Re: null-assign.tst failing in 'make check'

[John W. Eaton]

Yeah, the test needs to be updated.  I'll fix it.

jwe


On 09/01/2016 12:13 PM, Rik wrote:
> On 09/01/2016 03:38 AM, address@hidden wrote:
> > Subject:
> > make check fails w/ recent hg
> > From:
> > "Dmitri A. Sergatskov" <address@hidden>
> > Date:
> > 09/01/2016 01:47 AM
> >
> > To:
> > Octave maintainers mailing list <address@hidden>
> >
> > List-Post:
> > <mailto:address@hidden>
> > Precedence:
> > list
> > MIME-Version:
> > 1.0
> > Message-ID:
> > <address@hidden>
> > Content-Type:
> > multipart/alternative; boundary=001a1142b98ed021e4053b6e445e
> > Message:
> > 5
> >
> >
> > changeset:   22415:a338dbafcc09
> >
> > >> test null-assign.tst
> > ***** test
> >  a = ones (3); b = []; fail ("subsasgn (a, substruct ('()',
> > {':',1:2}), b)", ".");
> > !!!!! test failed
> > expected error <.> but got none
>
> This is probably due to this changeset
>
> changeset:   22390:6cbbd550aa91
> user:        John W. Eaton <address@hidden>
> date:        Thu Aug 25 18:18:20 2016 -0400
> summary:     allow subsasgn (x, idx, zeros(0,0)) to remove elements (bug
> #48867)
>
> Either the test suite in null-assign.tst needs to be updated for the new
> behavior, or it has caught some sort of regression.
>
> --Rik