segfault with univariate polynomial, realfield, complexfield

[ncalexan]

This is with a modified

| Sage Version 4.1.rc1, Release Date: 2009-07-07                     |

Mac OS X, Intel hardware.

sage: RealField(300)['x']( [ 1, ComplexField(300).gen(), 0 ])


------------------------------------------------------------
Unhandled SIGSEGV: A segmentation fault occured in SAGE.
This probably occured because a *compiled* component
of SAGE has a bug in it (typically accessing invalid memory)
or is not properly wrapped with _sig_on, _sig_off.
You might want to run SAGE under gdb with 'sage -gdb' to debug this.
SAGE will now terminate (sorry).
------------------------------------------------------------


Process SAGE exited abnormally with code 1

Depends on #11900

CC: @orlitzky

Component: commutative algebra

Keywords: polynomial segfault

Author: Michael Orlitzky

Reviewer: Mike Hansen

Merged: sage-5.0.beta0

Issue created by migration from https://trac.sagemath.org/ticket/6777

[simon-king-jena]

comment:1

I don't obtain a segmentation fault, but this is on Debian GNU/Linux.

On sage-devel, it is discussed that one gets a rather strange phenomenon: Before the traceback starts, some ERROR is printed.

sage: RealField(300)['x']( [ 1, ComplexField(300).gen(), 0 ])
ERROR: An unexpected error occurred while tokenizing input
The following traceback may be corrupted or invalid
The error message is: ('EOF in multi-line statement', (1375, 0))

ERROR: An unexpected error occurred while tokenizing input
The following traceback may be corrupted or invalid
The error message is: ('EOF in multi-line statement', (1375, 0))

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)

I get this with unpatched sage (hence, no segfault), and Michael Orlitzky reports the same after fixing the segfault.

However, with some of my patches applied, one simply gets a straight forward traceback:

sage: RealField(300)['x']([ ComplexField(300).gen() ]) 
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
...
TypeError: Unable to convert x (='1.00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000*I') to real number.

Could you test whether #9138 (or perhaps #11900) actually fixes the problem already?

[orlitzky]

Dependencies: #11900

[orlitzky]

comment:2

The fix is in #11900, and it has a positive review, so I'm adding it as a dependency before a doctest can be added.

[orlitzky]

Author: Michael Orlitzky

[mwhansen]

Patch to add a doctest expecting a TypeError

[mwhansen]

Reviewer: Mike Hansen

[mwhansen]

comment:5

Attachment: sage-trac_6777.patch.gz

mjo's patch looks good to me. I updated the formatting slightly.

[jdemeyer]

Merged: sage-5.0.beta0