#!/usr/bin/env python
import sys
import os
import argparse
import quikr
import numpy as np

def main():

    parser = argparse.ArgumentParser(description=
    "Quikr returns the estimated frequencies of batcteria present when given a \
    input FASTA file. \n \
    A default trained matrix will be used if none is supplied \n \
    You must supply a kmer and default lambda if using a custom trained \
    matrix.")

    parser.add_argument("-f", "--fasta", help="fasta file", required=True)
    parser.add_argument("-o", "--output", help="output path (csv output)", required=True)
    parser.add_argument("-t", "--trained-matrix", help="trained matrix", required=True)
    parser.add_argument("-l", "--lamb", type=int, help="the default lambda value is 10,000")
    parser.add_argument("-k", "--kmer", type=int, required=True,
        help="specifies which kmer to use, must be used with a custom trained database")


    args = parser.parse_args()
    
    # our default lambda is 10,000
    lamb = 10000

    # Make sure our input exist
    if not os.path.isfile(args.fasta):
        parser.error( "Input fasta file not found")

    if not os.path.isfile(args.trained_matrix):
        parser.error("Custom trained matrix not found")
    
    # use alternative lambda
    if args.lamb is not None:
        lamb = args.lamb
    
    trained_matrix = quikr.load_trained_matrix_from_file(args.trained_matrix)
    xstar = quikr.calculate_estimated_frequencies(args.fasta, trained_matrix, args.kmer, lamb)

    np.savetxt(args.output, xstar, delimiter=",", fmt="%f")
    return 0

if __name__ == "__main__":
    sys.exit(main())