#!/usr/bin/env python
# -*- coding=utf-8 -*-
import os
import glob
import commands
import argparse

from collections import OrderedDict


__version__ = '1.0'
__author__ = 'suqingdong'


def human_readable_size(size):

    if size > 1024 * 1024 * 1024:
        size = str(round(size / 1024. / 1024. / 1024., 1)) + 'G'
    elif size > 1024 * 1024:
        size = str(round(size / 1024. / 1024., 1)) + 'M'
    elif size > 1024:
        size = str(round(size / 1024., 1)) + 'K'
    elif 0 < size < 1024:
        size = str(size)
    else:
        print 'size must greater then zero'

    return size


def get_size(file, human_readable=True):

    if not os.path.exists(file):
        size = '\033[31;5mError\033[0m'
    elif os.path.getsize(file) == 0:
        size = '\033[31;5m0\033[0m'
    else:
        size = human_readable_size(os.path.getsize(file)) if human_readable else os.path.getsize(file)

    return size


def print_glob_size(path):

    path_list = [path] if type(path) != list else path

    for path in path_list:
        for f in glob.glob(path):
            f = f.lstrip('./')
            size = get_size(f)
            print '| {:6s}\t{}'.format(size, f)


def check_raw_clean(name, data_path='.'):

    print '\033[33m|{:=^100s} \033[0m'.format(' Checking %s ' % name)
    sample_nums = commands.getoutput('ls {}/{} | wc -l'.format(data_path, name))
    print '\033[323m|-- {}样本数: {}\033[0m'.format(name, sample_nums)

    # fq_nums = commands.getoutput('ls %s/%s/*/*.fq.gz | wc -l' % (data_path, name))
    fq_nums = len(glob.glob('%s/%s/*/*.fq.gz' % (data_path, name)))

    cmd = 'wc -l %s/%s/*/MD5.txt | grep -vE "(total)|(总用量)" | awk \'{print $1}\'' % (data_path, name)
    # print cmd
    md5_nums = sum(int(n) for n in commands.getoutput(cmd).split('\n'))

    print '|-- fq.gz文件数:', fq_nums
    print '|-- MD5总行数:', md5_nums

    if fq_nums != md5_nums:
        print '\033[31m|-- fq.gz文件数和MD5总行数不相等，请检查!!!\033[0m'
    elif fq_nums == 0:
        print '\033[31m|-- fq.gz文件不存在，请检查!!!\033[0m'

    print_glob_size('{}/{}/*/*.fq.gz'.format(data_path, name))

    for md5 in glob.glob('{}/{}/*/MD5.txt'.format(data_path, name)):
        md5 = md5.lstrip('./')
        md5_num = commands.getoutput('wc -l %s | awk \'{print $1}\'' % md5)
        md5_num = '\033[31;5m{}\033[0m'.format(md5_num) if md5_num == '0' else md5_num
        print '| {:6s}\t{}'.format(md5_num, md5)


def check_bam(name='Mapping', data_path='Mapping'):

    print '\033[33m|{:=^100s} \033[0m'.format(' Checking %s ' % name)
    sample_nums = commands.getoutput('ls -l {} | grep "^d" | wc -l'.format(data_path))
    print '\033[323m|-- Mapping样本数: {}\033[0m'.format(sample_nums)

    print_glob_size('{}/*/*.bam*'.format(data_path))


def check_variation(name='Variation', data_path='Variation'):

    print '\033[33m|{:=^100s} \033[0m'.format(' Checking %s ' % name)
    sample_nums = commands.getoutput('ls -l {} | grep "^d" | wc -l'.format(data_path))
    print '\033[323m|-- Variation样本数: {}\033[0m'.format(sample_nums)

    for each in ('SNP', 'InDel', 'SV', 'CNV'):
        if glob.glob('{}/*/{}'.format(data_path, each)):
            print '|-- checking %s ...' % each
            print_glob_size([
                '{}/*/{}/*.*'.format(data_path, each),
                '{}/*/{}/Circos/*'.format(data_path, each),
            ])


def check_advance(name='Advance', data_path='Advance'):

    print '\033[33m|{:=^100s} \033[0m'.format(' Checking %s ' % name)

    analysis_items = sorted(str(d) for d in os.listdir(data_path) if 'Total' not in d)

    print '|-- analysis items:\n|  ' + '\n|  '.join(analysis_items)

