time_range_regex.py

"""
A pretty nice time parsing regex.
@author Richard Caceres, @rchrd2
"""

import re

# The history for this regex is stored in my gist 
# https://gist.github.com/rchrd2/9773922
time_range_regex = re.compile(r'''
    \b # make sure it's either at start of string or word boundary
    (
        # Time Ranges get higher precedence which has slightly different logic (ie 6-7pm)
        (?:
            (?:[0-1]?[0-9])            # first part of time ie: (3):00, (12):00
            (?::[0-5][0-9])?           # second part including the colon ie 12(:30)
            
            # am/pm is optional either way for the first part of a time range
            (?:                # am/pm part
                [\s]{0,10}     # limit how many spaces between time and am/pm
                (?:am|pm|a|p)
            )?
            (?:
                [\s]{0,10}
                (?:-|to)
                [\s]{0,10}

                (?:[0-1]?[0-9])    # first part of time ie: (3):00, (12):00

                \:?
                (?:[0-5][0-9])?    # second part including the colon ie 12(:30)
                (?:                # am/pm part
                    [\s]{0,10}     # limit how many spaces between time and am/pm
                    (?:am|pm|a|p)
                )
            )
        )
        |
        # a time that's not a timerange (ie 6pm but not 6-7pm)
        (?:
            (?:[0-1]?[0-9])            # first part of time ie: (3):00, (12):00
            (?:
                # if colon is there, am/pm is optional
                (?:
                    (?::[0-5][0-9])    # second part including the colon ie 12(:30)
                    (?:                # am/pm part
                        [\s]{0,10}     # limit how many spaces between time and am/pm
                        (?:am|pm|a|p)
                    )?
                )
                |
                # if colon is not there am/pm is required
                (?:[0-5][0-9])?    # second part including the colon ie 12(:30)
                (?:                # am/pm part
                    [\s]{0,10}     # limit how many spaces between time and am/pm
                    (?:am|pm|a|p)
                )
            )
        )
    )
    \b # make sure it's either at end of string or word boundary
    ''', 
    re.IGNORECASE|re.U|re.VERBOSE|re.DOTALL)

time_range_regex.findall("6-7p 6p-7p 6:00pm - 7:00pm foo bar bazz 3pm to 5pm")
time_range_regex.findall("6p 6 p 7 pm 7:00 pm 7:00 pmBar fun today at 3:00pm  300pm 400pm 3p 10a 23p (4:00pm) 11am whatthe10afoo 10afoo foo4p 5p 7-8")
time_range_regex.findall("""7-9pm 7 - 9 pm 7-9 pm 7pm-9pm 7-9:00pm 7-9p  7-9 p 6:00 PM to 11:00 PM (PDT)""")