12-hour to 24-hour clock conversion

Alan Kilborn

This seems to do it:

Find: ^(?:(01)|(02)|(03)|(04)|(05)|(06)|(07)|(08)|(09)|(10)|(11)|(12))\.(\d\d)\.(\d\d)\.(\d{3}) (?:(A)|(P))M
Repl: (?{12}?{16}00:12)(?{1}?{17}13:01)(?{2}?{17}14:02)(?{3}?{17}15:03)(?{4}?{17}16:04)(?{5}?{17}17:05)(?{6}?{17}18:06)(?{7}?{17}19:07)(?{8}?{17}20:08)(?{9}?{17}21:09)(?{10}?{17}22:10)(?{11}?{17}23:11):$13:$14:$15
Search mode: Regular expression

I tried to be “fancy” and do the “AM” and “PM” part with “named groups” so that the replacement is more “readable”, but I ran into some trouble with that. If I get it working, and it truly is more readable, I may post that as well…

Alan Kilborn

I got it to work with named groups. The problem was, and it makes it less elegant, is that mixing numbered and named capture groups seems to mess things up. So I had to artificially name group # 1 as “one” instead of just using its number. I’ll let the reader decide if this version is more readable than the previous:

Find: ^(?:(?<one>01)|(?<two>02)|(?<three>03)|(?<four>04)|(?<five>05)|(?<six>06)|(?<seven>07)|(?<eight>08)|(?<nine>09)|(?<ten>10)|(?<eleven>11)|(?<twelve>12))\.(?<mins>\d\d)\.(?<secs>\d\d)\.(?<ms>\d{3}) (?:(?<am>A)|(?<pm>P))M

Repl: (?{twelve}?{am}00:12)(?{one}?{pm}13:01)(?{two}?{pm}14:02)(?{three}?{pm}15:03)(?{four}?{pm}16:04)(?{five}?{pm}17:05)(?{six}?{pm}18:06)(?{seven}?{pm}19:07)(?{eight}?{pm}20:08)(?{nine}?{pm}21:09)(?{ten}?{pm}22:10)(?{eleven}?{pm}23:11):$+{mins}:$+{secs}:$+{ms}

Search mode: Regular expression

(And I fully expect a “holy cow batman!” from @Meta-Chuh on that one!)

(And yes, doing 12 differently from the others in the replace was intentional, as it is a little “out of place” in the conversion)

Ekopalypse

@Alan-Kilborn

LOL - if I could I would upvote it 12 times. :-D

Alan Kilborn

@Ekopalypse

Sometimes things get upvoted too quickly for people to have tried it out to verify/duplicate results. This is somewhat bad as if something doesn’t work (and the thread dies right there) it looks to future readers like a good solution…

BTW I think the “replace” is very readable with the named groups, but the “find” suffers…slightly…

Ekopalypse

@Alan-Kilborn

:-D for me this looks like what we call - bohemian villages.
But I can participate on that thread now - I’ve tested it, and it looks good to me :-)
Let’s see what the OP thinks about it.

Sophia Cruz

@Alan-Kilborn it works perfectly. Thank you so much!

Sophia Cruz

2nd solution is less confusing but both still produces the desired output. :D

Meta Chuh

@Alan-Kilborn

holy cow batman! 😄👍

Sometimes things get upvoted too quickly for people to have tried it out to verify/duplicate results.

i guess sometimes your writing style is so worth an upvote, that people give you one, even if they have tested it’s content to fail.

greetings,
robin

Alan Kilborn

@Meta-Chuh said:

your writing style…worth an upvote…

My writing style? Nah. I don’t have the flowery prose of a @PeterJones or the diplomatic panache of a @Meta-Chuh . [Basically I don’t wanna put that kind of effort in–to the typing part–obviously with the above regexes staring us in the face I’ll put a little time/effort into THAT aspect. Trying to provide some hopefully accurate help, maybe without the greatest english composition accompanying.]

guy038

Hi, @sophia-cruz, @alan-kilborn, @eko-palypse, @meta-chuh and All,

Here is a variant of the Alan’s solution, which is less restrictive :

It just needs that the hour should be located after a non-word, instead of the beginning of line, due to the \b syntax
It does not care of the synbol between the different parts of the hour ( just changing the \. syntax by . ! )
It allows the milliseconds part to be absent ( .MMM ), in the hour, with the (?:.(\d{3}))? syntax
It allows the AM or PM part to come next to the digits, with \x20?
It does not care of the case , allowing the four syntaxes am, pm, AM and PM with the regex (?i:(AM)|PM)

Note :
In the remplacement regex, when group 17 exists ( AM ), I just rewrite the hour part ( \1 ), except when hour = 12, where the logic is reversed !

