[Bioperl-l] tests failing miserably

Steve Chervitz sac at pw600a.bioperl.org
Thu May 15 17:24:52 EDT 2003


For Bio::Tools::Blast the API is very different from SearchIO::blast,  
so masquerading it would require writing some adaptor code. It's a neat  
idea though, from the point of view of keeping unmaintained legacy code  
happy.

Downside: It runs the risk of confusing newcomers, and it increases  
maintenance since now you have to make sure the masquerading continues  
working. If we do masquerade any modules, I'd recommend issuing a  
deprecation warning and not support the masquerading indefinitely.

Steve

On Thursday, May 15, 2003, at 11:07 US/Pacific, Hilmar Lapp wrote:

>
> The hmmer tests pass meanwhile. Over the long run I'd agree though -  
> if there is no functionality difference then we'll deprecate the old  
> parser.
>
> In fact, to hide such moves from legacy scripts, I thought could we  
> masquerade  the old parser to be in fact the new one, by delegation,  
> or re-blessing or whatever? Same applies to Bio::Tools::Blast.pm
>
>       -hilmar
>
> -----Original Message-----
> From:	Jason Stajich [mailto:jason at cgt.duhs.duke.edu]
> Sent:	Thu 5/15/2003 6:30 AM
> To:	Hilmar Lapp
> Cc:	Bioperl
> Subject:	Re: [Bioperl-l] tests failing miserably
> The tree stuff needs me to back out my changes - I've still not been  
> able
> to make time to fix the bug I was intending to fix there.  This is  
> root of
> the molphy and tree tests failing.
>
> Somewhere a month or so ago some feature changes went in - that broke
> Bio::Tools::HMMer code for some reason.  I am actually a proponent of
> dropping those modules anyways if SearchIO::hmmer will fit the bill for
> most people.  Ewan wrote Tools::HMMer and I know old versions of it run
> Pfam, but I don't know if I really want to support 2 HMMer parsers
> (actually there 3 since the fugu group committed a quick hack in
> Bio::Tools::Hmmpfam).
>
> -jason
>
> On Wed, 14 May 2003, Hilmar Lapp wrote:
>
>> I've got Genpred.t, Pictogram.t, est2genome.t, and hmmer.t fixed. In
>> SeqFeature.t there is one test left that fails which I'm looking into,
>> everything else passes. I'm clueless wrt CoordinateMapper.t, Molphy.t,
>> and Tree.t, and being short of time I won't touch them. I'll give the
>> Sim4.t failures a try, unless someone else wants to take that on.
>>
>> 	-hilmar
>>
>> On Wednesday, May 14, 2003, at 03:16  PM, Hilmar Lapp wrote:
>>
>>> These failures have been there now since more than 2 weeks. Either
>>> this is specific to my system (perl 5.6.0, Mac OSX 10.2.5), or I'm
>>> playing the bad guy now pulling out the whistle. This is after a  
>>> fresh
>>> cvs update.
>>>
>>> Failed Test          Stat Wstat Total Fail  Failed  List of Failed
>>> --------------------------------------------------------------------- 
>>> --
>>> --------
>>> t/CoordinateMapper.t  255 65280   170    2   1.18%  169-170
>>> t/Genpred.t                        34   ??       %  ??
>>> t/Molphy.t            255 65280    17    0   0.00%  ??
>>> t/Pictogram.t           2   512     2    2 100.00%  1-2
>>> t/SeqFeature.t                     66    5   7.58%  13-14 17-18 61
>>> t/Sim4.t                           27   24  88.89%  35 38 41 44 52 55
>>> 58 61 69
>>>                                                     72 75 78 86 89 92
>>> 95 103
>>>                                                     106 109 112 120
>>> 123 126 129
>>> t/Tree.t                           10    4  40.00%  4-5 8-9
>>> t/est2genome.t                     60   25  41.67%  4 37-60
>>> t/hmmer.t                         134    2   1.49%  130 134
>>> 15 subtests skipped.
>>> Failed 9/149 test scripts, 93.96% okay. -39/6872 subtests failed,
>>> 100.57% okay.
>>>
>>> CoordinateMapper.t:
>>> t/CoordinateMapper...........ok 168/170Can't call method "start" on  
>>> an
>>> undefined value at blib/lib/Bio/Coordinate/Utils.pm line 136.
>>> t/CoordinateMapper...........dubious
>>>         Test returned status 255 (wstat 65280, 0xff00)
>>> DIED. FAILED tests 169-170
