note 8504 added to preg_match

example no. 3 seems bogus to me. Besides, trying it with php 4.02 and PWS on W95, the output produced is : domain name is: http://www.p which looks more or less ok as far as pcre is concerned (the first pranthesized expression, (.*), matches the nul string) however I can't seem to find any explanation about the /U modifier, it's not in the perlre manpage. I tried to change it to /i and /x (just to see what happens) and got the following output : domain name is: x.html which again is ok, only that now the first pranthesis is behaving normally and is greedy, eating everything it can eat while still allowing the rest of the pattern to match. anyway, it does not produce the desired output http://www.php.net/manual/function.preg-match.php