#!/usr/bin/env ruby
require 'optparse'
require 'genevalidator'
require 'genevalidator/version'

opt = {}
opt_parser = OptionParser.new do |opts|
  opts.banner = <<BANNER

USAGE:
    $ genevalidator [OPTIONS] Input_File

ARGUMENTS:
    Input_File: Path to the input fasta file containing the predicted sequences.

OPTIONAL ARGUMENTS
BANNER

  opt[:validations] = ['all']
  opts.on('-v', '--validations <String>', Array,
          'The Validations to be applied.',
          'Validation Options Available (separated by coma):',
          '  all   = All validations (default),',
          '  lenc  = Length validation by clusterization,',
          '  lenr  = Length validation by ranking,',
          '  merge = Analyse gene merge,',
          '  dup   = Check for duplications,',
          '  frame = Open reading frame (ORF) validation,',
          '  orf   = Main ORF validation,',
          '  align = Validating based on multiple alignment') do |lst|
    opt[:validations] = lst
  end

  opt[:db] = 'swissprot -remote'
  opts.on('-d', '--db [BLAST_DATABASE]',
          'Path to the BLAST database',
          'GeneValidator also supports remote databases:',
          'e.g.   $ genevalidator -d "swissprot -remote" Input_File') do |db|
    opt[:db] = db
  end

  opts.on('-x', '--blast_xml_file [FILE]',
          'Provide GeneValidator with a pre-computed BLAST XML output',
          'file (BLAST -outfmt option 5).') do |blast_xml_file|
    opt[:blast_xml_file] = blast_xml_file
  end

  opts.on('-t', '--blast_tabular_file [FILE]',
          'Provide GeneValidator with a pre-computed BLAST tabular output',
          'file. (BLAST -outfmt option 6).') do |blast_tabular_file|
    opt[:blast_tabular_file] = blast_tabular_file
  end

  # default blast tabular columns
  opts.on('-o', '--blast_tabular_options [Options]',
          'Custom format used in BLAST -outfmt argument',
          'See BLAST+ manual pages for more details') do |blast_table_options|
    opt[:blast_tabular_options] = blast_table_options
  end

  opt[:num_threads] = 1
  opts.on('-n', '--num_threads num_of_threads',
          'Specify the number of processor threads to use when running',
          'BLAST and Mafft within GeneValidator.') do |num_threads|
    opt[:num_threads] = num_threads
  end

  opt[:fast] = false
  opts.on('-f', '--fast',
          'Run BLAST on all sequences together (rather than separately)',
          'to speed up the analysis.',
          'However, this means that there will be a longer wait before the',
          'results can be viewed (as GeneValidator will need to run BLAST',
          'on all sequences before producing any results).',
          'The speed difference will be more apparent on larger input files') do
    opt[:fast] = true
  end

  opts.on('-r', '--raw_sequences [raw_seq]',
          'Supply a fasta file of the raw sequences of all BLAST hits present',
          'in the supplied BLAST XML or BLAST tabular file.') do |raw_seq|
    opt[:raw_sequences] = raw_seq
  end

  opts.on('-m', '--mafft_bin [MAFFT_PATH]',
          'Path to MAFFT bin folder (is added to $PATH variable)') do |mafft|
    opt[:mafft_bin] = mafft
  end

  opts.on('-b', '--blast_bin [BLAST_PATH]',
          'Path to BLAST+ bin folder (is added to $PATH variable)') do |blast|
    opt[:blast_bin] = blast
  end

  opts.on('--version', 'The version of GeneValidator that you are running.') do
    puts GeneValidator::VERSION
    exit
  end

  opts.on('-h', '--help', 'Show this screen.') do
    puts opt_parser
    exit
  end
end

begin
  opt_parser.parse!(ARGV)
  if ARGV.length > 1
    $stderr.puts 'Error: you must specify a single fasta input file instead' \
                 ' of ' + ARGV.length.to_s + ".\n"
    exit 1
  elsif ARGV.length == 0
    puts opt_parser
    exit 1
  end
rescue OptionParser::ParseError
  $stderr.print 'Error: ' + $ERROR_INFO.to_s + '\n'
  exit 1
end

start = Time.now
opt[:input_fasta_file] = ARGV[0]

(GeneValidator::Validation.new(opt)).run

puts "Total running time: #{(Time.now - start).round(3)}s"
puts # a blank line