Since it seems that most of the problems everyone now has are caused
by the mwrank.spkg -> cremona.spkg switchover, and given that 99.9% of
the extra functionality has not yet been wrapped into Sage and is
hence unavailable from Sage anyway, why don't we just chuck out the
cremona.spkg into an optional extra for the time being, as no-one wil
notice anyway?
Otherwise everytime I look at sage-devel I just feel guilty at
apparently being the cause of everytone's grief while giving them no
apparent benefit.
The only problem I can see with this is that the various edits which
have had to be done to my source code to make it gcc4.2+ compatible
have not been made to the mwrank sources.
John
On 05/12/2007, mabshoff
<[EMAIL PROTECTED]> wrote:
>
>
>
> On Dec 5, 6:34 am, "Justin C. Walker" <[EMAIL PROTECTED]> wrote:
> > Hi, all,
> >
> > On Dec 4, 2007, at 20:07 , mabshoff wrote:
> >
> > [snip]
> >
> > > Those are all cause by mwrank. The issue seems to be the same issue as
> > > Justin Walker reported in the 2.8.15 thread in sage-support.
> > > Interestingly nobody ever reported it and I now finally have a
> > > testcase that also produces the issue on sage.math under valgrind. I
> > > have send John Cremona two valgrind traces and hope that it will help
> > > to fix the issues.
> >
> > An update:
> >
>
> Hello Justin,
>
> > FWIW, I ran mwrank on the same input as you used for your test, and
> > it completed without a peep, at least until I quit.
> >
> > 'mwrank' doesn't like it if you just exit (by typing "^D" or, more or
> > less equivalently, by "ending a file").
> >
> > Thus, if you run 'mwrank' and give it a file containing, for example,
> > "[0,0,1,-1,0]" and nothing else, it will barf at the end. If your
> > file contains "[0,0,0,0,0]", the program sweetly closes up shop (when
> > it reads this) and quits.
> >
> > The error I get is
> > bad ZZ input
> > Abort trap
> >
> > Anyone know why this doesn't show up in the test logs? Is it worth
> > tracking down?
>
> Yep, any crash is worth fixing and I assume it is still present in the
> newer cremona.spkg codebase.
>
> > The issue is that the terminating condition for input processing in
> > 'getcurve()' is a "null curve" (""[0,0,0,0,0]""), rather than EOF.
> > An EOF is an error condition, hence the abort().
> >
> > A C++ mavin may know how to fix this (I don't). Trac #1403.
>
> Sorry, I overwrote #1402 with what I thought was the real issue until
> realizing that your error is an orthogonal crash. The above two tickts
> (#1402, #1403) might explain why mwrank on Cygwin behaved really badly
> and just crashed more or less randomly. We blamed this on crappy
> Cygwin, but the fault may lie somewhere else.
>
> Cheers,
>
> Michael
>
> > Justin
> >
> > --
> > Justin C. Walker, Curmudgeon-At-Large
> > Institute for the Absorption of Federal Funds
> > --------
> > Men are from Earth.
> > Women are from Earth.
> > Deal with it.
> > --------
> >
>
--
John Cremona
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---
