Python creating date objects to filter certain dates

Question

I am reading different log files with different date formats. I am using python to read in the files line by line and then parse the line. I want to parse the line for dates and then formulate them into a date object to do comparisons on.

For example, say that I have 2 log files both with different date formats. How do I read them into an object to compare them to a known date. Assume for example, I wanted to discard all dates before a certain time.

Let's assume the first log file just has one line:

invalid access 2015-01-04 14:23:15 on IP 5.5.5.5

How do I read in 2015-01-04 14:23:15 into a dateobject (so I can do comparisons)

What if the date format was different? How would I read in that?

if the time string represents local time then you need to convert it to UTC or POSIX time to do comparisons because local time is non-monotonous. See [Find if 24 hrs have passed between datetimes - Python](http://stackoverflow.com/a/26313848/4279) — jfs, Feb 10 '15 at 23:29
related: [How to parse ISO formatted date in python?](http://stackoverflow.com/q/127803/4279) — jfs, Feb 10 '15 at 23:36

score 1 · Answer 1 · answered Feb 09 '15 at 22:07

You can use datetime.datetime.strptime:

In [1]: from datetime import datetime
In [2]: d = '2015-01-04 14:23:15'
In [3]: datetime.strptime(d, '%Y-%m-%d %H:%M:%S')
Out[3]: datetime.datetime(2015, 1, 4, 14, 23, 15)

For other formats, check out the documentation

score 0 · Answer 2 · edited Feb 09 '15 at 22:08

0

dateutil can usually parse whatever

import dateutil.parser as p
print p.parse("2015-01-04 14:23:15")

this assumes you can isolate your datestring

$ easy_install python-dateutil 
$ pip install python-dateutil

or simply attainable at https://pypi.python.org/pypi/python-dateutil/2.4.0 if you need the source ...

edited Feb 09 '15 at 22:08

mgilson

300,191
65
633
696

answered Feb 09 '15 at 21:59

Joran Beasley

110,522
12
160
179

A link to [`dateutil`](https://pypi.python.org/pypi/python-dateutil/2.4.0) might be helpful as it isn't in the stdlib. – mgilson Feb 09 '15 at 22:01

score 0 · Answer 3 · answered Feb 10 '15 at 23:34

Using re module directly might be more efficient than using it implicitly via datetime.strptime() (measure it to find out whether it matters in your case):

>>> import datetime, re   
>>> s = '2015-01-04 14:23:15'
>>> datetime.datetime(*map(int, re.findall('\d+', s)))
datetime.datetime(2015, 1, 4, 14, 23, 15)

Python creating date objects to filter certain dates

3 Answers3