[SOLVED] Regex Problem

[jjacquay712]

this regex should return test.html, but its getting this error: Warning: preg_match_all() [function.preg-match-all]: Delimiter must not be alphanumeric or backslash in lol.php on line 6

<?php
$code = "<a href=\"test.html\">test</a>";
preg_match_all("\/href=\"(.*)\"/", $code, $array);
echo '<pre>';
print_r($array);
echo '</pre>';
?>

[jjacquay712]

bump

[Psycho]

$code = "<a href=\"test.html\">test</a>";
preg_match_all("/href=\"(.*)\"/i", $code, $array);
echo '<pre>';
print_r($array);
echo '</pre>';

 

EDIT: DO not "bump" after only 10 minutes.

[MadTechie]

mjdamato: That will work find for a the example give but will fail for this

<a href="test.html" class="myclass">test</a>

 

use lazyiness

'/href="(.*?)"/i'

ie

preg_match_all("/href=\"(.*?)\"/i", $code, $array);

 

and on another note jjacquay712 read the rules.. wait an hour before bumping, if you don't get help after 2 bumps you may want to re-think the question posted,

 

many members scan down the page looking for people with no replies in an effort to help, by bumping yourself too soon mean some members (myself included) would probably miss you and thus costing you more time.

 

in anycase i hope it helps

[.josh]

note that unless you are retrieving your string from some text file or something where the escape quotes will physically be there, you need to remove the escape quotes from your regex.  For example, scraping a rendered webpage will not have escaped quotes in it. 

 

[MadTechie]

That would be true if we used single quotes CV..

 

with double quote we still need to escape the quotes

ie

<?php
$code= '<a href="test.html" class="myclass">test</a>';
preg_match("/href=\"(.*?)\"/i", $code, $array);
echo $array[1];
?>

 

But to save confusion it is always better to contain the RegEx in singles

<?php
$code= '<a href="test.html" class="myclass">test</a>';
if (preg_match('/href="(.*?)"/i', $code, $regs))
{
$result = $regs[1];
}else{
    $result = "Not found";
}
echo $result;
?>

[.josh]

What I mean is, if you're going to for example scrape a webpage using file_get_contents, you wouldn't be using quotes to assign the data to a var at all, so quoting is not an issue.  But in the regex above, you are taking into consideration escaped quotes, which would be non-existent in a scraped page from file_get_contents, since the data would already be rendered.

 

Point is, when you are trying to create a regex pattern using example data in a string, you're going to have to do some quote juggling, be it escaping, or using single/double, or whatever.  But that won't really necessarily be the case/issue for where your data is coming from in production environment.

 

Look at the OP's example string:

 

$code = "<a href=\"test.html\">test</a>";

 

It would not look like that if that link was coming from a page scrape.  It would not be escaped like that.  But he made his regex to look for escaped quotes. 

[effigy]

use lazyiness

'/href="(.*?)"/i'

 

Even better, be specific: '/href="([^"]*)"/i'

 

in the regex above, you are taking into consideration escaped quotes

 

The double quotes were escaped in the regex because they were used to delimit the string; they will match an unescaped quote:

 

<pre>
<?php
  $str = '"';
  echo preg_match("/\"/", $str);
?>
</pre>