Yes it looks like you're right. I have fixed this in the repository.
Thanks for reporting this!
Patrick
On Fri, Apr 11, 2008 at 11:46:57AM +0200, Thord Andersson wrote:
> Hi,
>
> In rowcol_source.c , functions gsl_matrix_subrow() and
> gsl_matrix_subcolumn() , there are overflow checks which I think compare
> with the wrong matrix dimension.
> Possible fixes are indicated in inline comments. My applications works
> now....
>
> Thank you for GSL! It's great!
> /Thord Andersson
>
> QUALIFIED_VIEW(_gsl_vector,view)
> FUNCTION (gsl_matrix, subrow) (QUALIFIED_TYPE(gsl_matrix) * m, const size_t
> i, const size_t offset, const size_t n)
> ..
> ..
> else if (offset + n > m->size1) // Should be: (offset + n > m->size2)
> (compare with length of row -->number of columns)
> {
> GSL_ERROR_VAL ("dimension n overflows matrix", GSL_EINVAL, view);
> }
> ..
> ..
>
> and
>
> QUALIFIED_VIEW(_gsl_vector,view)
> FUNCTION (gsl_matrix, subcolumn) (QUALIFIED_TYPE(gsl_matrix) * m, const
> size_t j, const size_t offset, const size_t n)
> ..
> ..
> else if (offset + n > m->size2) // Should be: (offset + n > m->size1)
> (compare with length of column -->number of rows)
> {
> GSL_ERROR_VAL ("dimension n overflows matrix", GSL_EINVAL, view);
> }
> ..
>
>
>
>
> _______________________________________________
> Bug-gsl mailing list
> [email protected]
> http://lists.gnu.org/mailman/listinfo/bug-gsl
_______________________________________________
Bug-gsl mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-gsl
