lesson 101 Dates Numbers & Regex · intermediate

Read and write dates and times

A timestamp in a file is just text. `datetime.strptime` reads it with a format you spell out, giving a value you can ask for its weekday or compare; `strftime` writes it back in any layout.

You'll be able to

Parse timestamps with datetime.strptime(text, format) using %Y %m %d %H %M %S
Read parts of a datetime (year, month, weekday()) and format it with strftime
Recognise the ValueError for text that doesn't match, and the silent %m/%M mix-up

The idea

"2024-01-03 18:35:10" looks like a time, but to Python it's ten digits and some punctuation: you can't ask it for the weekday, compare it properly with "2024-1-4 9:00", or add a day to it.

from datetime import datetime, then datetime.strptime(text, format) (string parse time) reads the text using a format made of codes: %Y four-digit year, %m month number, %d day, %H hour (00–23), %M minute, %S second, %b a short month name like Jan. Everything else in the format must appear in the text exactly. The result is a datetime, and value.strftime(format) (string format time) writes it back out in whatever layout a report needs.

Run each example to see what it prints. Running examples doesn't count toward progress.

example-1.py · An order timestamp, taken apart

from datetime import datetime

ordered = datetime.strptime("2024-01-03 18:35:10", "%Y-%m-%d %H:%M:%S")
print(ordered)
print(ordered.year, ordered.month, ordered.day, ordered.hour)
print(ordered.strftime("%d %b %Y, %H:%M"))
print(ordered.strftime("%A"), ordered.weekday())

example-2.py · Text that doesn't match the format

from datetime import datetime

due = datetime.strptime("01/03/2024", "%Y-%m-%d")

Once timestamps are datetimes, questions about them become one line each. This function counts orders per weekday, the kind of figure that decides when a shop schedules staff.

example-3.py · Orders per weekday

from datetime import datetime

def orders_per_weekday(timestamps):
    counts = {}
    for text in timestamps:
        day = datetime.strptime(text, "%Y-%m-%d %H:%M:%S").strftime("%A")
        counts[day] = counts.get(day, 0) + 1
    return counts

print(orders_per_weekday([
    "2024-01-03 18:35:10",
    "2024-03-13 17:16:09",
    "2024-08-23 04:08:40",
    "2024-12-03 23:16:00",
    "2024-12-21 19:58:19",
]))

When to use it: whenever dates arrive as text (CSV columns, logs, forms) and you need to compare them, group them or show them differently. ISO text like 2024-01-03 18:35:10 can also be read with datetime.fromisoformat(text). Mistakes to avoid: %M (minutes) where you meant %m (month), which parses without any error and gives the wrong date; and guessing whether 01/03 is day-first or month-first instead of knowing where the data came from.

Review

datetime.strptime(text, format) parses; value.strftime(format) formats. The format must match the text exactly.
%Y %m %d %H %M %S are year, month, day, hour, minute, second; %m and %M are easy to swap.
weekday() is 0 for Monday through 6 for Sunday.
Next: a time only means something with its time zone.

Practise this

Convert US-style invoice timestamps