Change one part of the selection with Regex

Question

I would like to insert a class with regex and preg_replace

echo preg_replace("/<li\>\s*<p\>[a-z]\)\s/", "/<li class=\"inciso\"\>\s*<p\>[a-z]\)\s/", $documento);

This is the model of text that I haveEste é o modelo das linhas do meu documento:

<li>
  <p>a) long text</p>
</li>
<li>
  <p>b) long text</p>
</li>
<li>
  <p>c) long text</p>
</li>

New example, let´s say that is not a HTML, is just a simple list, and you wanna a change from this:

a) long text
b) long text
c) long text

To this:

a) new text long text
b) new text long text
c) new text long text

echo preg_replace("/[a-z]\)\s/", "/[a-z]\)\snew\stext/", $documento);

Is this correct?

[Don't parse HTML with regex!](http://stackoverflow.com/a/1732454/418066) — Biffen, May 05 '15 at 13:39
It depends. If you don't have to **parse** it, it's fine and maybe easier. It depends on the task and on the document you have. — ColOfAbRiX, May 05 '15 at 13:44
Yes, you can do it. Few questions about the part you added: you are replacing "every word followed by a space. Is that what you want? Or do you need to replace just the word "long"? — ColOfAbRiX, May 05 '15 at 13:52
Yes, it has a space, and the word "long", is just a sample, the content changes from this part. The only part that stay stable is the beginning to the space after the ")" — Alê Moraes, May 05 '15 at 13:55

ColOfAbRiX · Answer 1 · 2015-05-05T13:54:37.307

IF, and I emphasize again IF, the input text you have is like the one you posted here, then you can assume you can find a safe pattern to replce, as you won't see this pattern somewhere else:

preg_replace("/<li>/", "<li class=\"inciso\"\/>", $documento);

This will replace every occurrence of <li> with the modified version. If there are <li> that you won't replace then it becomes more difficult and you should use a DOM or SAX parser

UPDATE after your update: You can match a word and add something before it with:

preg_replace("(long)", "new text $1", $documento);

Have a look at backreferences

Lisa · Accepted Answer · 2015-05-05T13:51:28.193

0

use str_replace instead.

$find    = '<li>';
$replace = '<li class="inciso">';


echo str_replace($documento, $find, $replace);

edited May 05 '15 at 13:51

answered May 05 '15 at 13:46

Lisa

54
1
5

But I have others instance of
, that don´t matches what I want.

Alê Moraes

May 05 '15 at 13:48

than describe better what you want. (the exact strings). In general you should not use regexp for that. – Lisa May 05 '15 at 13:49

Sorry Lisa. ;) . But if I can´t use regex, what are my option? – Alê Moraes May 05 '15 at 13:57

If you have a way to uniquely identify in one go all the `

` you need (like if they are always followed by a character, or if they have specific attributes and so on...) to replace then you can use regexes, otherwise no

– ColOfAbRiX May 05 '15 at 13:57

Change one part of the selection with Regex

2 Answers2