What is the best way to sort a csv file in python with multiple columns on date?

Import your file using the pandas library, which will likely be less effort than using pure python.

"Note that sorted() is applied to each line and not to the column entries. You would need to import the operator module and use operator.itemgetter(). Since the entries in a CSV file are strings, the sorting will place a nmber like "90" as 'higher' than a nuber like "864". This should not affect your dates, but would cause a problem with values. This would have to be handled seperately. I have not yet examined sorted() to determine if this can be avoided without conversion. import csv import operator ifile =open('myfile.csv', 'rb') infile = csv.reader(ifile) # The first entry is the header line infields = infile.next() statindex = infields.index('Desired Header') # create the sorted list sortedlist = sorted(infile, key=operator.itemgetter(statindex), reverse=True) ifile.close # open the output file - it can be the same as the input file ofile = open('myoutput.csv, 'wb') # write the header outfile.writerow(infields) # write the sorted list for row in sortedlist: outfile.writerow(row) # processing finished, close the output file ofile.close() UPDATE: If any of the rows in the entire file contain non-numeric data, then the sorted MUST sort by a text method. This includes if the header line is not row 1 (for example follows a title line). or if there is a blank entry in the column. If all entries in that column are numeric then it can be sorted as such. For example: try: sortedlist = sorted(infile, key=lambda d: float(d[statistic]), reverse=True) except ValueError: sortedlist = sorted(infile, key=lambda d: d[statistic], reverse=True)