#! /usr/bin/ruby
#
# Author:: Pjotr Prins
# Copyright:: August 2010
# License:: Ruby License
#
# Copyright (C) 2010,2011 Pjotr Prins <pjotr.prins@thebird.nl>


USAGE = <<EOM
  Fetch and assemble mRNAs, or CDS and print in FASTA format. 

    gff3-fetch [--no-cache] mRNA|CDS [filename.fa] filename.gff

  Where:

    --no-cache      : do not load everything in memory (slower)
    mRNA            : assemble mRNA
    CDS             : assemble CDS 

  Multiple GFF3 files can be used. For external FASTA files, always the last
  one before the GFF file is used.

  Examples:

    Find mRNA and CDS information from test.gff3 (which includes sequence information)

      gff3-fetch mRNA test/data/gff/test.gff3
      gff3-fetch CDS test/data/gff/test.gff3

    Find CDS from external FASTA file

      gff3-fetch CDS test/data/gff/MhA1_Contig1133.fa test/data/gff/MhA1_Contig1133.gff3

    Find mRNA from external FASTA file, without loading everything in RAM

      gff3-fetch --no-cache mRNA test/data/gff/test-ext-fasta.fa test/data/gff/test-ext-fasta.gff3   

  If you use this software, please cite http://dx.doi.org/10.1093/bioinformatics/btq475

== Copyright

Copyright (C) 2010,2011 Pjotr Prins <pjotr.prins@thebird.nl> 

EOM

rootpath = File.dirname(File.dirname(__FILE__))
$: << rootpath+'/lib'
$: << rootpath+'/../bioruby/lib'

require 'bio-gff3'

$stderr.print "BioRuby GFF3 Plugin Copyright (C) 2010,2011 Pjotr Prins <pjotr.prins@thebird.nl>\n\n"

if ARGV.size == 0
  print USAGE
end

gfftype = ARGV.shift
caching = true
if gfftype == "--no-cache"
  caching = false
  gfftype = ARGV.shift
end
raise "Unknown GFF type '#{gfftype}'" if gfftype !~ /mrna|cds|exon/i

fastafn = nil

ARGV.each do | fn |
  if File.extname(fn) =~ /fa|fas|fasta/i
    fastafn = fn
    next
  end
  options = {:validate => false}
  options = {:validate => false, :cache_components => :cache_none, :cache_records => :cache_none} if caching == false
  options[:fasta_filename] = fastafn if fastafn

  gffdb = Bio::GFFbrowser::GFFdb.new(fn,options)
  gff = gffdb.assembler
  case gfftype.downcase
    when 'mrna' 
          gff.each_mRNA_seq do | id, seq |
            puts ">"+id
            puts seq
          end
    when 'exon'
          gff.each_exon_seq do | id, seq |
            puts ">"+id
            puts seq
          end
    when 'cds'
          gff.each_CDS_seq do | id, seq |
            puts ">"+id
            puts seq
          end
    else
      raise "Unknown action <#{gfftype}>"
  end
end