octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Test failures associated with license.m


From: Daniel J Sebald
Subject: Re: Test failures associated with license.m
Date: Sat, 04 Oct 2014 15:25:15 -0500
User-agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.24) Gecko/20111108 Fedora/3.1.16-1.fc14 Thunderbird/3.1.16

On 10/04/2014 03:00 PM, Carnë Draug wrote:
On 4 October 2014 19:41, Daniel J Sebald<address@hidden>  wrote:
I'm seeing some test failures.  One with polygcd, which I think is a known
random occurrence failure.  (Should we change it to "known failure"?)  Then
there are five failures for the output of the license command:

processing
/usr/local/src/octave/octave-complex_operators/octave/scripts/miscellaneous/license.m

***** assert ((license ("inuse", "octave")).feature, "octave")
!!!!! test failed
cellfun: C must be a cell array

etc.

At first I couldn't replicate this. The problem is that you don't have
any package installed and 'pkg ("describe", "all")' is returning nothing
instead of an empty cell array. I guess this should be fixed in pkg instead.

Actually, I'd arguee that it should be fixed so that it returns a
struct array, rather than a cell array of scalar structs but I'm
afraid that may break old code.

This one confounded me, so now I know what is happening.  I'm getting

octave:63> [out1 out2] = pkg ("describe", "all")
warning: describe: some elements in list of return values are undefined
warning: called from
    describe at line 86 column 1
    pkg at line 537 column 31
out1 = [](0x0)
out2 = {}(0x0)
octave:64> class(out1)
ans = double
octave:65> class(out2)
ans = cell

With or without the fix, if the user has no packages installed then there isn't much of a test. No real problem, as I'm sure there are so many already with installed packages that it would be caught quick. However, is there is an easy way to create a bogus or empty package from a script in some temporary file to install before running the test?

Dan



reply via email to

[Prev in Thread] Current Thread [Next in Thread]