Re: patch for io.tst

[c.]

On 23 Dec 2013, at 20:15, Rik <address@hidden> wrote:

> On 12/23/2013 10:59 AM, c. wrote:
>> On 23 Dec 2013, at 19:20, Rik <address@hidden> wrote:
>> 
>>> On 12/23/2013 04:08 AM, c. wrote:
>>>> Here's a modified changeset, which branch is most appropriate for it?
>>> Carlo,
>>> 
>>> I'll let jwe decide on that one.  I can see good reasons to put it on
>>> stable so that 'make check' doesn't fail, but this isn't affecting everyone
>>> and it would be nice to have RC2 transition to being the true release,
>>> which argues to put it on gui-release.
>>> 
>>> --Rik
>> Pushing to gui-release sounds fine as it seems I was the only 
>> one affected by this problem ...
>> 
>> I wonder why the tests were failing for me and not others ...
>> 
>> Do the tests in io.tst work for you if you do 'make install' and then 
>> run '__run_test_suite__' as non-root user?
> You're right.  I get the failures if I install octave to /usr/local/ and
> then use __run_test_suite__.  Normally, I have some bleeding edge installs
> and these I put in my /home directory which is writable by me.  Hence, I
> didn't see the problem.  Since __run_test_suite__ is an internal function
> not meant to be called by ordinary users I think it would still be okay to
> wait and puit it on gui-release.
> 
> --Rik

I agree this is not important enough to justify another release candidate.
I pushed to gui-release.
c.