From 563e929ab685d91bcf13842c2feb03861a4368df Mon Sep 17 00:00:00 2001
From: Chris Fields <cjfields@bioperl.org>
Date: Tue, 3 Aug 2010 15:05:01 -0500
Subject: [PATCH] [bug 2984] end was calculated based on arbitrary symbols;
 punt and let LocatableSeq do this, for consistency

---
 Bio/AlignIO/phylip.pm | 12 ++++++------
 Bio/LocatableSeq.pm   |  7 +++++++
 Changes               |  4 +++-
 t/AlignIO/phylip.t    | 16 +++++++++++++++-
 4 files changed, 31 insertions(+), 8 deletions(-)

diff --git a/Bio/AlignIO/phylip.pm b/Bio/AlignIO/phylip.pm
index 3707b02c6..efe22d30d 100644
--- a/Bio/AlignIO/phylip.pm
+++ b/Bio/AlignIO/phylip.pm
@@ -44,7 +44,7 @@ Bio::AlignIO::phylip - PHYLIP format sequence input/output stream
 =head1 DESCRIPTION
 
 This object can transform Bio::SimpleAlign objects to and from PHYLIP
-fotmat. By deafult it works with the interleaved format. By specifying
+format. By default it works with the interleaved format. By specifying
 the flag -interleaved =E<gt> 0 in the initialization the module can
 read or write data in sequential format.
 
@@ -226,8 +226,8 @@ sub next_aln {
 	    $str =~ s/\s//g;
 	    $count = scalar @names;
 	    $hash{$count} .= $str;
-	} elsif( $entry =~ /^(.{$idlen})\s+(.*)\s$/ ||
-		 $entry =~ /^(.{$idlen})(\S{$idlen}\s+.+)\s$/ # Handle weirdnes s when id is too long
+	} elsif( $entry =~ /^(.{$idlen})\s*(.*)\s$/ ||
+		 $entry =~ /^(.{$idlen})(\S{$idlen}\s+.+)\s$/ # Handle weirdness when id is too long
 		 ) {
 	    $name = $1;
 	    $str = $2;
@@ -289,8 +289,8 @@ sub next_aln {
 	    $seqname=$name;
 	    $start = 1;
 	    $str = $hash{$count};
-	    $str =~ s/[^A-Za-z]//g;
-	    $end = length($str);
+#	    $str =~ s/[^A-Za-z]//g;
+	    #$end = length($str);
 	}
 	# consistency test
 	$self->throw("Length of sequence [$seqname] is not [$residuecount] it is ".CORE::length($hash{$count})."! ")
@@ -299,7 +299,7 @@ sub next_aln {
 	$seq = Bio::LocatableSeq->new('-seq'           => $hash{$count},
 				      '-display_id'    => $seqname,
 				      '-start'         => $start,
-				      '-end'           => $end,
+				      (defined $end) ? ('-end'           => $end) : (),
 				      '-alphabet'      => $self->alphabet,
 				      );
 	$aln->add_seq($seq);
diff --git a/Bio/LocatableSeq.pm b/Bio/LocatableSeq.pm
index f48aa8e57..942b4801e 100644
--- a/Bio/LocatableSeq.pm
+++ b/Bio/LocatableSeq.pm
@@ -224,6 +224,13 @@ sub _ungapped_len {
     return CORE::length($string)/($map_res/$map_coord) + $offset/($map_coord/$map_res);
 }
 
+#sub length {
+#    my $self = shift;
+#    return unless my $string = $self->seq;
+#    $string =~ s{[$GAP_SYMBOLS$FRAMESHIFT_SYMBOLS]+}{}g;
+#    return CORE::length($string);
+#}
+
 =head2 strand
 
  Title   : strand
diff --git a/Changes b/Changes
index b3097a1dd..c2475a99c 100644
--- a/Changes
+++ b/Changes
@@ -41,7 +41,9 @@ CPAN releases are branched from 'master'.
     * bug 3120 - bp_seqfeature_gff3.pl '-y' option [genehack, David Breimann]
     * bug 3126 - catch description [Toshihiko Akiba]
     * bug 2983 - fix score/percent ID mixup [Alexie Papanicolaou]
- 
+    * bug 2984 - let LocatableSeq decide on length of phylip aln [Adam Witney, 
+                 cjfields]
+
     [Deprecated]
     * Bio::Expression modules - these were originally designed to go with the
       bioperl-microarray suite of tools, however they have never been completed
diff --git a/t/AlignIO/phylip.t b/t/AlignIO/phylip.t
index 5d44fc013..be4cf044a 100644
--- a/t/AlignIO/phylip.t
+++ b/t/AlignIO/phylip.t
@@ -7,7 +7,7 @@ BEGIN {
 	use lib '.';
     use Bio::Root::Test;
     
-    test_begin(-tests => 11);
+    test_begin(-tests => 16);
 	
 	use_ok('Bio::AlignIO::phylip');
 }
@@ -31,6 +31,20 @@ $strout = Bio::AlignIO->new(
 $status = $strout->write_aln($aln);
 is $status, 1, "phylip output test";
 
+# check the LocatableSeq start/end/strand etc
+my $ls = $aln->get_seq_by_pos(2);
+is($ls->display_id, 'Pan_panisc');
+is($ls->start, 1);
+is($ls->end,47);
+
+# bug 2984
+TODO: {
+    local $TODO = 'problems with default strand, length?';
+    # shouldn't this be 0?
+    is($ls->strand,0);
+    is($ls->length,47);
+}
+
 # PHYLIP sequential/non-interleaved
 $strout = Bio::AlignIO->new('-file'  => test_input_file('noninterleaved.phy'),
 			    '-format' => 'phylip');
-- 
2.11.4.GIT