Apply PHP regex replace on a multi-line repeated pattern

Question

Let's say I have this input:

I can haz a listz0rs!
# 42
# 126
I can haz another list plox?
# Hello, world!
# Welcome!

I want to split it so that each set of hash-started lines becomes a list:

I can haz a listz0rs!
<ul>
    <li>42</li>
    <li>126</li>
</ul>
I can haz another list plox?
<ul>
    <li>Hello, world!</li>
    <li>Welcome!</li>
</ul>

If I run the input against the regex "/(?:(?:(?<=^# )(.*)$)+)/m", I get the following result:

Array
(
    [0] => Array
    (
        [0] => 42
    )
    [1] => Array
    (
        [0] => 126
    )
    [2] => Array
    (
        [0] => Hello, world!
    )
    [3] => Array
    (
        [0] => Welcome!
    )
)

This is fine and dandy, but it doesn't distinguish between the two different lists. I need a way to either make the quantifier return a concatenated string of all the occurrences, or, ideally, an array of all the occurrences.

Ideally, this should be my output:

Array
(
    [0] => Array
    (
        [0] => 42
        [1] => 126
    )
    [1] => Array
    (
        [0] => Hello, world!
        [1] => Welcome!
    )
)

Is there any way of achieving this, and if not, is there a close alternative?

Jan Goyvaerts · Accepted Answer · 2010-05-02 03:12:08Z

1

If you want to do this with regular expressions, you'll need two. Use the regex ^(#.*\r?\n)+ to match each list and add tags around it. Within each list (as matched by the first regex), search-and-replace ^#.* with <li>$0</li> to add tags around each list item. Both regexes require ^ to match at line breaks (/m flag in PHP).

In PHP you can use preg_replace_callback and preg_replace to achieve this in just a few lines of code.

$result = preg_replace_callback('/^(#.*\r?\n)+/m', 'replacelist', $subject);

function replacelist($groups) {
  return "<ul>\n" .
    preg_replace('/^#.*/m', '    <li>$0</li>', $groups[0])
    . "</ul>\n";
}

answered May 2, 2010 at 3:12

Jan Goyvaerts

22.1k7 gold badges63 silver badges72 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Amber · Accepted Answer · 2010-05-01 04:52:18Z

0

I'd say don't try to do it all in a single regex - instead, first use a regex to match sets of consecutive lines that all begin with # signs and wrap those lines with a <ul></ul> pair. Then use a second regex (or not even a regex at all - you could just split on line breaks) to match each individual line and convert it to <li></li> format.

answered May 1, 2010 at 4:52

Amber

531k89 gold badges643 silver badges558 bronze badges

2 Comments

Hussain Over a year ago

I thought of doing that as well, but the problem is that you can't quantify the lines, so when you wrap them in ul tags, you'll be wrapping each line instead of the entire set.

Amber Over a year ago

With a multiline regex, you could match multiple lines at once. You'd just need to match the whitespace newline characters between them too.

Syntax Error · Accepted Answer · 2010-05-01 04:54:29Z

0

If it was me I would:

explode("\n", $input) into an array where 1 key = line
foreach through that array
whenever you get a line that doesn't start with a #, that's when you add your closing/opening ul tags

Add a little more to deal with unexpected input (like two non hash lines in a row) and you're good.

answered May 1, 2010 at 4:54

Syntax Error

4,5473 gold badges24 silver badges33 bronze badges

Comments

Community · Accepted Answer · 2017-05-23 10:32:56Z

0

Looks like Syntax Error has already explained what I'm doing. But here goes the link to a working example.

edited May 23, 2017 at 10:32

CommunityBot

11 silver badge

answered May 1, 2010 at 5:07

codaddict

457k83 gold badges502 silver badges537 bronze badges

Comments

artlung · Accepted Answer · 2010-05-01 05:11:18Z

0

With structured content like this, I would not do this as a regex. How about another approach?

$your_text = <<<END
I can haz a listz0rs!
# 42
# 126
I can haz another list plox?
# Hello, world!
# Welcome!
END;

function printUnorderedList($temp) {
    if (count($temp)>0) {
        print "<ul>\n\t<li>" .implode("</li>\n\t<li>", $temp) . "</li>\n</ul>\n";
    }
}

$lines = explode("\n", $your_text);
$temp = array();
foreach($lines as $line) {
    if (substr($line, 0, 1) == '#') {
        $temp[] = trim(substr($line,1));
    } else {
        printUnorderedList($temp);
        $temp = array();
        echo $line . "\n";
    }
}
printUnorderedList($temp);

answered May 1, 2010 at 5:11

artlung

33.5k19 gold badges76 silver badges124 bronze badges

Comments

Henrik · Accepted Answer · 2010-05-01 05:12:00Z

You could avoid regex altogether, and simply try a simpler approach by having it read the file, line by line (an array of lines), and every time it encounters a non-hash-started line, it starts a new list. Like so:

// You can get this by using file('filename') or 
// just doing an explode("\n", $input)
$lines = array(
    'I can haz a listz0rs!',
    '# 42',
    '# 126',
    'I can haz another list plox?',
    '# Hello, world!',
    '# Welcome!'
);

$hashline = false;
$lists = array();
$curlist = array();
foreach ($lines as $line) {
    if ($line[0] == '#')
        $curlist[] = $line;
    elseif ($hashline) {
        $lists[] = $curlist;
        $curlist = array();
        $hashline = false;
    }
}

A little clean-up may be in order, but hopefully it helps.

(after reading new answers, this is basically an indepth explanation of Syntax Error's answer.)

EDIT: You may want it to strip off the # at the beginning of each line too.

Collectives™ on Stack Overflow

Apply PHP regex replace on a multi-line repeated pattern

6 Answers 6

Comments

2 Comments

Comments

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

6 Answers 6

Comments

2 Comments

Comments

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related