Fix code to extracts multiple values from a string, then sends all them to a function and replaces t-CodePudding

import re

input_text_to_check = "En ese sitio [(2022-09-04 15:10 pm)to(2022-09-04 04:30 am)][] vi que hay algunos campamentos y [(2022-09-06 04:00 am)to(2022-09-05 04:15 am)][fuimos a caminar hasta las 22:00 pm montanas, de alli] cerca de hasta las 23:00 pm hs"

I need to rearrange the datetime patterns if the time periods are reversed, for example :

Input: [(2022-09-04 15:10 pm)to(2022-09-04 04:30 am)]

Correct output: [(2022-09-04 04:30 am)to(2022-09-04 15:10 pm)]

Input: [(2022-09-06 04:00 am)to(2022-09-05 04:15 am)]

Correct output: [(2022-09-05 04:15 am)to(2022-09-06 04:00 am)]

And this is the function that contains the replacement logic, which must receive the substring with the values of year, month, day, hour, minute, and am or pm

def reorder_inconsistent_time_periods(input_text_to_check):
    print(repr(input_text_to_check))

    #Extract dates and hour values
    #-------------------
    date_in_numbers = r"\d{4}[\s|]*-[\s|]*\d{2}[\s|]*-[\s|]*\d{2}"
    time_in_numbers = r"(\d{2})[\s|]*(?::|)[\s|]*(\d{2})[\s|]*(?:am|pm)"
    date_and_time_re_pattern = "[("   date_in_numbers   " "   time_in_numbers   ")to("   date_in_numbers   " "   time_in_numbers   ")]"
    time_groups = []
    m0 = re.search(date_and_time_re_pattern, input_text_to_check, re.IGNORECASE)
    if m0:
        time_groups = m0.groups()
        print(time_groups)
    #-------------------

    #This time 1 has to be less than time 2
    year_1, month_1, day_1, hour_1, minute_1, am_or_pm_1 = time_groups[0], time_groups[1], time_groups[2], time_groups[3], time_groups[4], time_groups[5]
    year_2, month_2, day_2, hour_2, minute_2, am_or_pm_2 = time_groups[6], time_groups[7], time_groups[8], time_groups[9], time_groups[10], time_groups[11]
    same_year, same_month, same_day, same_hour, same_minute = False, False, False, False, False
    reorder_time_period = ""

    if(int(year_2) < int(year_1)):
        #reorder times
        reorder_time_period = "[("   year_2   "-"   month_2   "-"   day_2   " "   hour_2   ":"   minute_2   " "   am_or_pm_2   ")to("   year_1   "-"   month_1   "-"   day_1   " "   hour_1   ":"   minute_1   " "   am_or_pm_1   ")]"
    else:
        if(int(year_2) == int(year_1)): same_year = True

        if(int(month_2) < int(month_1)):
            #reorder times
            reorder_time_period = "[("   year_2   "-"   month_2   "-"   day_2   " "   hour_2   ":"   minute_2   " "   am_or_pm_2   ")to("   year_1   "-"   month_1   "-"   day_1   " "   hour_1   ":"   minute_1   " "   am_or_pm_1   ")]"
        else:
            if(int(month_2) == int(month_1)): same_month = True

            if(int(day_2) < int(day_1)):
                #reorder times
                reorder_time_period = "[("   year_2   "-"   month_2   "-"   day_2   " "   hour_2   ":"   minute_2   " "   am_or_pm_2   ")to("   year_1   "-"   month_1   "-"   day_1   " "   hour_1   ":"   minute_1   " "   am_or_pm_1   ")]"
            else:
                if(int(day_2) == int(day_1)): same_day = True

                if(int(hour_2) < int(hour_1)):
                    #reorder times
                    reorder_time_period = "[("   year_2   "-"   month_2   "-"   day_2   " "   hour_2   ":"   minute_2   " "   am_or_pm_2   ")to("   year_1   "-"   month_1   "-"   day_1   " "   hour_1   ":"   minute_1   " "   am_or_pm_1   ")]"
                else:
                    if(int(day_2) == int(day_1)): same_hour = True

                    if(int(minute_2) < int(minute_1)):
                        #reorder times
                        reorder_time_period = "[("   year_2   "-"   month_2   "-"   day_2   " "   hour_2   ":"   minute_2   " "   am_or_pm_2   ")to("   year_1   "-"   month_1   "-"   day_1   " "   hour_1   ":"   minute_1   " "   am_or_pm_1   ")]"
                    else:
                        if(int(minute_2) == int(minute_1)): same_minute = True

