Use RegEx to get (parsed) times from string - php

Question

I need some direction on how to alter my current RegEx line.
Example string:
9:00 AM - 9:30 AM
Desired output:
900 930 (each can be an index in an array or whatever)..
Current approach:
preg_match('/^([0-9]{1,2}):([0-9]{1,2}) ?([ap]m)/i', trim($presentationtime), $matches);
however this only seems to get me the FIRST (stripped/parsed) time.
results:
$presentationtime = $matches[1] . $matches[2];
echo 'Matches check: '. $matches[0] . '<br>';
echo 'Matches check: '. $matches[1] . '<br>';
returns:
Matches check: 9
Matches check: 00
How can I alter my regex to get BOTH times (stripped/parsed the same way)..
I'm expecting a 6 index array..but can only get a 3 index/count array

Answer 1

as #anubhava says use preg_match_all
$presentationtime = '9:00 AM - 9:30 AM';
preg_match_all('/([0-9]{1,2}):([0-9]{1,2}) /', trim($presentationtime), $matches);
print_r($matches);
results:
Array
(
[0] => Array
(
[0] => 9:00
[1] => 9:30
)
[1] => Array
(
[0] => 9
[1] => 9
)
[2] => Array
(
[0] => 00
[1] => 30
)
)
Edit to answer the comment:
very lazy workaround to get one dimensional array, banal regex
$presentationtime = '9:00 AM - 9:30 PM';
preg_match('/([0-9]{1,2}):([0-9]{1,2})\s([apm]+)\s-\s([0-9]{1,2}):([0-9]{1,2})\s([apm]+)/i', trim($presentationtime), $matches);
result
Array
(
[0] => 9:00 AM - 9:30 PM
[1] => 9
[2] => 00
[3] => AM
[4] => 9
[5] => 30
[6] => PM
)

Answer 2

You have to simply use preg_match_all (as stated by anubhava in the comment) and remove the ^ at the start of the regex:
preg_match_all('/([0-9]{1,2}):([0-9]{1,2}) ?([ap]m)/i', trim($presentationtime), $matches);
Then $matches will be like:
Array
(
[0] => Array
(
[0] => 9:00 AM
[1] => 9:30 AM
)
[1] => Array
(
[0] => 9
[1] => 9
)
[2] => Array
(
[0] => 00
[1] => 30
)
[3] => Array
(
[0] => AM
[1] => AM
)
)
If you want to refine the regex you may use:
preg_match('/([0-9]{1,2}):([0-5]\d)\s*([ap]m)/i', trim($presentationtime), $matches);
The minute section is will match 2 digits from 00 to 59, the space section \s* is optional and matches more than one whitespace char (space, tab, CR, FF...)

Answer 3

One way of doing that:
$data = '9:00 AM - 9:30 AM';
print str_replace('-', ' ', preg_replace("/[^0-9-]/","", $data));
Result:
900 930
keep digits and dash
replace dash with space
In the previous example, I was using limited regex with preg_replace. To keep things simple and using time with AM/PM, I'd just explode (as you already thought about),
$data = '9:00 AM - 9:30 PM';
$timeArray = array_map('trim', explode('-', $data)); // trim each array item
$timeArray = array_map('Time24Hr', $timeArray); // clean up time
print_r($timeArray);
function Time24Hr($time) {
$time_ampm = explode(' ', $time);
if (strtolower(trim($time_ampm[1])) === 'am') {
// A
return $time_ampm[0];
// B
//return str_replace(':', '', $time_ampm[0]);
// C
//return $time;
}
$hr_min = explode(':', trim($time_ampm[0]));
// A
return ($hr_min[0] + 12) . ':' . $hr_min[1];
// B
//return ($hr_min[0] + 12) . $hr_min[1];
// C
//return ($hr_min[0] + 12) . ':' . $hr_min[1]. ' ' . $time_ampm[1];
}
Results:
Result A:
Array
(
[0] => 9:00
[1] => 21:30
)
Result B:
Array
(
[0] => 900
[1] => 2130
)
Result C: -- obviously this is insane, but you can easily make it sane. This is just an example
Array
(
[0] => 9:00 AM
[1] => 21:30 PM
)
Of course, this simplistic method can be streamlined much further. This method doesn't need regex.