[Bioperl-l] Makefile; Funkiness with _rearrange issues and - on alignments

[Ewan Birney]

I am in serious 1.2 debugging mode. There are actually alot of failing
tests on my platform at the least (alpha, 5.6.1). Somethings:

Makefile
--------

I have put in the suggested fix from the MakeMaker team and I believe this
means that our Makefile.PL will work on all platforms. Could other Irix
and Dec people check this out.


Seq/Alphabet


Lincoln and Hilmar - I guess you haven't been running make test across the
entire platform when you were changing _rearrange (bad, bad).


If you had you would have noticed that AlignIO was mysteriously failing on
an internal sequence. after *alot* of messing around I discovered that
Bio::PrimarySeq() called rearrange- someone thought it was a good idea to
be able to pass in arguments like

   $pseq_obj->seq( -seq => $seq_string, -alphabet => 'dna');

however, most people passed arguments like:

   $pseq_obj->seq($seq_string)

which works fine *except* for cases like 

   >AKH_HAEIN/114-431
   -----------------VEDAVKATIDCRGEKLSIAMMKAWFEARGY-S--VHIVDPVKQ

which, with new the fixes to _rearrange triggered the
anti-misnumbered-rearrange fix just popped in.


After some humming and harring I decided that seq on primary_seq should
not be this leathly complicated argument detecting thing, and removed it.


  Does anyone *USE* the syntax

   $primary_seq_object(-seq => $seq_as_string, -alphabet => 'dna');

(notice that lots of people use this for new, but not for seq).



Fix is committed, which fixed lots of AlignIO and BLAST test failing.


I still have the following tests failing which I will get around to
tomorrow if noone else beats me to it.



Failed Test    Stat Wstat Total Fail  Failed  List of Failed
-------------------------------------------------------------------------------
t/BlastIndex.t   -1    11    13   11  84.62%  3-13
t/Index.t        79 20224    12   12 100.00%  1-12
t/PhylipDist.t                6   ??       %  ??
5 subtests skipped.
Failed 3/132 test scripts, 97.73% okay. 11/5854 subtests failed, 99.81%
okay.





-----------------------------------------------------------------
Ewan Birney. Mobile: +44 (0)7970 151230, Work: +44 1223 494420
<birney@ebi.ac.uk>. 
-----------------------------------------------------------------