class CSVUtils:
def __init__(self, dir_path, file_name=None):
self.validator = Validator()
self.validator.is_folder(dir_path)
self.file_name = file_name
self.dir_path = dir_path
self.delimiter = ';'
self.sep = ';'
def get_data(self, has_header=None, header_line=None, transpose=False):
# this funcion helps to retrive data from csv
file = self.create_name()
self.validator.is_csv(file)
header = []
df = pd.read_csv(file, header=header_line, delimiter=self.delimiter)
if transpose is True:
df = df.T
data = df.values.tolist()
if has_header:
header = np.transpose(data[0])
data.pop(0)
data = np.transpose(data)
return data, header
def save(self,
data,
header_read=None,
header_insert=False,
float_format=Constants().get_default_float_format(),
index=False):
# this function help to save data in a CSV
file = self.create_name()
is_file = self.validator.check_is_csv(file)
if is_file:
self.validator.is_csv(file)
df = pd.read_csv(file,
header=header_read,
delimiter=self.delimiter)
df.loc[len(df)] = data
df.to_csv(file,
sep=self.sep,
float_format=float_format,
header=header_insert,
index=index)
else:
df = pd.DataFrame(data).T
df.to_csv(file,
sep=self.sep,
float_format=float_format,
header=header_insert,
index=index)
def close_csv(self, file=None):
# this function perform a CSV clousure
if file is None:
file = self.create_name()
pd.set_option('display.precision', 15)
df = pd.read_csv(file,
header=None,
delimiter=self.delimiter,
dtype=np.float64)
df[0] = df[0].astype(int)
index = 1
columns_number = len(df.columns)
while index < columns_number:
df[index] = df[index].astype(np.float64)
index = index + 1
df.to_csv(file, sep=self.sep, header=False, index=False)
def create_name(self, file_name=None):
# utils to return file name
if file_name is None:
file_name = self.file_name
dir_path = self.dir_path
ext = '.' + file_name.split('.')[-1].lower()
if ext != Constants().get_csv_format():
file = dir_path + file_name + Constants().get_csv_format()
else:
file = dir_path + file_name
return file
def get_csv_files(self, dir_path=None):
# utils to read all CSV in a folder
if dir_path is None:
dir_path = self.dir_path
return [
f for f in listdir(dir_path)
if self.validator.check_is_csv(join(dir_path, f))
]
def generate_pairs(self):
files = self.get_csv_files()
return [(files[i], files[j]) for i in range(len(files))
for j in range(i + 1, len(files))]
def generate_combinations(self):
pairs = self.generate_pairs()
all_files_tuple = tuple(self.get_csv_files())
pairs.append(all_files_tuple)
return pairs
def merge_csv(self,
csv_tuple,
final_dir_path,
final_name,
float_format=Constants().get_default_float_format()):
# merge differnte CSV in one
df_list = []
final_name = final_dir_path + final_name
for csv in csv_tuple:
file = self.create_name(csv)
df = pd.read_csv(file,
header=None,
delimiter=self.delimiter,
dtype=np.float64)
df_list.append(df)
final_df = pd.concat(df_list)
final_df.to_csv(final_name,
sep=self.sep,
float_format=float_format,
header=False,
index=False)
self.close_csv(final_name)
def generate_csv_merged_name(self, csv_tuple):
name = ''
for file_name in csv_tuple:
name += file_name.split(Constants().get_csv_format())[0].upper()
return name + Constants().get_csv_format()
def merge_csv_combinations(self, final_dir_path):
# merge all CSV combinations in a folder
csv_list = self.generate_combinations()
for csv_tuple in csv_list:
name = self.generate_csv_merged_name(csv_tuple)
self.merge_csv(csv_tuple=csv_tuple,
final_dir_path=final_dir_path,
final_name=name)
