asked on

I need to grab an additional row based on a variable result

when l1.completed > startDate AND a.name LIKE '%Disable%'

Open in new window

true of the first row in the result set, I need to grab the previous row in the database and include it in the result set

CREATE PROCEDURE reportFreeCoolingTrackerDateTest (
        IN fromDate varchar (50),
        IN toDate varchar (50),
        IN timeZone varchar (50))

    BEGIN
            DECLARE startDate varchar (50);
            DECLARE endDate varchar (50);
            DECLARE mylogID Int;
            
             SET startDate = FROM_UNIXTIME(fromDate/1000);
             SET endDate = FROM_UNIXTIME(toDate/1000);
             
            
SELECT 
    CASE
        when l1.completed > startDate AND a.name LIKE '%Disable%' 
            THEN (time_to_sec(timediff(l1.completed,(CONVERT_TZ( (FROM_UNIXTIME(fromDate/1000)),'UTC', timeZone) ))) / 3600)
       END AS newTime,l1.recordId,
    a.activityId, 
    DATABASE() AS databaseName, ((l1.item31985 - l1.item31987)*(time_to_sec(timediff(t2.completed, l1.completed)))) / 3600  AS kwDifference,
    ((l1.item31985 - l1.item31987) *  (substr(l.details, instr(l.details , ':' ) +1))) AS cost,
    (((l1.item31985 - l1.item31987) *  (substr(l.details, instr(l.details , ':' ) +1)))
    *(time_to_sec(timediff(t2.completed, l1.completed)) / 3600))  AS costT,
      time_to_sec(timediff(t2.completed, l1.completed)) / 3600 AS coolingHours,
      time_to_sec(timediff(endDate, startDate)) / 3600 AS totalTimeRange,l1.completed, 
     (CONVERT_TZ( (FROM_UNIXTIME(fromDate/1000)),'UTC', timeZone) )AS StartingDate, 
     (CONVERT_TZ( (FROM_UNIXTIME(toDate/1000)),'UTC', timeZone) ) AS EndingDate,   
      timeZone  AS timeZonePassed

    
  FROM logs l 
      INNER JOIN groups g ON g.groupId = l.groupId
      LEFT JOIN groups g1 ON g.parentId = g1.groupId
      LEFT JOIN groups g2 ON g1.parentId = g2.groupId
      LEFT JOIN groups g3 ON g2.parentId = g3.groupId                
      INNER JOIN activities a ON l.logId = a.logId 
      INNER JOIN log1644 l1 ON a.activityId = l1.activityId 
      INNER JOIN log1644 t2 ON t2.recordId = l1.recordid + 1
      INNER JOIN items i ON l.logId = i.logId AND i.name LIKE '%KW%'
      INNER JOIN users u ON l1.userId = u.userId AND i.name LIKE '%KW%'
     WHERE i.itemID = "31985"  
      AND l1.started
        BETWEEN startDate
            AND endDate 
     ORDER BY l1.recordId,l1.started;
 
    END //

DELIMITER ;

Open in new window

gplana

You should create a cursor to access to the rows of the select. By creating a cursor you can process rows one by one: save the current row to a variable, maintaining the previous row on another variable (so you will need two row variables).

Please see this article about cursors on MySQL: http://dev.mysql.com/doc/refman/5.0/en/cursors.html

portlight

ASKER

I have not used cursors before, is it difficult to convert what I have to function as you are talking of.

gplana

Have you read the article? I think it shouldn't be difficult.

I would do it for you but the problem is I don't really understand what exactly you want. You said: "if (this) true of the first row in the result set, I need to grab the previous row in the database and include it in the result set ". Never exist a previous row of first row.

gplana

Also you can get first row just by adding a "TOP 1" just after SELECT word. This will make the SELECT to return just the first record instead of all records.

Then, you can also use variables to assign the value of columns when the SELECT statement returns exactly one row:

SELECT @myvar1=mycolumn1, @myvar2=mycolumn2, ...., @myvarN = mycolumnN
FROM bla bla....

portlight

ASKER

If the initial query grabs 10 of 100 rows - the first row of the result set being row 24 AND that first row meets this condition
when l1.completed > startDate AND a.name LIKE '%Disable%'

Then I need to figure out how to include row 23 in the result set. Even if the started time is outside of the between dates

gplana

Then I think you should do this:

store the recordId of the found record on a variable (for example v_id).

then make this query:

SELECT @v_previous_id = MAX(recordId)
FROM logl644
WHERE recordId < @v_id;

Then you can get the previous record by issuing this sentence:

SELECT *
FROM your_tables_and_join
WHERE ll.recordId = @v_previous_id;

portlight

ASKER

The problem that i have been running into is this creates two return data sets when the query has been run - is there any way to incorporate this into one return set.

ASKER CERTIFIED SOLUTION

gplana

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial