#6456: Upgrade cvxopt in sage from 0.9 to 1.1.2
--------------------------------+-------------------------------------------
Reporter: was | Owner: mabshoff
Type: defect | Status: needs_work
Priority: major | Milestone: sage-4.5.2
Component: packages | Keywords:
Author: schilly, dimpase | Upstream: N/A
Reviewer: | Merged:
Work_issues: |
--------------------------------+-------------------------------------------
Comment(by drkirkby):
Replying to [comment:78 dimpase]:
> Dave,
>
> your spkg-check is buggy. You probably end up calling a very wrong
python to run the examples. E.g. if I cd to
> cvxopt-1.1.2/src/examples/doc/chap8
> make a symbolic link mcsdp.py to mcsdp, fire up sage, and do
>
> sage: load('mcsdp.py')
>
> it happily runs without any errors etc. And this is one of examples your
spkg-check fails on with weird error messages. So the problem does not
seem to be in cvxopt here, but rather in your script...
>
> Dima
>
Dima,
I made it very clear that "I've spent as long on this as I am able to for
a while, so I thought I'd post what I have got." I also made it clear I'd
not committed the changes - a clear reflection I was not confident of them
all.
http://trac.sagemath.org/sage_trac/ticket/6456#comment:58
If the wrong python is being called, that's a bug, as the first in the
path, which is the one in Sage, should be called. One could work around
that with "$SAGE_LOCAL/nib/python"
I'm aware of the issue with calling the .bin file, but as I remarked
above, that's in the noise compared to the most significant issues of
warnings from the compiler.
Feel free to improve the test suite. Rather than execute them all in a
loop, perhaps its better to cd to the directory and run them from there.
But those changes are not going to get around the more serious issue,
which is seen well before the test suite is run.
I'm involved in many things at Sage at the moment.
* This ticket
* #9343 - the upgrade of Pari, which is a non-trivial issue
* #9281 - trying to get more self-tests into Sage.
* #8059 - update Singular SPKG to newest upstream release. Another big
package
In addition, I've involved in same way with other less time consuming
tickets
* #9533: Update GSL to the latest upstream release (1.14) & permit
parallel building.
* #9568: Update IML to the newest upstream release, and improve spkg-
install
* #9603: Force iconv to build + install on HP-UX. Currently it is only
installed on Solaris and Cygwin.
I really don't have a huge amount of time to devote to this one. I'm less
inclined to devote time to improving the self-tests on this package, as
its clear the package is more seriously broken when its compiled.
I seem to get cc'ed on a number of tickets where people can't be bothered
to build on Solaris, so they think I will do the work for them. I'm
tending to do that less now. There should be a much larger disk on t2
soon, so there will be even less excuse for people to pester me do do the
Solaris checking. I never pester others other check code on Linux or OS X.
I just do it myself.
To be fair Dima, this is not aimed at you, as you have checked code on
Solaris many times.
With all the comments above on this ticket, how many people have actually
made an effort to test cvxopt on a few computers with a few different
operating systems. Compare that with
#9533: Update GSL to the latest upstream release (1.14) & permit parallel
building.
where a lot of people have tested the code on multiple operating systems,
with multiple compilers and run all the self-tests and run all the
doctests on popular platforms. That's been done on
* Cygwin
* FreeBSD
* HP-UX
* Linux
* OpenSolaris
* OS X
* Solaris
Dave
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/6456#comment:79>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