>>>         Failed 2/170 tests, 98.82% okay
>>>
>>> Genpred.t: I fixed this. (Msg: Don't know which tests failed: got 35
>>> ok, expected 34)
>>>
>>> Molphy.t:
>>> t/Molphy.....................ok 1/17Modification of non-creatable
>>> array value attempted, subscript -8 at
>>> blib/lib/Bio/TreeIO/TreeEventBuilder.pm line 242.
>>> t/Molphy.....................dubious
>>>         Test returned status 255 (wstat 65280, 0xff00)
>>>         after all the subtests completed successfully
>>>
>>> Pictogram.t: I fixed this. (no graceful skip if SVG.pm not installed)
>>>
>>> Seqfeature.t:
>>> not ok 13
>>> # Test 13 got: <UNDEF> (t/SeqFeature.t at line 103)
>>> #    Expected: 'exon'
>>> not ok 14
>>> # Test 14 got: <UNDEF> (t/SeqFeature.t at line 104)
>>> #    Expected: 'internal'
>>> not ok 17
>>> # Test 17 got: <UNDEF> (t/SeqFeature.t at line 107)
>>> #    Expected: 'other'
>>> not ok 18
>>> # Test 18 got: <UNDEF> (t/SeqFeature.t at line 108)
>>> #    Expected: 'program_a'
>>> not ok 61
>>> # Test 61 got: '1' (t/SeqFeature.t at line 284)
>>> #    Expected: '1595'
>>>
>>> Sim4.t:
>>> lots of warnings in the form
>>> -------------------- WARNING ---------------------
>>> MSG: Did not defined the number of conserved matches in the HSP
>>> assuming conserved == identical (118)
>>> ---------------------------------------------------
>>>
>>> Failures are all 0/1 mismatches like
>>> not ok 112
>>> # Test 112 got: '0' (t/Sim4.t at line 217)
>>> #     Expected: '1'
>>>
>>> Tree.t:
>>>
>>> not ok 4
>>> # Failed test 4 in t/Tree.t at line 76
>>> not ok 5
>>> # Failed test 5 in t/Tree.t at line 79
>>> not ok 8
>>> # Test 8 got: '-1' (t/Tree.t at line 100)
>>> #   Expected: '0'
>>> not ok 9
>>> # Test 9 got: '-1' (t/Tree.t at line 103)
>>> #   Expected: '1'
>>> -------------------- WARNING ---------------------
>>> MSG: Node requested for reroot is already the root node!
>>> ---------------------------------------------------
>>>
>>> est2genome.t:
>>>
>>> not ok 4
>>> # Failed test 4 in t/est2genome.t at line 56
>>>
>>> The rest of the failures stem from the fact that it only prints 36
>>> tests, not 60 as expected.
>>>
>>> hmmer.t:
>>>
>>> ok 121
>>>
>>> -------------------- WARNING ---------------------
>>> MSG: Could not find SEED 33 103 unit even though I am reading it in.
>>> ugh!
>>> ---------------------------------------------------
>>>
>>> -------------------- WARNING ---------------------
>>> MSG: Could not find SEED 124 194 unit even though I am reading it in.
>>> ugh!
>>> ---------------------------------------------------
>>> ok 122
>>>
>>> -------------------- WARNING ---------------------
>>> MSG: Could not find Methylase_M 280 481 unit even though I am reading
>>> it in. ugh!
>>> ---------------------------------------------------
>>>
>>> not ok 130
>>> # Test 130 got: <UNDEF> (t/hmmer.t at line 267)
>>> #     Expected: '-105.2'
>>> not ok 134
>>> # Test 134 got: '0' (t/hmmer.t at line 279)
>>> #     Expected: '604'
>>>
>>>
>>> Quite frankly I have little clue as to what is going on in those that
>>> I haven't fixed (all except 2). A couple of them look like old
>>> features that used to work, so it smells like a knock-on effect that
>>> some recent change caused. Maybe someone who changed something
>>> recently has an idea where to start?
>>>
>>> 	-hilmar
>>>
>>> --
>>> -------------------------------------------------------------
>>> Hilmar Lapp                            email: lapp at gnf.org
>>> GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
>>> -------------------------------------------------------------
>>>
>>> _______________________________________________
>>> Bioperl-l mailing list
>>> Bioperl-l at bioperl.org
>>> http://pw600a.bioperl.org/mailman/listinfo/bioperl-l
>>>
>>
>
> --
> Jason Stajich
> Duke University
> jason at cgt.mc.duke.edu
>
>
>
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at bioperl.org
> http://pw600a.bioperl.org/mailman/listinfo/bioperl-l
>



More information about the Bioperl-l mailing list