Hello Nikos,

Sunday, November 30, 2003, 10:29:19 AM, you wrote:

NG>  Can somebody tell me what is wrong with the following?
NG>  $body=eregi_replace("<a href=\"(.+)\"([^>]*)>(.+)</a>", "<a href=\"\\1\"
NG>  target=\"_blank\" class=\"down_txt\"><strong>\\3</strong></a>", $body);

1) /.+/ is too greedy. You should use something like /[^"]+/ or /.+?/ instead
   (second is supported only in Perl-compatible regexps). The best solution
   could be /[^'"]+?/ however.
2) /<a / doesn't match "\n" after "a".

   $body = preg_replace(
       '#<[Aa]\s[^> ]*?[Hh][Rr][Ee][Ff]\s*=\s*([\'"])([^\'">]+?)\1[^>]*?>' .
           '([^<]*?)</a[^> ]*?>#s',
       '<A href="\2" target="_blank" class="down_txt"> ' .
           '<strong> \3</strong></a>',
       $body
   );

   This regexp is not perfect, but I hope it has enough margin of safety, at
   least if your HTML code is more or less correct. E. g., all href values
   must be quoted and '<', '> ', '"' and "'" without special meaning must be
   written as HTML entities.
   
-- 
Best regards,
 Ivan                            mailto:johnny@[...].ru

-- 
PHP Database Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php