Some very useful remarks from Mark...
---------- Forwarded message ----------
From: Mark Watkins <mark.watk...@sydney.edu.au>
Date: Tue, Jun 7, 2016 at 11:43 AM
Subject: Re: proposal: possibly remove Watkins Sympow from Sage
To: William Stein <wst...@gmail.com>
William Stein wrote:
>
> Hi,
>
> Long ago in 2006, I put Mark Watkins amazing C program "sympow" in Sage:
>
> [...]
>
> I just tried to use Sympow, since it would be extremely useful for
> some research I'm doing with Barry
> Mazur during my visit to Harvard this week. However, it just
> segfaults in both sage-6.10 **and** sage-7.3.beta.
> It silently fails when doing the above "not tested" thing, but if you
> try it at the command line it segfaults:
>
> -----
> (sage-sh) 95d92fa7cb50414ea35d9897eabe44de@compute4-us:sympow$ sympow
> -new_data 2
> Make data for symmetric power 2
> Running the new_data script for -sp 2
> Making the datafiles for -sp 2
>
> Rewarping the param_data file
> Left with 13 entries in param_data
> Segmentation fault (core dumped)
The relevant code (in my version) is:
fclose(F); printf("Left with %i entries in param_data\n",i);
F=fopen("datafiles/param_data","w");
So it likely can't open "datafiles/" for writing.
> I think nothing else in Sage depends on sympow.
>
> So... after some discussion, if there is any, I'm going to recommend
> that we remove sympow. There's no point in building, compiling, and
> distributing something that doesn't work at all. This just wastes
> time and maybe makes the release managers life slightly harder.
AFAIK, there were (major) changes made by Sage maintainers over the
years, firstly I had to help Michael Abshoff even build it (for Suns?)
back in 2007, and then a number of issues with working directories, etc.
In any event, the principal "application" of sympow was modular degrees,
and I think Bill Allombert might have something for that in PARI.
For your work with Barry, you could just download
http://magma.maths.usyd.edu.au/~watkins/sympow.src.tar.bz2
http://magma.maths.usyd.edu.au/~watkins/sympow.tar.bz2
I am about to leave for Europe in 8 hours, so I can't say that I will
necessarily be in email contact.
===
Mark Watkins
watk...@maths.usyd.edu.au
--
William (http://wstein.org)
--
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.
