There are two ways that you can do this.
PHP
Keep a running total of all of the "pay" values, and add it into your table at the bottom. For example:
$i=0;
while ($list = mysql_fetch_assoc($query)) { // for each row in your results
if ($id != $list['EmployeeId']) { // We only enter this loop if the EmployeeId doesn't equal $id. This can happen because either $id doesn't exist yet, or it doesn't match the previous EmployeeId
$i++; // increase $i by 1
if($i>1) { // Enter this loop only if $i is greater than or equal to 2 (if it is less than two, then this is our first time running this script, and adding a footer row wouldn't make any sense).
create footer (EmployeeId, Hours, Pay); // Log Id is irrelevant here
}
// reset your variables here
$id = $list['EmployeeId']; // set $id = the first or the new Employee ID
$total_pay = $list['pay']; // This is our first time for this Employee, so don't just add it to the running total
create header (EmployeeId, Hours, Pay) // Create the top half of your table
} else { // The EmployeeId has been established: we only need to change the running total
$total_pay = $total_pay + $list['pay'];
}
// add a data row for each LogId. This executes every time we go through the loop
create_normal_row(LogId, EmployeeId, Hours, Pay)
}
// At this point, both Employees have a header, and all data rows. However, we left the loop before we could add the last Employee's footer row
// Let's add one more footer row for the last user
create_footer (Logid, Hours, Pay);
SQL
MySQL has a function that does something very similar to what you are trying to do called ROLLUP. You can read more about it here:
http://dev.mysql.com/doc/refman/5.0/en/group-by-modifiers.html
Basically, you would change your query to work like this:
SELECT LogId, EmployeeId, SUM(Hours), SUM(Pay) FROM `log`
GROUP BY empid, logid WITH ROLLUP
This query will return a dataset that looks like this:
---------------------------------------
| Logid | EmployeeId| Hours | Pay |
---------------------------------------
| 1001 | 1 | 10 | 50 |
---------------------------------------
| 1002 | 1 | 2 | 10 |
---------------------------------------
| NULL | 1 | 12 | 60 |
---------------------------------------
| 1003 | 2 | 3 | 9 |
---------------------------------------
| 1004 | 2 | 6 | 18 |
---------------------------------------
| NULL | 2 | 9 | 27 |
---------------------------------------
| NULL | NULL | 21 | 87 |
---------------------------------------
Whenever $list['Logid'] is null, you know that you have a "total" row. Be careful though, this will add a "sum of all employees" row at the bottom of your dataset. If $list['EmployeeId'] is null, then you know you're in this "total" row.
On a related note (I'm not sure if this is what you're asking for), you can show this stuff in a table by using HTML <table> elements.
Each row would look like this:
<table> <!-- shown at the beginning of each table -->
<tr> <!-- shown at the beginning of each row -->
<td> <!-- shown at the beginning of each table cell -->
Your text goes here
</td> <!-- shown at the end of each table cell -->
<td>
More text can go here
</td>
</tr> <!-- shown at the end of each row -->
</table> <!-- shown at the end of each table -->
<tr>s can be repeated indefinitely within each <table>, and <td>s can be repeated within <tr>s.
