User:Graham87/Logs.py

This page contains a quick script I wrote to enumerate the number of entries for each day in the old log pages; see this entry in my personal timeline for November 2016 for more details.

#!python3
import datetime
from dateutil import parser #Parse the dates
import fileinput #To process a list of files to input
import sys
dates={}
files=[]
fin=open('protlist.txt') #A file containing a list of log files in plain text format, one file per line; I happened to be working on the protection log when I last modified this file
for line in fin:
    files.append(line.strip())
fin.close()
for line in fileinput.input(files):
    if '[[User:' in line and '200' in line:
        try:
            date=str(parser.parse(line[line.find(':')-2:line.find('[')]).date())
        except ValueError:
            print(fileinput.filename(), fileinput.fileno())
            print(line)
            sys.exit()
        dates[date]=dates.get(date,0)+1#Count the number of instances of each date
fout=open('protdates.txt','w') #The file containing the dates
date_list=list(dates)
date_list.sort()
for item in date_list:
    fout.write(item+chr(9)+str(dates[item])+'\n')