Re: PHP function inside a regex - possible?

Hi Kev, Try putting a g after the /e in your regex. Perl regular expressions stop at newlines by default - the g modifier makes the regex consider newlines to be just another character. You might also need to make the expression non-greedy to keep it from eating up everything between the first <b> and the last </b> in the text. If this is the case add a U after the g. Good Luck! Zak At 06:47 PM 6/1/00 +0000, kev wrote:

Aaarrrrgh!! Don't you always get the most problems when it's urgent?! Running a substantial block of text through this (about 2000+ characters) doesn't seem to make any change to the text at all. So, my question now, is: how can I modify the regex to modify every bolded term throughout the entire text, and is there a size limit on a string being put through a regex? If there is, how can I get round it without having to physically split it up into sections, which will be very messy as I would have to ensure that it wasn't split in the middle of <B> or </B> tag? thanks, - Kev kev wrote: This isn't giving the desired results though when there is more than one term to be modified, eg with this string: $text = "Turn <B>term 1</B> and <B>term 2</B> to a link"; It turns everything in between the first <B> and the last </B> into a link! How do I guard against that, but ensure that it still modifies every bolded term? thanks, - Kev Andreas Pour wrote:

Hi, Kevin, As to your example, you have to have the replacement string be suitable for an "echo" statement, I think. The following works, anyway: $newtext = preg_replace("/(<B>(.*)<\/B>)/e", "\"<A href=\" . '\\2' . \"><B>\" . '\\2' . \"</B></A>\"", $text ); Good luck, Andreas kev wrote:

Zak McGregor wrote:

In php 3.x, no. In php 4.x try: preg_replace("/<B>(.*)?<\/B>/", "<A HREF=\"whatever.php3?search=".urlencode('\\1')."\">".'\\1'."<\/A>");

I'm using PHP4. I don't understand how that could work, how will it know what '\\1' is in the urlencode() ? The function has not even been entered at that point, so surely urlencode('\\1') will just equate literally to '\\1' urlencode()ed. Anyway, I tried this little example as a starting point to build up to the full thing eventually: $text = "Turn <B>bold</B> to a link"; $newtext = preg_replace( "/<B>(.*)?<\/B>/e",

            "<A href=\\1><B>\\1</B></A>",
            $text );

echo "\$text = $text<BR>"; echo "\$newtext = $newtext<BR>"; Please try this example, and you'll see it doesn't work. Can you see what's wrong with it? Also, is question mark (?) needed? In Perl, I use $1 for a back-reference, is \\1 the PCRE equivalent? thanks, - Kev -- PHP General Mailing List (http://www.php.net/) To unsubscribe, e-mail: php-general-unsubscribe@lists.php.net For additional commands, e-mail: php-general-help@lists.php.net To contact the list administrators, e-mail:

php-list-admin@lists.php.net

-- PHP General Mailing List (http://www.php.net/) To unsubscribe, e-mail: php-general-unsubscribe@lists.php.net For additional commands, e-mail: php-general-help@lists.php.net To contact the list administrators, e-mail: php-list-admin@lists.php.net

-- PHP General Mailing List (http://www.php.net/) To unsubscribe, e-mail: php-general-unsubscribe@lists.php.net For additional commands, e-mail: php-general-help@lists.php.net To contact the list administrators, e-mail: php-list-admin@lists.php.net -- PHP General Mailing List (http://www.php.net/) To unsubscribe, e-mail: php-general-unsubscribe@lists.php.net For additional commands, e-mail: php-general-help@lists.php.net To contact the list administrators, e-mail: php-list-admin@lists.php.net