However, I believe that the trigger should return an error to the client. INSERT INTO RawData (LineItemID, PreState,CtimeIn,CtimeOut,PostState,Name,Value) VALUES (NULL, '1','1131405338.64','1131405338.6416','','Enter','11')Īs a application developer, this problem can be fixed by correcting the trigger to be The second should demonstrate the problem. Add three columns (and add a space between them) into one 'Address' column: SELECT CONCATWS (' ', Address, PostalCode, City) AS Address. Instead, I believe that this trigger should error out under such a condition and return an error to the client.Ĭreate the tables listed in description. When the row contains a null value, FieldNameValues becomes null itself.
However, The trigger has some abnormal behavior when we insert a row into RawData that contains a null value.
So, when there is a value in the column that is not null, that will be concatenated. The COALESCE function returns the first non-Null value. To avoid this, you can use the COALESCE function. Normally, when you insert a row into rawData, you get a trigger that inserts a row into ReplicationLog with a fieldNamevalues looking something like this When you concatenate any string with a NULL value, it will result in NULL. In addition, I have the following trigger that occurs after RawData inserts: An exception is made if the outer query has aggregates without any GROUP BY this means that it has a non-empty result. TableName varchar(64) NOT NULL default '', This change is made because, when the table used by the outer query is found empty at optimization, this sets the join column to NULL even when defined as not nullable, which leads to a conflict when optimizing the subquery. PostState varchar(100) NOT NULL default '', PreState varchar(100) NOT NULL default '', LineItemID int(10) unsigned default NULL,ĬtimeIn varchar(100) NOT NULL default '',ĬtimeOut varchar(100) NOT NULL default '', ID int(10) unsigned NOT NULL auto_increment,
0 kommentar(er)
