如何优化查询以更快地运行（查询内的查询）

Question

This query will take around 5 seconds to complete. I mean when I refresh or navigate to the page it will take 5 seconds to complete the browser loading and to display the total counts.

Here is what I want to achieve, FIRST is to get the MAX value of the systemID (ID) based on the empID. But before the First query ends I made another query the will get the row where the empID has a JUMP value on the eStatus col.

This is to compare the year difference of the latest data startDate of the empID and to his previous JUMP endDate.

Here is my table

---|-------|-----------|------------|---------
ID | empID | startDate |   endDate  | eStatus
---|-------|-----------|------------|---------
1  |   10  | 2001-1-31 | 2001-12-31 |
2  |   10  | 2002-1-31 | 2002-12-31 | 
3  |   22  | 2001-1-31 | 2001-12-31 |
4  |   10  | 2003-1-31 | 2003-12-31 | JUMP
5  |   10  | 2004-1-31 | 2004-12-31 |
6  |   22  | 2002-1-31 | 2002-12-31 | JUMP
7  |   10  | 2005-1-31 | 2005-12-31 |
8  |   22  | 2003-1-31 | 2003-12-31 |
9  |   22  | 2004-1-31 | 2004-12-31 |
10 |   10  | 2006-1-31 | 2006-12-31 | JUMP
11 |   10  | 2007-1-31 | 2007-12-31 |
12 |   10  | 2008-1-31 | 2008-12-31 |
13 |   10  | 2009-1-31 | 2009-12-31 | JUMP
14 |   10  | 2010-1-31 | 2010-12-31 |
15 |   10  | 2011-1-31 | 2011-12-31 |

the First query will get the max ID by group of empID.

---|-------|-----------|------------|---------
ID | empID | startDate |   endDate  | eStatus
---|-------|-----------|------------|---------
15 |   10  | 2011-1-31 | 2011-12-31 |
9  |   22  | 2004-1-31 | 2004-12-31 | 

the Second query will get the empID row that has a JUMP data on the eStatus Col

---|-------|-----------|------------|---------
ID | empID | startDate |   endDate  | eStatus
---|-------|-----------|------------|---------
4  |   10  | 2003-1-31 | 2003-12-31 | JUMP
6  |   22  | 2002-1-31 | 2002-12-31 | JUMP
10 |   10  | 2006-1-31 | 2006-12-31 | JUMP
13 |   10  | 2009-1-31 | 2009-12-31 | JUMP
6  |   22  | 2002-1-31 | 2002-12-31 | JUMP

Now I can compute the date difference from startDate of 1st query and enddate of 2nd query. If it is greater than 2 the it will count to my final count. THANK YOU SO MUCH IN ADVANCE and KEEP SAFE. Here is my code:

<?php
$counterA= 0;
$counterB= 0;
$finalCount = 0;
                  
  $result = mysqli_query($con,"SELECT empID,endDate FROM tablerecord 
  WHERE ID IN (SELECT MAX(ID) FROM tablerecord GROUP BY empID)");
  while ($row=mysqli_fetch_assoc($result )) {
    $counterA++; //count how many result based on the above query
    $emp_max = $row['empID']; //Get emp ID based on max ID
    endDate_result = $row['endDate']; //Get endDate based on max ID
        
    $resultPrevious=mysqli_query($con,"SELECT empID,startDate FROM tablerecord
    WHERE empID = '$emp_max' AND eStatus = 'JUMP' ");
    while ($rowPrevious=mysqli_fetch_assoc($resultPrevious)) {
                      
        $counterB++; //count how many result based on the above query
                                            
        $dateA=date_create($rowPrevious['startDate']);
        $dateB=date_create($endDate_result);
                            
        $diff2=date_diff($dateA,$dateB);
        $numLenght = $diff2->y;
                      
                      if ($numLenght > 2) {
                        $finalCount++;
                      }
    }
   }
   echo $counterA;
   echo "<br>";
   echo $counterB;
   echo "<br>";
   echo $finalCount;

?>

Answer 1

if I understand correctly, what you want is to select all data that have eStatus value jump .

why not simply SELECT empID,startDate FROM tablerecord WHERE empID IN(SELECT MAX(ID) FROM tablerecord) AND eStatus = 'JUMP' ?

or, if you want the unique value of empID , SELECT empID,startDate FROM tablerecord WHERE empID IN(SELECT DISTINCT(ID) FROM tablerecord) AND eStatus = 'JUMP'

all my query above will not produce any result you might want to try multiple select in your query for a crude way to get it all in one go

select * from tablerecord where empID in(SELECT empID FROM tablerecord WHERE ID IN (SELECT MAX(ID) FROM tablerecord group by empID)) and eStatus = 'JUMP'

in this way, you only query the DB one time and the rest can be done with your PHP code

Answer 2

I would use this query to get all needed data :

# Get first and last jump for each empID
SELECT * 
FROM tablerecord t1
WHERE t1.ID = (
    SELECT MIN(t2.ID)
    FROM tablerecord t2
    WHERE t1.empID==t2.empID
) or t1.ID = (
    SELECT MAX(t3.ID)
    FROM tablerecord t3
    WHERE t1.empID==t3.empID
)

And then some PHP to run through, comparing each pairs

$result = mysqli_query($con,"the_previous_query....");
$row_to_compare_with =false;
while ($row=mysqli_fetch_assoc($result )) {
        if ($row_to_compare_with == false) {
          $row_to_compare_with = $row;
        } else {
           // Compare $row_to_compare_with with $row, both having same empId
           // do your thing here...
           $row_to_compare_with = false;
        }
}