utils.py

'''
Copyright (C) e2its - All Rights Reserved
 * Unauthorized copying of this file, via any medium is strictly prohibited
 * Proprietary and confidential
 *
 * This file is part of gDayF project.
 *
 * Written by Jose L. Sanchez <e2its.es@gmail.com>, 2016-2019
'''

from hashlib import md5 as md5
from hashlib import sha256 as sha256
from pandas import read_json
from json import dumps
from copy import deepcopy
from numpy.random import rand
from gdayf.common.dfmetada import compare_dict


def hash_key(hash_type, filename):
    if hash_type == 'MD5':
        try:
            return md5(open(filename, 'rb').read()).hexdigest()
        except IOError:
            raise
    elif hash_type == 'SHA256':
        try:
            return sha256(open(filename, 'rb').read()).hexdigest()
        except IOError:
            raise



def decode_json_to_dataframe(json_string, orient='split'):
        return read_json(json_string, orient=orient)



def decode_ordered_dict_to_dataframe(ordered_dict, orient='split'):
    json_string = dumps(ordered_dict)
    return decode_json_to_dataframe(json_string, orient='split')



def compare_list_ordered_dict(list1, list2):
    if list1[0] is None or list2[0] is None:
        return list1[0] is None and list2[0] is None
    elif len(list1) != len(list2):
        return False
    else:
        for i in range(0, len(list1)):
            if not compare_dict(list1[i], list2[i]):
                return False
        return True



def compare_sorted_list_dict(list1, list2):
    if list1 is None or list2 is None:
        return list1 is None and list2 is None
    elif len(list1) != len(list2):
        return False
    else:
        for i in range(0, len(list1)):
            if not (list1[i] == list2[i]):
                return False
        return True

def get_model_fw(model):
    return list(model['model_parameters'].keys())[0]



def get_model_ns(model):
    return deepcopy(model['normalizations_set'])



def pandas_split_data(df, train_perc=0.9):
    df['train'] = rand(len(df)) < train_perc
    train = df[df.train == 1].drop('train', axis=1)
    test = df[df.train == 0].drop('train', axis=1)
    return train, test



def xstr(s):
    if s is None:
        return ''
    return str(s)