查询中的MySQL查询

Question

I have the following query which works well to query asterisk cdr for all missed queue calls but it is duplicating. That is, if there are two extensions in the queue call, then it's classified as two missed calls.

The query:

SELECT count( * ) as today_total_8
    FROM cdr 
    WHERE channel LIKE '%from-queue%' 
        AND disposition='NO ANSWER' 
        AND DATE(`calldate`) = DATE(CURDATE())

The column calldate is the same value for all two results; how can I add this to the above query to display count to one missed call?

Data returned on SELECT without count:

calldate            channel                 disposition     
------------------------------------------------------------------------
2014-02-08 01:15:02     Local/201@from-queue-00000012;2     NO ANSWER
2014-02-08 01:15:02     Local/200@from-queue-00000013;2     NO ANSWER
2014-02-08 00:18:04     Local/201@from-queue-00000010;2     NO ANSWER
2014-02-08 00:18:04     Local/200@from-queue-00000011;2     NO ANSWER

How can I do this, as I want my count to be 2?

Answer 1

SELECT count(distinct calldate ) as today_total_8
FROM cdr 
WHERE channel LIKE '%from-queue%' 
    AND disposition='NO ANSWER' 
    AND DATE(`calldate`) = DATE(CURDATE())
GROUP BY calldate

Answer 2

Change the SELECT list of the query to something like this:

SELECT COUNT(DISTINCT SUBSTRING_INDEX(cdr.channel,'@',1) AS today_total_8
  FROM ...

That will prevent "duplicate" values of the extension (the portion of channel preceding the '@' sign) from being counted more than once.

This is using the SUBSTRING_INDEX function to return just the "extension", eg 'Local/201' .

The DISTINCT keyword inside the COUNT aggregate adds 1 to the count for for each unique value of "extension".