remove commented code from SeqIO::fasta, but modify test grammar and …

…get a 2x-3x increase in parsing
cjfields · Mar 26, 2017 · d1f1af2 · d1f1af2
1 parent 335c484
commit d1f1af2
Show file tree

Hide file tree

Showing 2 changed files with 11 additions and 30 deletions.
diff --git a/lib/Bio/Grammar/Fasta.pm6 b/lib/Bio/Grammar/Fasta.pm6
@@ -15,17 +15,17 @@ grammar Bio::Grammar::Fasta {
     token record {
         <description_line> <sequence>
     }
-    token description_line    {
-        ^^\> <identifier> <description>? \n
+    rule description_line    {
+        ^^\><identifier><.ws><description>?\n
     }
     token identifier   {
         #assume we going to parse NCBI specific id for reference number and gi numbers
         \S+
-    }    
+    }
     token description  {
         \N+
     }
-    token sequence     {
+    rule sequence     {
         <-[>]>+
     }
 }
diff --git a/lib/Bio/SeqIO/fasta.pm6 b/lib/Bio/SeqIO/fasta.pm6
@@ -15,7 +15,7 @@ class Bio::Grammar::Fasta::Actions::PrimarySeq {
             )
         );
     }
-    
+
     method sequence($/)             {
         $/.make(~$/.subst("\n", '', :g))
     }
@@ -26,37 +26,18 @@ class Bio::SeqIO::fasta does Bio::Role::SeqStream does Bio::Role::IO {
     has $!actions = Bio::Grammar::Fasta::Actions::PrimarySeq.new();
     has $.width = 60;
     has $.block = $!width;  # NYI
-    
+
     multi method initialize-io(:$!width?, :$fh?, :$file?, *%args) {
         # we reset the input record sep here
         callwith(:nl-in("\n>"), :$fh, :$file, |%args);
     }
-    
+
     # TODO: one key future optimization requires implementation in Rakudo:
     # 1) Grammar parsing of a stream of data (e.g. Cat), which is now considered
     #    a post-6.0 update
-    
+
     # 2) Make next-Seq() a multi, which allows it to take arguments for various iterators
-    #method next-Seq() {
-    #    return if $.fh.eof();
-    #    while $.fh.get -> $chunk {
-    #        #my ($header, $seq) = $chunk.split("\n", 2);
-    #        #$header.=subst(/^^\>/, '');
-    #        #my ($display_id, $desc) = $header.split(' ', 2);
-    #        #$seq.=subst("\n", '', :g);
-    #        #return Bio::PrimarySeq.new(
-    #        #    seq             => $seq,
-    #        #    description     => $desc || '',
-    #        #    display_id      => $display_id
-    #        #);
-    #        if $chunk !~~ /^^\>/ {
-    #            return Bio::Grammar::Fasta.parse( ">$chunk", actions => $!actions, rule => 'record').ast;
-    #        } else {
-    #            return Bio::Grammar::Fasta.parse( "$chunk", actions => $!actions, rule => 'record').ast;
-    #        }
-    #    }
-    #}
-
+
     method next-Seq() {
         return if $.fh.eof();
         while $.fh.get -> $chunk {
@@ -76,9 +57,9 @@ class Bio::SeqIO::fasta does Bio::Role::SeqStream does Bio::Role::IO {
             #}
         }
     }
-    
+
     method write-Seq(Bio::PrimarySeq $seq) {
         self.fh.say(sprintf(">%s %s", $seq.display_id, $seq.description));
         self.fh.say($seq.seq.subst( /(.** {$.width})/, { "$0\n" }, :g));
     }
-}
+}