It seems what's logged by show_sql is not what's actually sent to the data base (in this case, it's a DB2 instance).
This is logged by the application:
Hibernate:
insert
into
table
(id, col1, col2, col3, col4)
values
(default, ?, ?, ?, ?)
(id is an IDENTITY column).
But the DB2 server reports this query:
select id from final table (insert into table (id, col1, col2, col3, col4) values (default, ?, ?, ?, ?))
This is perfectly reasonable - hibernate needs the id that was generated by the db to update the entity, and this syntax seems to be the recommended way for DB2. However, I expected that the actual query would be logged.
This was observed in a Spring Boot application using JPA repositories, with hibernate as the implementation.
If I run the insert with a database client, or if I specify the insert manually (#Query(..., nativeQuery=true)) then the DB2 log shows just the insert.
My guess is that the db2 driver changes the query, probably because of the need to retrieve the newly generated id. Is that correct? And is there any way to log the actual queries, without using an interceptor such as those used to view the bound parameters?
This behavior is controlled by the useIdentityValLocalForAutoGeneratedKeys property turned off by default.
useIdentityValLocalForAutoGeneratedKeys
Specifies whether the IBM Data Server Driver for JDBC and SQLJ uses only the SQL built-in function IDENTITY_VAL_LOCAL to determine automatically generated key values.
The data type of this property is boolean.
Possible values are:
true
Specifies that the IBM Data Server Driver for JDBC and SQLJ always uses the SQL built-in function IDENTITY_VAL_LOCAL to determine automatically generated key values. The driver uses IDENTITY_VAL_LOCAL even if it is possible to use SELECT FROM INSERT.
Specify true if the target data server supports SELECT FROM INSERT, but the target objects do not. For example, SELECT FROM INSERT is not valid for a table on which a trigger is defined.
false
Specifies that the IBM Data Server Driver for JDBC and SQLJ determines whether to use SELECT FROM INSERT or IDENTITY_VAL_LOCAL to determine automatically generated keys. false is the default.
Common IBM Data Server Driver for JDBC and SQLJ properties for Db2 servers
Yes this is probably the driver that changes the request. You can check which query is sent to the server by the application by using VisualVm. It has an option to profile the requests sent by the driver :
Just connect to the application, start the JDBC Profiler and launch the request. You will see the request in the 'profiling results'
Related
I'm running a script to insert some data in a MySQL database, and it runs properly in the MySQL workbench. However, when I try to run it from Java via the JDBC, I get an error. The script is:
INSERT INTO `pa_record` (`username`, `pa_record_type`, `record_time`) VALUES (?, ?, CURRENT_TIMESTAMP);
SET #record_id := 1;
INSERT INTO `pa_crud` (`pa_record_id`, `table_name`) VALUES (#record_id, ?);
The error I get from the JDBC is
You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'SET #record_id := 1;
INSERT INTO pa_crud (pa_record_id, table_name) VALUES' at line 2
Any ideas?
Many (most?) database access libraries do not allow multiple statements in a single execute; those that do, usually need to have the feature activated with a setting change. It's likely not complaining about the SET, but that you had anything after the end of the first query at all.
I have a java application that does a SQL query against an Oracle database, that for some reason gives way less values when executed from the SQL Developer and from the application itself.
Now to the technicalities. The application produces a connection to the db using a wrapper library that employs c3p0. The c3p0 config has been checked, so we know that this things can't be:
-Pointing to wrong database/schema
-Restricted user
Then there's the query:
select to_char(AGEPINDW.TRANSACTION.TS_TRANSACTION,'yyyy-mm') as Time,result, count(*) as TOTAL, sum(face_value) as TOTAL_AMOUNT
from AGEPINDW.TRANSACTION
where (ts_transaction >= to_timestamp(to_char(add_months(sysdate,-1),'yyyy-mm'),'yyyy-mm')
and ts_transaction < to_timestamp(to_char(sysdate,'yyyy-mm'),'yyyy-mm')) and service_id in (2,23)
group by to_char(AGEPINDW.TRANSACTION.TS_TRANSACTION,'yyyy-mm'), result;
It doesn't have any parameter and is executed via your standard PreparedStatement. Yet the return from the app is wrong and I don't know what may be. Any suggestions?
