This is incorrect. If you add the 's' switch to the end of the line then you can search across newlines, and ^,$ will match only at the very beginning and end of the string.

For example:

$contents = '<table>
    <record>
    <fld>25015</fld>
    </record>
    </table>';
$contents =~ /(<record>.+<\/record>)(.*)(\1)/si;
print 'Result: $1, $2, $3';

You can also include \n as a possible match in the regex.

Example (from HTML):
<ul>(\n|.)*</ul>

Result: will match all the text between the <ul> tags, even if it spans multiple lines.

Hopefully, you can expand this to other applications.