please interprete this ereg_replace !

kbrij

I find this ereg arguments so difficult to understand. Could someone please interpret it for me. It is very important for me to know what this replacement does from the original string.

1.
$description = ereg_replace('<[bB][rR][^>]*>', "\n",
$vars['description']);

2.
$description = addslashes(ereg_replace('</?([^{aA/]|[a-zA-Z_]{2,})[^>]*>',} '', $vars['description']));

3.
$subject = addslashes(ereg_replace('<[^>]*>', '',

$vars['subject']));

Thank you.
Brij.

epp_b

replaces any HTML line breaks ( , , , or ) with \n in the variable $vars['description']
(assuming those forward slashes should be backslashes) Will replace anything that begins with <? (php tags), with any alpha character that is 2 or more characters in length, followed by a > with nothing in the variable $vars['description'] (in other words, will delete these strings). And afterwards, it will escape any necessary characters (such as quotes) with backslashes using the addslashes() function.
Will replace any < characters, any > characters with nothing preceeding them, and any > with anything preceeding them with nothing in the variable $vars['subject']. Then, the addslashes() function will escape the necessary characters.

I may be a little rusty on my regex, so I'm not 100% sure on all of these (maybe 92.3% 😉).

See http://www.php.net/manual/en/pcre.pattern.syntax.php as well as http://www.phpbuilder.com/columns/dario19990616.php3 for futher reading on regular expressions.

kbrij

Thank you for the ereg analysis.
I will have to master the skills to understand this regex in depth.
At times they get so complicated.

Basically i was looking in the original php code, where does it stripoff all the HTML tags from description and subject.
I thought the above ereg_replace does that, but there should be another function somwhere doing the trick.

The final result of the description is a plain text with no HTML tags, that are keyed in from the user.
e.g. bold 
is replaced by

bold

Thank you.
Bij.

Weedpacket

I'd disagree with epp_b on number 2 and 3. In two, it's a difference in interpretation: if the forward slashes are supposed to be forward slashes, then the statement would delete any tag (opening or closing) that is not an <a> tag. In 3, the given replacement would delete any tags of any sort (presumably, those that are left over after replacing tags of types 1 and 2).

The first two could be better written using eregi_replace:

1.  <br[^>]*>
2.  </?([^a/]|[a-z_]{2,})[^>]*>

And the second in particular is crying out for the smarter (though more complicated) PCRE syntax:

preg_replace('#</?(?!a[>\s])[^>]*>#i', '', $vars['description']);

Which says "any tag (opening or closing) so long as it's not an "a" tag (which can be identified because it starts with "a" and is followed by either whitespace or a >).

On the other hand, (2) and (3) could just be done with strip_tags() anyway.

epp_b

Oh, silly me. Weedpacket is right, it makes a lot more sense that way.