$: << File.expand_path(File.join(File.dirname(__FILE__)))

require 'scbi_mapreduce'
require 'scbi_blast'
require 'json'
require 'sequence'
require 'fl_string_utils'
require "lcs" # like the class simliar of seqtrim, return the longest common sequence
require "test_code"

require 'fl_analysis'
include FlAnalysis

require 'nc_rna'
include NcRna

class MyWorker < ScbiMapreduce::Worker

	def starting_worker
		
		# $WORKER_LOG.info "Loading actions"
		rescue Exception => e
		puts (e.message+ e.backtrace.join("\n"))
		
	end
		
	def receive_initial_config(obj)
		
		# Reads the parameters
		# $WORKER_LOG.info "Params received: #{obj.to_json}"
		@options = obj
		
	end
		
	def process_object(obj)

		full_lenghter2(obj)
		return obj
		
	end
		
	def closing_worker
		 
	end
	
	# ejecuta blast utilizando los parametros fichero de entrada, base de datos, fichero de salida y tipo de blast
	def run_blast(input, database, blast_type, evalue)
		
		blast=BatchBlast.new("-db #{database}",blast_type,"-evalue #{evalue} -max_target_seqs 1")
		blast_result = blast.do_blast_seqs(input, :xml)
		
		return blast_result
	end
	
	
	def full_lenghter2(seqs)
		
		# -------------------------------------------- User database
		# if the user has included his own database in the parameters entry,
		# the location of the database is tested, and blast and the results analysis is done
		
		if (@options[:user_db])
		
			if (@options[:user_db] =~ /\//)
				user_db_name = @options[:user_db].sub(/.+\//,'')
			end
		
			if !File.exists?("#{File.expand_path(@options[:user_db])}.psq")
				puts "user database: #{@options[:user_db]} was not found"
				exit
			end
		
			# do blast
			my_blast = run_blast(seqs, "#{@options[:user_db]}", 'blastx', '1e-6')
			
			# split and parse blast
			seqs.each_with_index do |seq,i|
				analiza_orf_y_fl(seq, my_blast.querys[i], @options, user_db_name)
			end
			
			new_seqs=seqs.select{|s| s.get_annotations(:complete).empty?}
			
		else
			new_seqs = seqs
		end
		
		# -------------------------------------------- UniProt (sp)
		# blast
		sp_path=File.join("sp_#{@options[:tax_group]}","sp_#{@options[:tax_group]}.fasta")
		my_blast = run_blast(new_seqs, sp_path, 'blastx', '1e-6')
		
		# split and parse blast
		new_seqs.each_with_index do |seq,i|
			analiza_orf_y_fl(seq, my_blast.querys[i], @options, "sp_#{@options[:tax_group]}")
		end
		
		new_seqs=seqs.select{|s| s.get_annotations(:complete).empty?}
		
		# -------------------------------------------- UniProt (tr)
		# blast
		tr_path=File.join("tr_#{@options[:tax_group]}","tr_#{@options[:tax_group]}.fasta")
		my_blast = run_blast(new_seqs, tr_path, 'blastx', '1e-6')
		
		# split and parse blast
		new_seqs.each_with_index do |seq,i|
			analiza_orf_y_fl(seq, my_blast.querys[i], @options, "tr_#{@options[:tax_group]}")
		end
		
		# -------------------------------------------- Test Code
		# the sequences without a reliable similarity with an orthologue are processed with Test Code
		testcode_input=seqs.select{|s| !s.get_annotations(:apply_tcode).empty?}

# active this line to test tcode, and comment all lines above in this function
# testcode_input=seqs
	
		testcode_input.each do |seq|
			TestCode.new(seq)
		end
		
		# -------------------------------------------- nc RNA
		unknown_seqs=seqs.select{|s| !s.get_annotations(:tcode_unknown).empty?}
		# run blastn
		ncrna_path=File.join('nc_rna_db','ncrna_fln_100.fasta')
		my_blast = run_blast(unknown_seqs, ncrna_path, 'blastn', '1e-3')
		
		# split and parse blast
		unknown_seqs.each_with_index do |seq,i|
			find_nc_rna(seq, my_blast.querys[i])
		end
		# ---------------------------------------------------
		
	end
	
end