We are dealing with sybase database in core java code. We are using org.eobjects.metamodel.DataContext to parse query.
String sCol[]=table.getColumnNames();
Query query=dataContext.query().from(table.getName()).select(sCol).toQuery();
return new QueryIterator(dataContext.executeQuery(query).iterator());
But it executing query. Same code working fine with Oracle database to parse and execute query.
Some of query example generated are :
select City.CityName from ownername.City
select City.CityName from City
select CityName from ownername.City
select CityName from City
select ownername.City.CityName from ownername.City
SELECT "City"."CityName" FROM ownername."City"
select * from ownername.City
No any of above query executed. We are getting these errors :
Not capable of parsing FROM token: "ownername"."City"
Could not execute query: The column prefix '"City"' does not match with a table name or alias name used in the query. Either the table is not specified in the FROM clause or it has a correlation name which must be used instead.
Not capable of parsing SELECT token: ownername.City.CityName
How can we execute query using metamodel wih SYBASE database OR is there any other way to execute sybase queries?
Oracle (and Microsoft) use a schema logical construct that Sybase ASE does not. In SAP/Sybase ASE, all tables and columns in a database are in the same schema. It is possible to users to have their own objects in the database though, so there is the possibility of imitating some of the schema behavior using user ownership, but it would require an extra level of effort.
For Sybase the proper query syntax would be:
SELECT [ Col1, Col2 | * ]
FROM [dbName].[ownerName.]TABLE
In your case
SELECT CityName
FROM dbName.ownername.City
In Sybase ASE, it's typically best practice to have all objects owned by 'dbo', so in that case you can omit the owner from the query:
SELECT CityName
FROM dbName..City
Full query syntax and information can be found in the Query Section of the Transact SQL Users Guide in the documentation.
The error messages you're getting are coming from MetaModel's query parser layer. It is searching for matching column and table names in the metadata of your database before it's even firing the query.
I notice that you're using namespace "org.eobjects.metamodel". You should upgrade to Apache MetaModel ("org.apache.metamodel") if possible since a lot has been improved in MetaModel since it's introduction into Apache. Including a lot of query parser improvements.
I have 2 tables A and B with reporting_date_time as timestamp(6).
When I query table A executes fine but not on table B
This is my query:
select * from A where reporting_date_time = ?
I use JDBC PreparedStatements and pass the argument through: prepStmt.setTimestamp(1, myTimestampVar);.
private java.sql.Timestamp myTimestampVar;
myTimestampVar.toString shows the value as '2014-11-26 00:00:00.0'
Strange thing is that same application gets data from both tables in user testing, while it fails to do so in production even though it is the same JDBC driver on both environments.
I got oracle trace enabled and found that in table A timestamp is passed to the query as 26-NOV-14 12.00.00 AM, while that in table B it is passed as 26-NOV-14 00.00.00.
Why would it is sending it in two different formats?
This query works when I input it through phpmyadmin.
INSERT INTO conversation (user_id) VALUES (?);
INSERT INTO conversation (conversation_id, user_id)
VALUES ((SELECT LAST_INSERT_ID()), ?)
However when I send that query using jdbc and java I get an error -
You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'INSERT INTO conversation (conversation_id, user_id) VALUES ((SELECT LAST_INSERT_' at line 1"
I am using the exact same query. I checked by calling toString on the PreparedStatement and copying and pasting it into phpmyadmin and executing it and it worked fine. It just doesn't work through java. Any ideas whats wrong?
By default, you cannot execute multiple statements in one query through JDBC. Splitting it into two calls will work, as will changing the allowMultiQueries configuration property to True.
JDBC Configuration Properties — allowMultiQueries:
Allow the use of ';' to delimit multiple queries during one statement (true/false), defaults to 'false', and does not affect the addBatch() and executeBatch() methods, which instead rely on rewriteBatchStatements.
Default value: false