make check failure from current sources

[Ben Abbott]

On Oct 31, 2008, at 7:12 AM, Jaroslav Hajek wrote:

> On Fri, Oct 31, 2008 at 11:30 AM, Jaroslav Hajek <highegg at gmail.com>  
> wrote:
>> On Fri, Oct 31, 2008 at 11:28 AM, Ben Abbott <bpabbott at mac.com>  
>> wrote:
>>>
>>> On Oct 31, 2008, at 6:16 AM, Jaroslav Hajek wrote:
>>>
>>>> On Fri, Oct 31, 2008 at 11:02 AM, Ben Abbott <bpabbott at mac.com>  
>>>> wrote:
>>>>>
>>>>> On Oct 30, 2008, at 7:28 PM, Ben Abbott wrote:
>>>>>
>>>>>> I pulled this evening, my tip is
>>>>>>
>>>>>>> changeset:   8295:16819ca24fa5
>>>>>>> tag:         tip
>>>>>>> user:        John W. Eaton <jwe at octave.org>
>>>>>>> date:        Thu Oct 30 18:03:37 2008 -0400
>>>>>>> summary:     fix typo in docs
>>>>>>
>>>>>> During my building went fine, but "make check" ...
>>>>>>
>>>>>> [snip]
>>>>>>
>>>>>> scripts/general/dblquad.m ..............................error:  
>>>>>> `pso'
>>>>>> undefined near line 248 column 23
>>>>>> error: evaluating argument list element number 1
>>>>>> error: called from:
>>>>>> error:   ./fntests.m at line -1, column -1
>>>>>>
>>>>>> make[2]: *** [check] Error 1
>>>>>> make[1]: *** [check] Error 2
>>>>>> make: *** [check] Error 2
>>>>>>
>>>>>> As the file dblquad.m only has 68 lines and does not contain  
>>>>>> the string
>>>>>> `pso', I'm at a loss.
>>>>>>
>>>>>> Anyone have an idea?
>>>>>>
>>>>>> Ben
>>>>>
>>>>> Running the first test of dblquad from the command line ...
>>>>>
>>>>> octave:1> assert (dblquad (@(x,y) 1 ./ (x+y), 0, 1, 0, 1),  
>>>>> 2*log(2),
>>>>> 1e-6)
>>>>> error: A(I): Index exceeds matrix dimension.
>>>>> error: called from:
>>>>> error:
>>>>>
>>>>> /Users/bpabbott/Development/mercurial/octave-3.1.51/scripts/ 
>>>>> general/quadgk.m
>>>>> at line 374, column 7
>>>>> error:
>>>>>
>>>>> /Users/bpabbott/Development/mercurial/octave-3.1.51/scripts/ 
>>>>> general/dblquad.m
>>>>> at line 61, column 10
>>>>> error:    at line 54, column 26
>>>>> error:    at line 275, column 30
>>>>> error: evaluating argument list element number 1
>>>>> error:
>>>>>
>>>>> /Users/bpabbott/Development/mercurial/octave-3.1.51/scripts/ 
>>>>> general/quadgk.m
>>>>> at line 413, column 5
>>>>> error: `warn_state' undefined near line 374 column 19
>>>>> error: evaluating argument list element number 1
>>>>> error: called from:
>>>>> error:
>>>>>
>>>>> /Users/bpabbott/Development/mercurial/octave-3.1.51/scripts/ 
>>>>> general/quadgk.m
>>>>> at line 374, column 7
>>>>> error:
>>>>>
>>>>> /Users/bpabbott/Development/mercurial/octave-3.1.51/scripts/ 
>>>>> general/dblquad.m
>>>>> at line 54, column 5
>>>>> error: evaluating argument list element number 1
>>>>>
>>>>> Ben
>>>>>
>>>>>
>>>>
>>>> Can anyone else reproduce this problem? It doesn't happen in my  
>>>> fresh
>>>> build. If not, could you please track the problem down a little?
>>>> There is no indexing operation on line 374 of quadgk. It would  
>>>> help if
>>>> you identified the exact inputs that cause quadgk to fail.
>>>
>>> I tried saving the variables to a file when quadgk is called ...
>>>
>>>       save quadgk.mat f a b varargin
>>>
>>> Unfortunately ...
>>>
>>> octave:5> assert (dblquad (@(x,y) 1 ./ (x+y), 0, 1, 0, 1),  
>>> 2*log(2), 1e-6)
>>> error: octave_base_value::save_ascii(): wrong type argument  
>>> `<unknown type>'
>>> warning: save: no such variable `f'
>>> warning: save: no such variable `a'
>>> warning: save: no such variable `b'
>>> warning: save: no such variable `varargin'
>>> error: called from:
>>> error:
>>> /Users/bpabbott/Development/mercurial/octave-3.1.51/scripts/ 
>>> general/quadgk.m
>>> at line 123, column 1
>>> error:
>>> /Users/bpabbott/Development/mercurial/octave-3.1.51/scripts/ 
>>> general/dblquad.m
>>> at line 61, column 10
>>> error:    at line 54, column 26
>>> error:    at line 279, column 30
>>> error: evaluating argument list element number 1
>>> error:
>>> /Users/bpabbott/Development/mercurial/octave-3.1.51/scripts/ 
>>> general/quadgk.m
>>> at line 417, column 5
>>> error: `warn_state' undefined near line 378 column 19
>>> error: evaluating argument list element number 1
>>> error: called from:
>>> error:
>>> /Users/bpabbott/Development/mercurial/octave-3.1.51/scripts/ 
>>> general/quadgk.m
>>> at line 378, column 7
>>> error:
>>> /Users/bpabbott/Development/mercurial/octave-3.1.51/scripts/ 
>>> general/dblquad.m
>>> at line 54, column 5
>>> error: evaluating argument list element number 1
>>>
>>> I'm flying to Asia today. I'll do some more debugging on my  
>>> loooong flight.
>>>
>>
>> Have a nice trip, then. I just found a defect in idx-vector.h that  
>> may
>> be (by my wild guess) the cause of your problem. Currently compiling
>> with a patch. I'll push it shortly (after make check).
>>
>
> Does this patch:
> http://hg.savannah.gnu.org/hgweb/octave/rev/7e87d3d76a56
>
> fix your problem?
>
> cheers

No. The problem remains. I'll be unusually busy for the next several  
days (at the least), but will try to dig a bit deeper into this problem.

If anyone else is building the developer's sources on Mac OSX, please  
let me/us know if you see this problem as well.

Ben