    # Other analysis check needs to add here
    analysis_map = OrderedDict()
    analysis_map['FilterDB'] = [
        data_path + '/*FilterDB*/VCF/*',
        data_path + '/*FilterDB*/Filter/*/*'
    ]
    for each in ('ACMG', 'FilterCNV_SV', 'ModelF'):
        analysis_map[each] = data_path + '/*{}*/*/*'.format(each)
    analysis_map['Denovo'] = [
        data_path + '/*Denovo*/SNP_INDEL/Denovo*/*/*/*',
        data_path + '/*Denovo*/SNP_INDEL/IntersectResult/*xls',
        data_path + '/*Denovo*/SNP_INDEL/IntersectResult/*/*',
        data_path + '/*Denovo*/CNV_SV/*/*/*'
    ]
    for each in ('Noncoding', 'Network', 'PPI', 'Share'):
        analysis_map[each] = data_path + '/*{}*/*'.format(each)
    analysis_map['Pathway'] = [
        data_path + '/*Pathway*/*.*',
        data_path + '/*Pathway*/KEGG_maps/*.*',
        data_path + '/*Pathway*/KEGG_maps/png/*'
    ]
    analysis_map['Customiezd'] = data_path + '/*Customiezd*/*.*'

    for analysis_detail in analysis_items:
        for analysis in analysis_map:
            if analysis in analysis_detail:
                print '\033[33m|-- checking %s ...\033[0m' % analysis_detail
                print_glob_size(analysis_map[analysis])


def check_46(check_all=False):

    print '\033[36m|=== Check result for disease pipeline 4.6 ...\033[0m'

    # RawData and CleanData
    for each in ('RawData', 'CleanData'):
        if each in os.listdir('ReleaseResult/Data'):
            check_raw_clean(each, 'ReleaseResult/Data')

    # BamData
    if 'BamData' in os.listdir('ReleaseResult/Data'):
        check_bam('BamData', 'ReleaseResult/Data/BamData')

    # SampleVariation
    if 'SampleVariation' in os.listdir('ReleaseResult/PrimaryAnalysis'):
        check_variation('SampleVariation', 'ReleaseResult/PrimaryAnalysis/SampleVariation')

    # FilterAnalysis
    if 'FilterAnalysis' in os.listdir('ReleaseResult/PrimaryAnalysis'):
        print '\033[33m|{:=^100s} \033[0m'.format(' Checking FilterAnalysis ')
        print_glob_size('ReleaseResult/PrimaryAnalysis/FilterAnalysis/*/*/*')

    # FinalResult
    if 'FinalResult' in os.listdir('ReleaseResult'):

        check_advance('FinalResult', 'ReleaseResult/FinalResult')

        print '\033[33m|-- checking Total.candidate_gene.xls ...\033[0m'
        print_glob_size('ReleaseResult/FinalResult/*xls')

    # 4.6的Advance目录只用于信息自查，不释放
    if check_all:
        print '|' + '-' * 100
        print '|\033[33m{:-^100s}\033[0m'.format(' The below directories will not be released, just used for self checking! ')
        if 'Advance' in os.listdir('.'):
            check_advance()


def check_45():

    print '\033[36m|=== Check result for disease pipeline 4.5 ...\033[0m'

    directories = os.listdir('.')

    # RawData and QC
    for each in ('RawData', 'QC'):
        if each in directories:
            check_raw_clean(each)

    # Mapping
    if 'Mapping' in directories:
        check_bam()

    # Variation
    if 'Variation' in directories:
        check_variation()

    # Advance
    if 'Advance' in directories:
        check_advance()


def main():

    result_path = os.path.abspath(args.get('dir'))
    check_all = args.get('all')

    os.chdir(result_path)
    print '\033[32m|=== Release Directory: %s\033[0m' % os.getcwd()

    if 'ReleaseResult' in os.listdir('.'):
        check_46(check_all)
    else:
        check_45()


if __name__ == "__main__":

    parser = argparse.ArgumentParser(
        prog='Check',
        description='Check Result directory for disease pipeline',
        epilog='Contact: {0} <{0}@novogene.com>'.format(__author__),
        formatter_class=argparse.RawTextHelpFormatter,
        version=__version__
    )

    parser.add_argument('-d', '--dir', help='The data release path to check [default="%(default)s"]', default='.')
    parser.add_argument('-a', '--all', help='Whether check Advance directory or not for 4.6 pipeline', action='store_true')

    args = vars(parser.parse_args())

    main()