I'm using the following query to replace old link with a new one:
UPDATE my_table SET file = 'link' WHERE my_table.file ='old_link';
In my tests I can't duplicate that and I'm not sure what's wrong with that query, but apparently sometimes it leaves the old entry and inserts a new one instead of updating!
mysql ver:
5.6.12-56 Percona Server, table type: innodb
The query looks fine to me. UPDATE should never create new rows, only modify existing rows. The problem is probably in another part of the code.
Although an UPDATE won't fire an insert in its own, there could exist triggers in the database that would fire an INSERT whenever a record gets updated.
Here are some links that you should check:
CREATE TRIGGER Syntax
Trigger Syntax and Examples
Related
I want to fetch the BEFORE and AFTER data from TableProduct whenever there is an update. So I created a Before and After trigger queries in which the respected values from the trigger is stored in separate tables titled TableProduct_Before and TableProduct_After.
My challenge is I always get duplicates result whenever I try to INNER JOIN the three tables.
I have tried the DISTINCT and ORDER BY ID command but still the same challenge.
I made enquiry I was told I can't join trigger table values because it doesn't have a foreign key, I tried adding a foreign key but it didn't work saying foreign key already exists in TableProduct.
Please, I will appreciate any help thank you.
You are misunderstanding how triggers work. You can use either a before or after trigger.
The key is using new and old. Inside the body of the trigger, you have both the old and new values. In pseudocode, this looks like:
insert into archive (pk, old_value, new_value)
values (new.pk, old.value, new.value);
Both OLD and NEW data is available in both before and after trigger.
The difference is - BEFORE trigger fixes the attempt to update whereas AFTER fixes successful attempt result. I.e. BEFORE is executed anycase (but its changes may be rollbacked!), rather than AFTER.
See https://dbfiddle.uk/?rdbms=mysql_8.0&fiddle=dd19c611c7e2fbef88b2133dcaa61dfa
PS. Recent MySQL versions allows triggers chains - a lot of triggers on the same event fired one-by-one. So check that your saving trigger is the most last in this chain.
I had a trigger set in MySQL workbench that used to accomplish what I wanted. Recently, my workplace updated MySQL workbench to version 6.3.10 across all computers and the triggers were dropped during the update for some reason.
Basically, the trigger should update a column with the current date when a different column is changed. This was previously achieved by using the following syntax:
if (NEW.colname <> OLD.colname or (OLD.colname is null && NEW.colname is not null))
then
set NEW.othercolname = current_date()
The issue is when I try to recreate the same trigger in the new MySQL version and apply it, I am met with the error:
ERROR 1363: There is no OLD row in on INSERT trigger
It seems to me that MySQL is looking for a row named OLD, which does not exist; however, this syntax used to work in previous versions for the intended result. Can anyone suggest an alternative syntax to accomplish this?
First of all, the error message comes from mysql, not from mysql workbench, so mysql workbench's version is irrelevant here.
Secondly, the error message is pretty clear: you are trying to refer to the OLD row in an insert trigger. This is obviously impossible, since there is no old row when you insert a new one.
You need to change either the trigger to catch an update event or change the trigger logic.
ERROR 1363: There is no OLD row in on INSERT trigger
You've got MySQL Workbench set to create an INSERT trigger. As the error message says, there's no OLD row in an INSERT trigger.
What you want is an UPDATE trigger.
I want the epc column to always be earnings/clicks. I am using an AFTER UPDATE trigger to accomplish this. So if I were to add 100 clicks to this table, I would want the EPC to update automatically.
I am trying this:
CREATE TRIGGER `records_integrity` AFTER UPDATE ON `records` FOR EACH ROW SET
NEW.epc=IFNULL(earnings/clicks,0);
And getting this error:
MySQL said: #1362 - Updating of NEW row is not allowed in after trigger
I tried using OLD as well but also got an error. I could do BEFORE but then if I added 100 clicks it would use the previous # clicks for the trigger (right?)
What should I do to accomplish this?
EDIT - An example of a query that would be run on this:
UPDATE records SET clicks=clicks+100
//EPC should update automatically
You can't update rows in the table in an after update trigger.
Perhaps you want something like this:
CREATE TRIGGER `records_integrity` BEFORE UPDATE
ON `records`
FOR EACH ROW
SET NEW.epc=IFNULL(new.earnings/new.clicks, 0);
EDIT:
Inside a trigger, you have have access to OLD and NEW. OLD are the old values in the record and NEW are the new values. In a before trigger, the NEW values are what get written to the table, so you can modify them. In an after trigger, the NEW values have already been written, so they cannot be modified. I think the MySQL documentation explains this pretty well.
Perhaps you could write two separate statements in that transaction
update record set clicks=...
update record set epc=...
or you could put them inside a function, say updateClick() and just call that function. By doing it this way you can easily alter your logic should the need arise.
Putting the logic inside a trigger might create a situation where debugging and tracing are made unnecessarily complex.
I'm trying to write a mini-auditing system for individual tables in MySQL.
I had good luck working with basic INSERT/UPDATE/DELETE commands, however now a table I'm interested in auditing uses INSERT ... ON DUPLICATE KEY UPDATE.
By using a trigger ON BEFORE INSERT I can tell this event is occurring, however I can only get half the data I'm interested in. NEW.values are readily available, but I've no idea how to get the OLD.values that came before. I suppose I could do a query using the NEW.ID in existing table, but I'm not sure about performance and reliability.
I need the OLD.values because I'm storing both old and new values for each change event since I read somewhere that was a good idea for collapsing data etc...
Is there a way in MySQL 5.0 (or newer GA release) to reliably retrieve these values as though I were in an UPDATE trigger?
EDIT: Another wrinkle:
Looks like the NEW.values do not match the data after update. They match the INSERT statement not the ON DUPLICATE KEY UPDATE data that will actually go into the record.
It looks like the trigger event ON AFTER UPDATE also catches the ON DUPLICATE KEY UPDATE change. From here, I was able to get OLD/NEW values and perform the logging I needed to perform.
I am fairly new to using mysql. I have an application that performs some basic querying. I am also trying to run a simple delete statement -
delete from mydb.mytable
This table is a simple 2 column table with not keys or triggers or anything defined. For some reason, the delete is not being performed. If I run the statement from MySql Workbench in the query window, it works fine. From the code, it does nothing. I am not seeing any error messages. I created a user with select, insert, update and delete rights to the schema. I am able to do the insert fine, but the delete does not seem to be working.
Is there a setting for mysql that I am missing that will not allow me to perform the delete?
Thanks for any thoughts.
Fist of all, check if
you are connected to the right database ;
you are using transaction and forgetting 'commit' ;
the user you use have enough permissions to delete from the table .
As a side notice, if you want to delete all records, you should use truncate instead of delete
Are you using transactions? My first guess is that your code might be issuing a BEGIN TRANSACTION without a COMMIT.
We would have to see some of your code to answer the question.
My guess is that you are not calling commit from your code. You can configure MySQL to auto-commit your queries, but this is usually not what you want.