I had a similar requirement and so I wrote the following class
import os
import pathlib
import glob
import csv
from collections import defaultdict
class FileCsvExport:
"""Generate a CSV file containing the name and contents of all files found"""
def __init__(self, directory: str, output: str, header = None, file_mask = None, walk_sub_dirs = True, remove_file_extension = True):
self.directory = directory
self.output = output
self.header = header
self.pattern = '**/*' if walk_sub_dirs else '*'
if isinstance(file_mask, str):
self.pattern = self.pattern + file_mask
self.remove_file_extension = remove_file_extension
self.rows = 0
def export(self) -> bool:
"""Return True if the CSV was created"""
return self.__make(self.__generate_dict())
def __generate_dict(self) -> defaultdict:
"""Finds all files recursively based on the specified parameters and returns a defaultdict"""
csv_data = defaultdict(list)
for file_path in glob.glob(os.path.join(self.directory, self.pattern), recursive = True):
path = pathlib.Path(file_path)
if not path.is_file():
continue
content = self.__get_content(path)
name = path.stem if self.remove_file_extension else path.name
csv_data[name].append(content)
return csv_data
@staticmethod
def __get_content(file_path: str) -> str:
with open(file_path) as file_object:
return file_object.read()
def __make(self, csv_data: defaultdict) -> bool:
"""
Takes a defaultdict of {k, [v]} where k is the file name and v is a list of file contents.
Writes out these values to a CSV and returns True when complete.
"""
with open(self.output, 'w', newline = '') as csv_file:
writer = csv.writer(csv_file, quoting = csv.QUOTE_ALL)
if isinstance(self.header, list):
writer.writerow(self.header)
for key, values in csv_data.items():
for duplicate in values:
writer.writerow([key, duplicate])
self.rows = self.rows + 1
return True
Which can be used like so
...
myFiles = r'path/to/files/'
outputFile = r'path/to/output.csv'
exporter = FileCsvExport(directory = myFiles, output = outputFile, header = ['File Name', 'Content'], file_mask = '.txt')
if exporter.export():
print(f"Export complete. Total rows: {exporter.rows}.")
In my example directory, this returns
Export complete. Total rows: 6.
Note: rows
does not count the header if present
This generated the following CSV file:
"File Name","Content"
"Test1","This is from Test1"
"Test2","This is from Test2"
"Test3","This is from Test3"
"Test4","This is from Test4"
"Test5","This is from Test5"
"Test5","This is in a sub-directory"
Optional parameters:
header
: Takes a list of strings that will be written as the first line in the CSV. Default None
.
file_mask
: Takes a string that can be used to specify the file type; for example, .txt
will cause it to only match .txt
files. Default None
.
walk_sub_dirs
: If set to False, it will not search in sub-directories. Default True
.
remove_file_extension
: If set to False, it will cause the file name to be written with the file extension included; for example, File.txt
instead of just File
. Default True
.