tmp/sag
Perhaps this is https://trac.sagemath.org/ticket/26996#comment:23. Or at
least very similar.
On Thursday, March 7, 2019 at 10:50:10 AM UTC-6, kcrisman wrote:
>
> upgrading from some early 8.6 beta, I get something weird with file
> permissions on gap. Everything seems to go well, but:
>
>
>
> p
https://trac.sagemath.org/ticket/27388 fixed the permissions, but
because it's an upgrade from the state where there was a read-only
file, it breaks (on OSX only)
On Thu, Mar 7, 2019 at 4:50 PM kcrisman wrote:
>
> upgrading from some early 8.6 beta, I get something weird with file
> permissions
On Friday, March 8, 2019 at 3:24:45 AM UTC-5, Dima Pasechnik wrote:
>
> https://trac.sagemath.org/ticket/27388 fixed the permissions, but
> because it's an upgrade from the state where there was a read-only
> file, it breaks (on OSX only)
>
Would that mean that, in general, upgrades from beta
On Fri, Mar 8, 2019 at 12:40 PM kcrisman wrote:
>
>
>
> On Friday, March 8, 2019 at 3:24:45 AM UTC-5, Dima Pasechnik wrote:
>>
>> https://trac.sagemath.org/ticket/27388 fixed the permissions, but
>> because it's an upgrade from the state where there was a read-only
>> file, it breaks (on OSX only)
> this is too blanket :-)
> e.g. in this particular case it suffices to manually remove
> SAGE_LOCAL/share/gap
> and rebuild gap spkg
> (and then run make, naturally...)
>
>
Great, thanks! Apparently that sufficed.
> Certainly there is never a guarantee that a "git pull"+"make" would
>
On Friday, March 8, 2019 at 8:18:02 AM UTC-5, Dima Pasechnik wrote:
>
> On Fri, Mar 8, 2019 at 12:40 PM kcrisman >
> wrote:
> >
> >
> >
> > On Friday, March 8, 2019 at 3:24:45 AM UTC-5, Dima Pasechnik wrote:
> >>
> >> https://trac.sagemath.org/ticket/27388 fixed the permissions, but
> >>
this is a MacOS only bug. Someone running it should work on fix...
On Wed, 14 Aug 2019 20:35 kcrisman, wrote:
>
>
> On Friday, March 8, 2019 at 8:18:02 AM UTC-5, Dima Pasechnik wrote:
>>
>> On Fri, Mar 8, 2019 at 12:40 PM kcrisman wrote:
>> >
>> >
>> >
>> > On Friday, March 8, 2019 at 3:24:45
On Thu, Mar 7, 2019 at 5:50 PM kcrisman wrote:
>
> upgrading from some early 8.6 beta, I get something weird with file
> permissions on gap. Everything seems to go well, but:
>
>
>
> pkg/laguna-3.9.0 ->
> /Users/.../sage/local/var/tmp/sage/build/gap-4.10.0.p0/inst/Users/.../sage/local/share/gap
> That's expected. The package build directory is under SAGE_LOCAL
> ($SAGE_LOCAL/var/tmp/sage/build/gap-4.10.0.p0) but the file's
> destination path ($SAGE_LOCAL) is appended to this. This is how it's
> supposed to work (DESTDIR installation).
>
> It's probably too late to diagnose now (th
On Thu, Aug 15, 2019 at 2:22 PM kcrisman wrote:
>
>
>> That's expected. The package build directory is under SAGE_LOCAL
>> ($SAGE_LOCAL/var/tmp/sage/build/gap-4.10.0.p0) but the file's
>> destination path ($SAGE_LOCAL) is appended to this. This is how it's
>> supposed to work (DESTDIR installati
This also happened to me a few weeks ago. The underlying problem seems to
be that the uninstall script fails because it tries to call
sage-dist-helpers, but cannot find it. This file has recently been moved
from src/bin to build/bin, which the uninstall script cannot know about.
Once upgraded t
Yes, that fits. Now that I know to look, I see buried in my gap install log
Uninstalling existing 'gap'
Running pre-uninstall script for 'gap'
/Users/jpalmier/Desktop/Sage/git/sage/local/var/lib/sage/scripts/gap/spkg-prerm:
line 13: /Users/jpalmier/Desktop/Sage/git/sage/src/bin/sage-dist-helper
On Thu, Aug 15, 2019 at 6:58 PM Markus Wageringel
wrote:
>
> This also happened to me a few weeks ago. The underlying problem seems to be
> that the uninstall script fails because it tries to call sage-dist-helpers,
> but cannot find it. This file has recently been moved from src/bin to
> build
>
> > This has also been reported on devel [1,2] and might affect more Mac
> users when upgrading to 8.9. Perhaps a copy of sage-dist-helpers could be
> added back to src/bin for some deprecation period?
>
> A symlink could do it.
>
Very nice sleuthing all around. I do recommend something b
For a data point, I am currently trying to use binary-pkg to create a Sage
9.0 binary for Mac, and ran into *exactly* the same issue - presumably
because it (the version of Sage I had in binary-pkg, that is) had upgraded
from 8.9.beta6 or so. I guess it must have hit the window for not having
19 matches
Mail list logo
