Question

I am writing a python library that takes a paragrah that consists of different types of hours format. Paragraph is actually taken from different restaurant sites, they have hours of operation in various formats for each site. So, I decided to write a library that takes any format of (eg: M-Fri 10am – 12pm, M, Tue, Wed 00:00-18:00 etc.). My decision is to write different set of RegEx that takes a string and proceeds if it matches any of my pattern.

I have choosen RegEx pattern matching because, instead of writing own logics for every pattern and replacing just using str.replace may takes a lot of time. But I stuck in regex since I am new to it. So any of you can suggest me how can I do the following task?

As a first step I would like to shorten all the weekday formats to 3 letter shorten week days as below.
Example:
“Monday 09:00 AM – 5:00 PM, Tuesday 09:00 AM – 5:00 PM, Wednesday 09:00 AM – 5:00 PM, Thursday 09:00 AM – 5:00 PM, Friday 09:00 AM – 5:00 PM, Saturday Closed, Sunday Closed“

Output should be
“Mon 09:00 AM – 5:00 PM, Tue 09:00 AM – 5:00 PM, Wed 09:00 AM – 5:00 PM, Thu 09:00 AM – 5:00 PM, Fri 09:00 AM – 5:00 PM, Sat Closed, Sun Closed”

Thanks.

Answer 1

seems like a simple str.replace should work here:

new_paragraph = paragraph.replace('Monday','Mon')

You can even use the calendar module to make this a little easier:

import calendar

for day,abbr in zip(calendar.day_name,calendar.day_abbr):
    paragraph = paragraph.replace(day,abbr)

---

for really big paragraphs, it might be worthwhile to switch over to regex:

import re
import calendar

regex = re.compile('|'.join(calendar.day_name))
sub_dict = dict(zip(calendar.day_name,calendar.day_abbr))
new_paragraph = regex.sub(lambda match: sub_dict[match.group(0)],paragraph)

Of course, with calendar, the abbreviations/etc are locale dependent (but maybe that’s desirable anyway)