12-hour to 24-hour clock conversion

Alan Kilborn

It wasn’t for you, it was for future searchers of this site that may find this topic when what they are really looking for is the other way around. :)

BTW if Eko doesn’t solve your problem here in a reasonable amount of time, I’ll jump in. :)

Ekopalypse

@Alan-Kilborn

Alan, LOL, please feel free to do so - you are much more in this regex then I’m.

Alan Kilborn

@Sophia-Cruz

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