Python script to generate stats for a collection of files with the format DATE - Author 1, Author 2-Editor.doc

Stats.py

import glob
import re
from collections import Counter
import datetime
import csv

files = glob.glob('*.*')

print "There are ", len(files), " files."

editors = []
authors = []

for file in files:
    parts = file.split('.')
    ext = parts[1]
    filename = parts[0]
    info_parts = filename.split(' - ')
    if len(info_parts) < 2:
        print "Skipping ", info_parts, ""
        continue
    date = info_parts[0]
    try:
        editor = info_parts[1].split('-')[1]
        editors.append(editor)
    except IndexError:
        print "Can't find author in: ", file

    author = info_parts[1].split('-')[0]
    authors_names = re.split('[, &]', author)
    # print authors_names
    authors.extend(authors_names)

print Counter(editors)

authors = filter(None,authors)

# remove things like Author 1 and Author 2
is_integer = lambda s: not s.isdigit()
results = Counter(filter(is_integer, authors))

# todo: print the Counter in a nicer way
print results

for name, count in results.most_common():
    print "%3i | %s" % (count, name)

# export CSV file
date = datetime.datetime.today().strftime('%Y%m%d')
filename = date + "-stats.csv"
with open(filename,'w') as csvfile:
    fieldnames=['author','count']
    writer=csv.writer(csvfile)
    writer.writerow(fieldnames)
    for row in results.items():
        writer.writerow(row)

If there are 2 files:

• 180315 - Nuala, Tama-r.doc
• 180415 - Nuala, Edgeworth-r.doc

this script should count Nuala 2 times, and Tama and Edgeworth as 1 each.