--- Mike Brandonisio <[EMAIL PROTECTED]> wrote:
> That worked. Thank you. I'll play some more. I want to catch domains
> with hyphen and sub domain emails too like [EMAIL PROTECTED]
>
> Thanks again.
>
> Sincerely,
> Mike
Perhaps if I explained what this regex (regular expression) does, you could
make some intelligent modifications.
> > preg_match_all("/[EMAIL PROTECTED]/", $str, $output);
// The regex is surrounded by / characters. It could be another character
but the / is traditional. Occasionally I will use a pipe character (|)
when I have matching content which contains slashes (/) and I don't care
to use the backslash (\) to escape each one.
\w Match any letter [a-z] or [A-Z] or digit [0-9] ... [0-9A-Za-z]
\w+ The plus (+) quantifier says "1 or more" of the match to the left. In
this case one or more of any character matching [0-9A-Za-z].
\. The dot (.) normally means match any character. Since we want to match
the literal dot character, it must be escaped with the backslash. In
Perl regex it is often necessary to escape the @ symbol but not in PHP.
Effectively this pattern matches for:
{any [EMAIL PROTECTED] alphanum}.{any alphanum}
You could use something like:
[EMAIL PROTECTED]
to say
{any [EMAIL PROTECTED] char}.{any char}
Regex is greedy by default so it will try to make the longest match possible.
Hence, this could have undesired results as the any char could match a space or
other characters not legal in email addresses. Here's something a little more
complete:
[\w._-]+@([\w-]+\.)+[A-Za-z]+
After the @ we have alphanum or a dash followed by a dot. The parens are used
for grouping but can also have the effect of storing the matching content in a
regex memory location. The plus after the parens says one or more of the
pattern in the parens. Since top level domains are just letters right now, one
or more of those are matched for the end.
As you can see, regex can look like an alphabet soup of punctuation marks very
quickly.
James Keeline
