from code loop to Sql query? - mysql

Question

Edit:
I edit the post, to add some complexity. What if the box would contain repeated barcodes? Barcode 222222222222 exists twice in the box. And in the main table (or the warehouse) exists 3 times.
How would be the query so I delete only 2 barcodes?
The answer I chose groups the barcodes of the box, thus, only deletes one of each.
Hello all,
I have this table
Barcodes
+--+------------------+-------+
|ID|Barcode |DELETE |
+--+------------------+-------+
|1 | 11111111111111 | 0 |
|2 | 22222222222222 | 0 |
|3 | 22222222222222 | 0 |
|4 | 22222222222222 | 0 |
|5 | 33333333333333 | 0 |
|6 | 33333333333333 | 0 |
|7 | 33333333333333 | 0 |
|8 | 44444444444444 | 0 |
+--+------------------+-------+
ID is an autoincremental field.
Each line is a barcode which represents 1 unit of the same item. For example, I have 3 items represented by barcode 33333333333333
Everytime any user scans a barcode, and it matches, I mark the barcode to be deleted when the process is finished.
update barcodes
set delete=-1
where id=(select max(barcode) as maxi from barcodes where barcode like 'x' and delete>-1)
This works fairly fast. But sometimes, instead of scanning the items one by one, the user may scan a box which contains up to 150 items.
The back-end is a mysql table in a remote server, and the front-end is ms access.
BOX
+------------------+
|Barcode |
+------------------+
| 22222222222222 |
| 22222222222222 |
| 33333333333333 |
| 44444444444444 |
+------------------+
I use a loop, and for each item in the box, I repeat the sql query, but this takes up to 3 minutes for the 150 items box because of latency.
Do you know any kind of query that would update my barcodes tables with those in the read box?
The output of it would be something like:
+--+------------------+-------+
|ID|Barcode |DELETE |
+--+------------------+-------+
|1 | 11111111111111 | 0 |
|2 | 22222222222222 | -1 |
|3 | 22222222222222 | 0 |
|4 | 33333333333333 | -1 |
|5 | 33333333333333 | 0 |
|6 | 33333333333333 | 0 |
|7 | 44444444444444 | -1 |
+--+------------------+-------+
If the sale is finished, I would just delete those marked as delete=-1

Answer 1

What is the datatype of barcode? Avoid grouping to speed up updates. Note you are not deleting anything despite use of a column with that name. Remove like and > in the sql and replace them with = logic with something like the following for mysql server side code:
update barcodes
set delete=-1
where id=(select id from barcodes where barcode='x' and delete=0 ORDER BY id DESC LIMIT 1)
You'll have to adjust the above to suit the specifics of x in your example.

Answer 2

You can use the following query to pick the id values of the rows to be deleted:
SELECT MIN(id)
FROM Barcodes AS b1
INNER JOIN BOX AS b2 ON b1.Barcode = b2.Barcode
WHERE `DELETE` <> -1
GROUP BY b1.Barcode
If you use the above SELECT query in your original UPDATE query, then you have what you need:
UPDATE Barcodes
SET `DELETE` = - 1
WHERE id IN ( SELECT minID
FROM (SELECT MIN(id) AS minID
FROM Barcodes AS b1
INNER JOIN BOX AS b2 ON b1.Barcode = b2.Barcode
WHERE `DELETE` <> -1
GROUP BY b1.Barcode) AS t )
The above works in MySQL. See demo here.
Note that you need to wrap the SELECT query in a sub-query, since MySQL doen't let you specify target table 'Barcodes' for update in FROM clause.