                        if ( (same_year == True) and (same_month == True) and (same_day == True) and (same_hour == True) and (same_minute == True) ):
                            #It is not necessary to establish a period, you could only put a single date
                            reorder_time_period = "[("   year_1   "-"   month_1   "-"   day_1   " "   hour_1   ":"   minute_1   " "   am_or_pm_1   ")]"
                        else:
                            #There is no need to reorder anything because the time period in the input string is well established
                            reorder_time_period = input_text_to_check

    return reorder_time_period

Here I establish the regex that conditions the call to the function that will be in charge of reordering the data

date_in_numbers = r"\d{4}[\s|]*-[\s|]*\d{2}[\s|]*-[\s|]*\d{2}"
time_in_numbers = r"(\d{2})[\s|]*(?::|)[\s|]*(\d{2})[\s|]*(?:am|pm)"
date_and_time_re_pattern = "[("   date_in_numbers   " "   time_in_numbers   ")to("   date_in_numbers   " "   time_in_numbers   ")]"

input_text_to_check = re.sub(date_and_time_re_pattern, reorder_inconsistent_time_periods, input_text_to_check)

print(repr(input_text_to_check))

This should print the original string but with the time periods reordered :

"En ese sitio [(2022-09-04 04:30 am)to(2022-09-04 15:10 pm)][] vi que hay algunos campamentos y [(2022-09-05 04:15 am)to(2022-09-06 04:00 am)][fuimos a caminar hasta las 22:00 pm montanas, de alli] cerca de hasta las 23:00 pm hs"

the problem is that for some reason this error appears in the regex pattern line, although I don't understand why this is appear.

Traceback (most recent call last):
    m0 = re.search(date_and_time_re_pattern, input_text_to_check, re.IGNORECASE)
    raise source.error("unbalanced parenthesis")
re.error: unbalanced parenthesis at position 91

I hope you can help me, maybe even find a more convenient way to do this.

CodePudding user response：

Here is a solution using re.sub with a callback function. We can match the datetime and am/pm components in 4 separate capture groups. Then, parse each datetime string into a bona fide datetime. Finally, compare the two datetimes to determine the correct order. Note that your times are already in 24 hour format, so the am/pm values are superfluous and we don't need to parse them.

def repl(m):
    str1 = m.group(1)
    dt1 = datetime.strptime(m.group(1), '%Y-%m-%d %H:%M')
    p1 = m.group(2)
    str2 = m.group(3)
    dt2 = datetime.strptime(m.group(3), '%Y-%m-%d %H:%M')
    p2 = m.group(4)
    if dt1 < dt2:
        output = f"[({str1} {p1})to({str2} {p2})]"
    else:
        output = f"[({str2} {p2})to({str1} {p1})]"
    return output

input_text_to_check = "En ese sitio [(2022-09-04 15:10 pm)to(2022-09-04 04:30 am)][] vi que hay algunos campamentos y [(2022-09-06 04:00 am)to(2022-09-05 04:15 am)][fuimos a caminar hasta las 22:00 pm montanas, de alli] cerca de hasta las 23:00 pm hs"
output = re.sub(r'\[\((\d{4}-\d{2}-\d{2} \d{2}:\d{2}) (am|pm)\)to\((\d{4}-\d{2}-\d{2} \d{2}:\d{2}) (am|pm)\)\]', repl, input_text_to_check)
print(output)

This prints:

En ese sitio [(2022-09-04 04:30 am)to(2022-09-04 15:10 pm)][] vi que hay algunos campamentos y [(2022-09-05 04:15 am)to(2022-09-06 04:00 am)][fuimos a caminar hasta las 22:00 pm montanas, de alli] cerca de hasta las 23:00 pm hs