PHP line break where lower-case letters meets uppercase-letter

Question

I have a string, that is like the following:

$string = "New video gameSome TV showAnother item";

I want to be able to access each item individually, to get something like the following output:

New video game
Some TV show
Another item

How can I add a \n after each item name in the string or some other random character, that I can later explode into an array to access each item individually in the string?

you better have your string items delimited by a comma or sth! — M Reza Saberi, Mar 18 '14 at 05:05
Who is to tell that `TV` is not uppercase first letter and shouldn't split on that? — Hanky Panky, Mar 18 '14 at 05:06
yeah, this is how the string is given to me… i'm trying to delimit the items within it — user3430837, Mar 18 '14 at 05:06
Looks challenging but there is no logic behind this... Yeah it fails on `TV` case. — Shankar Narayana Damodaran, Mar 18 '14 at 05:08
You need to have a pattern to know where you should explode. If uppercase letter is the delimiter, all other letters should be in lowercase. — AyB, Mar 18 '14 at 05:09
can you do something with the ASCII values do you think? and check for it that way? @ShankarDamodaran — user3430837, Mar 18 '14 at 05:09
but all other letters are not lowercase and that's the problem. can you somehow look at the ascii value of the previous letter and compare that to the ascii value of the next?? but the runtime of that for a big string would be terrible… @ICanHasCheezburger — user3430837, Mar 18 '14 at 05:13

Dave Meybohm · Accepted Answer · 2014-03-18T05:30:43.603

4

$string = preg_replace('/([a-z])([A-Z])/', "\\1\n\\2", $string);

To answer your comment to include words that end in a closing parenthesis or a number:

$string = preg_replace('/([a-z0-9\)])([A-Z])/', "\\1\n\\2", $string);

edited Mar 18 '14 at 05:30

answered Mar 18 '14 at 05:15

Dave Meybohm

651
8
5

This is excellent!!! Can you also account for words that may end in numbers and a closing )? @Dave Meybohm – user3430837 Mar 18 '14 at 05:22

score 0 · Answer 2 · answered Mar 18 '14 at 05:14

We (humans) can differentiate the items and identify one from another based on what it means to us. but a computer must be given a criteria to do all that.

You need to change the way you store the string by simply adding a delimiter (comma or colon) and do not expect the computer to read our minds on this.

score 0 · Answer 3 · answered Mar 18 '14 at 05:16

0

what about this ,

$string = "New video game/nSome TV show/nAnother item";

$string = explode("/n", $string);

print_r( $string);

answered Mar 18 '14 at 05:16

Dimag Kharab

4,439
1
24
45

score 0 · Answer 4 · answered Mar 18 '14 at 05:31

This is not a good way to try to separate items. as has been said multiple time, save yourself the headache and use delimiters, That being said, this returns desied results from given string.

$my_string = "New video gameSome TV showAnother item";
preg_match_all('/[A-Z]{1}([A-Z]{2,5}|[a-z\s])+/',$my_string, $matches);
var_dump($matches);

But i'm sure you'll find more cases this doesn't work if you keep using patterns that don't make sense.

[A-Z]{1} - find one uppercase letter
()+ - next pattern one or more times
[A-Z]{2,5}|a-z\s - 2 -5  uppercase letters(for acronyms) OR lowercase letters and spaces

This does what you ask here. Good luck not breaking it.

var dump looks like - never mind the second part.

array(2) { [0]=> array(3) { [0]=> string(14) "New video game" [1]=> string(12) "Some TV show" [2]=> string(12) "Another item" } [1]=> array(3) { [0]=> string(1) "e" [1]=> string(1) "w" [2]=> string(1) "m" } }

score 0 · Answer 5 · answered Mar 18 '14 at 05:37

I do a bit comparison to efficiently look for uppercases, and I only consider uppercases past the first character that aren't preceded by another uppercase or a space.

<?php

$s = "New video gameSome TV showAnother item";
$i = 0;
$j = 0;
$phrases = array();
$cap_bit = pow(2, 5);
while($j < strlen($s))
{
    $n = ord($s{$j});

    if(($n & $cap_bit) == 0 && 
       ($j == 0 || (
            ord($s{$j - 1}) & $cap_bit) > 0 && 
            $s{$j - 1} != ' ') && 
       $j > 0)
    {
        $phrases[] = substr($s, $i, $j - $i);
        $i = $j;
    }

    $j++;
}


$phrases[] = substr($s, $i);
var_dump($phrases);

Result:

array(3) {
  [0]=>
  string(14) "New video game"
  [1]=>
  string(12) "Some TV show"
  [2]=>
  string(12) "Another item"
}

PHP line break where lower-case letters meets uppercase-letter

5 Answers5