Scans for all zipped files within a directory (recursively), opens archives and reads the csv files inside the archives. Append all csv files into a single one named 'output.csv'. Useful for creating a single file from thousands of different csv files. Skips headers.

merge_csv.py

import os
import fnmatch
import zipfile
import StringIO
import csv

print('Scanning for files ...')
count = 0
for r, d, f in os.walk(os.path.dirname(os.path.abspath(__file__))):
		for fn in fnmatch.filter(f, '*.zip'):
			count = count + 1
print('Found '+str(count) +' zipped files')
buff = raw_input('Press <Enter> to process all zipped files')
print('Generating single file (might take a while) ...')

progress = 0
with open('output.csv', 'wb') as o:
	for root, dirnames, filenames in os.walk(os.path.dirname(os.path.abspath(__file__))):
		for filename in fnmatch.filter(filenames, '*.zip'):
			with open(os.path.join(root, filename), 'rb') as filehandle:
				with zipfile.ZipFile(filehandle) as zfile:
					dataFile = filename.replace(filename[-3:], 'csv')
					data = StringIO.StringIO(zfile.read(dataFile))
					reader = csv.reader(data)

			next(reader, None)

			progress = progress + 1
			a = csv.writer(o, delimiter=',', lineterminator='\n')
			for row in reader:
				a.writerow(row)

			print('\n') * 100
			print('Writing file number '+ str(progress) +'/'+str(count)+' ...')

print('Completed. Enjoy.')