So, assuming this sample text :

12.00.47 AM
	12.58.18.387 am
01 07 28 421 AM
03:17:31:934am
05.43.26.582 AM
        07:23:03.732 am
09.31.08AM
11.57.59,003 am
~~~~~~~~~~~~~~~~~~~~
12.00.45 PM
	12.56.49.203 pm
01 04 51 387 PM
03:15:33:632pm
05.22.00.814 PM
        07:33:55.548 pm
09.45.26PM
11.59.11,247 pm

With the regex S/R below :

SEARCH :    \b(0(?:(1)|(2)|(3)|(4)|(5)|(6)|(7)|(8)|(9))|(10)|(11)|(12)).(\d\d).(\d\d)(?:.(\d{3}))?\x20?(?i:(AM)|PM)
              │    │   │   │   │   │   │   │   │   │    │    │    │     │      │         │                │
              1    2   3   4    5   6   7   8   9   10   11   12    13    14     15        16               17   <-- Groups

REPLACE :   (?2(?17\1:13))(?3(?17\1:14))(?4(?17\1:15))(?5(?17\1:16))(?6(?17\1:17))(?7(?17\1:18))(?8(?17\1:19))(?9(?17\1:20))(?10(?17\1:21))(?11(?17\1:22))(?12(?17\1:23))(?13(?{17}00:\1)):$14:$15(?16\:$16)

We would obtain :

00:00:47
	00:58:18:387
01:07:28:421
03:17:31:934
05:43:26:582
        07:23:03:732
09:31:08
11:57:59:003
~~~~~~~~~~~~~~~~~~~~
12:00:45
	12:56:49:203
13:04:51:387
15:15:33:632
17:22:00:814
        19:33:55:548
21:45:26
23:59:11:247

Best Regards,

guy038

guy038

Hi, All,

When I woke up this morning, I immediately saw a simplification ( in length ! ) of the replacement regex ;-))

So, my last version is :

SEARCH \b(0(?:(1)|(2)|(3)|(4)|(5)|(6)|(7)|(8)|(9))|(10)|(11)|(12)).(\d\d).(\d\d)(?:.(\d{3}))?\x20?(?i:(AM)|PM)

REPLACE (?13(?{17}00:\1):(?17\1:(?{2}13)(?{3}14)(?{4}15)(?{5}16)(?{6}17)(?{7}18)(?{8}19)(?{9}20)(?{10}21)(?{11}22)(?{12}23))):$14:$15(?16\:$16)

BR

guy038

guy038

Hi, all,

As the (?...:....) regex structure, of the Boost regex library is not so common, you may feel a bit lost about the replacement regex syntax !

So, here is, below, the algorithmic translation of the replacement regex, of my previous post :

If group 13 exists    #  Case hour = '12'

    then

        If group 17 exists    #  String 'AM' or 'am', after the digits
  
            then

                Write the string '00'

            else

                Rewrite the group 1 contents    #  Actually, the hour '12'

    else

        If group 17 exists    #  String 'AM' or 'am', after the digits

            then

                Rewrite the group 1 contents    #  The 'hour' digits, so the values  '01'  or  '02'  or  ...... '11'

            else    #  The 'hour' digits, when followed with the string 'PM' or 'pm'

                If group  2 exists then write the string '13'    #  Case hour = '01'

                If group  3 exists then write the string '14'    #  Case hour = '02'

                If group  4 exists then write the string '15'    #  Case hour = '03'

                If group  5 exists then write the string '16'    #  Case hour = '04'

                If group  6 exists then write the string '17'    #  Case hour = '05'

                If group  7 exists then write the string '18'    #  Case hour = '06'

                If group  8 exists then write the string '19'    #  Case hour = '07'

                If group  9 exists then write the string '20'    #  Case hour = '08'

                If group 10 exists then write the string '21'    #  Case hour = '09'

                If group 11 exists then write the string '22'    #  Case hour = '10'

                If group 12 exists then write the string '23'    #  Case hour = '11'

       endif

endif

Write the string ':'

Rewrite the group 14    #  Minutes digits

Write the string ':'

Rewrite the group 15    #  Seconds digits

if group 16 exits       #  Milleseconds digits

    then

        write the string ':'

        write the group 16 contents    # Milleseconds digits
endif

BR

guy038