Question

I have a query that pulled out the report depends on Audit Date, but I’m very confuse on how the GROUP command is not working on what I’m expecting for the output.

Here is my queries,

SELECT prd.fldemployeeno `EmployeeNo`,
       prd.fldorderid `OrderNo`,
       prd.fldstarttime `TimeProcessed`,
       COUNT(qua.seqid) `ErrorCount`,
       COALESCE(qua.fldstarttime,(SELECT fldstarttime FROM tblproductionitl p
                                   WHERE (p.fldglobalid = prd.fldglobalid)
                                     AND p.fldprojectgroup=prd.fldprojectgroup
                                     AND p.fldstarttime > prd.fldstarttime
                                     AND prd.fldemployeeno != p.fldemployeeno
                                   LIMIT 0,1)) AS `AuditDate`
  FROM tblproductionitl prd
 INNER JOIN tblisauditeditl aud
    ON prd.fldglobalid=aud.fldid
  LEFT JOIN tblqualityaudit qua
    ON prd.fldglobalid=qua.fldid
   AND prd.fldstarttime=qua.fldprodstarttime
 GROUP BY prd.fldemployeeno,prd.fldorderid
 HAVING `AuditDate` BETWEEN '2011-10-04 00:00:00' AND '2011-10-04 23:59:59'
 ORDER BY `AuditDate`

And the output of this is

+-------------+---------------+---------------------+------------+---------------------+
| EmployeeNo  |   OrderNo     | TimeProcessed       | ErrorCount |  AuditDate          |
+-------------+---------------+---------------------+------------+---------------------+
| PSAA50577   | 20110930n01   | 2011-10-04 10:41:23 |   3        | 2011-10-04 10:44:07 |   
| PSAA50576   | 20111003n01   | 2011-10-03 11:39:52 |   1        | 2011-10-04 10:58:48 |
| PSAA50515   | 20110930n01   | 2011-10-04 10:44:07 |   1        | 2011-10-04 11:12:03 |
| PSAA50577   | 20111003n02   | 2011-10-03 12:22:33 |   1        | 2011-10-04 16:47:16 |
| PSAA50577   | 20110930n10   | 2011-10-01 18:27:09 |   1        | 2011-10-04 18:29:29 |
+-------------+---------------+---------------------+------------+---------------------+

And then I have removed the prd.fldorderid in the GROUP command so that the report will be grouped base from EmployeeNo only. But the output returns only 1 row instead of three. Please see below query and output.

SELECT prd.fldemployeeno `EmployeeNo`,
       prd.fldorderid `OrderNo`,
       prd.fldstarttime `TimeProcessed`,
       COUNT(qua.seqid) `ErrorCount`,
       COALESCE(qua.fldstarttime,(SELECT fldstarttime FROM tblproductionitl p
                                   WHERE (p.fldglobalid = prd.fldglobalid)
                                     AND p.fldprojectgroup=prd.fldprojectgroup
                                     AND p.fldstarttime > prd.fldstarttime
                                     AND prd.fldemployeeno != p.fldemployeeno
                                   LIMIT 0,1)) AS `AuditDate`
  FROM tblproductionitl prd
 INNER JOIN tblisauditeditl aud
    ON prd.fldglobalid=aud.fldid
  LEFT JOIN tblqualityaudit qua
    ON prd.fldglobalid=qua.fldid
   AND prd.fldstarttime=qua.fldprodstarttime
 GROUP BY prd.fldemployeeno
 HAVING `AuditDate` BETWEEN '2011-10-04 00:00:00' AND '2011-10-04 23:59:59'
 ORDER BY `AuditDate`

And the output for this query is:

+------------+--------------+---------------------+--------------+---------------------+
| EmployeeNo |  OrderNo     | TimeProcessed       | ErrorCount   |  AuditDate          |
+------------+--------------+---------------------+--------------+---------------------+
| PSAA50576  |  20111003n01 | 2011-10-03 11:39:52 | 1            | 2011-10-04 10:58:48 |
+------------+--------------+---------------------+--------------+---------------------+

Can anyone help me to analyze this on how only 1 row returned in second query and on how could I group the output base from Employee no.

Answer 1

You should check the way you are grouping, from MySQL doc:

MySQL extends the use of GROUP BY so that the select list can refer to
nonaggregated columns not named in the GROUP BY clause. This means
that the preceding query is legal in MySQL. You can use this feature
to get better performance by avoiding unnecessary column sorting and
grouping. However, this is useful primarily when all values in each
nonaggregated column not named in the GROUP BY are the same for each
group. The server is free to choose any value from each group, so
unless they are the same, the values chosen are indeterminate.
Furthermore, the selection of values from each group cannot be
influenced by adding an ORDER BY clause. Sorting of the result set
occurs after values have been chosen, and ORDER BY does not affect
which values the server chooses

Edit to explain. Given this example:

  SELECT column1
       , column2
GROUP BY column1

If the table can have different column2 given one value of column1 this is unsafe because anytime you execute the query you could get a different value of column2.
In your subquery you are doing this and you should rewrite the query to avoid it.