Re: Simple example of N+1 selects
Magnus Stattin wrote: Hi How would I be able to avoid the N+1 selects if I have for example a departments table and a table with employees. All employees belongs to a department and I would like to get a list of department valueObjects where each departmentVo contains a collection of employee valueObjects. Is it possible to do this with a single select with Ibatis and how would the Sqlmap look like. I am looking for a simple example to get me started. Thank you Magnus Take a look at this wiki FAQ: http://wiki.apache.org/ibatis/How_20do_20I_20get_20around_20the_20n_2b1_20selects_20problem_3f Kris -- Kris Jenkins Email: [EMAIL PROTECTED] Blog: http://cafe.jenkster.com/ Wiki: http://wiki.jenkster.com/
Feedback
Hello all, I write this mail to share with you all my satisfaction in using iBATIS. The project I was working on until yesterday has been completed and delivered perfectly in time. Since it was a db-intensive web application (not much business logic, a lot of queries, a not usual db configuration) I consider iBATIS as an important factor that contributed to our success. Its use permitted us to be better focused on other application aspects, with simple XML changes and without recompiling the Java code. THIS, is the way J2EE should take. Thanks a lot, Clinton. Thanks to all of the people that work on this great project. Fabrizio Gianneschi Italy
Re: a few questions
I cannot speak for the rest of the team, but if it were up to me, probably not. :-/ Some of the most attractive things about iBATIS (for many users) are simplicity, the fact that it does not mess with your SQL unless it is absolutely neccessary, that it does not impose anything on your development, and that it is not database specific. Please do not take my comments the wrong way: I am not saying that what you are asking for is of no value, but many of them go directly against those points. Creating DDL, is VERY database specific, especially when you start talking about generating keys, datatypes, tablespaces, and so on. The differences between Oracle, MySQL, and M$SQL alone would be significant enough to eat up weeks of development time that could be used to do other things. The other two things on the list (3 4), IMO are more the domain of the DAO or business logic layer. But you may be able to convince me otherwise! I am not dead against us doing that if we can do it in a way that meets the criteria above. :-) One of the valuable things in iBATIS is that it is powerful, yet constructed in such a way that you can leverage it without becoming married to it. Let me see if I can explain that in a way that makes sense. Before using iBATIS, I used Hibernate. It sounded great - it was very powerful and did many of the things you described above. It would generate DDL, do optimistic locking, and inserts like you described worked. But, the problem was that I had to define my database according to Hibernate's rules. I also needed to have a stateful environment to keep persisted objects in to keep them fresh for Hibernate. I also had to implement Hibernate specific interfaces in my code to make certain features work. Pretty soon, I was not writing applications according to the standards of the company I worked for, but instead, according to the standards of Hibernate. That (to me) is an anti-pattern called Vendor Lock-in, and to be avoided. Now, I use iBATIS. My applications could not care less that I am using iBATIS, because the only part that has to be aware of it is the DAO layer. Yes, I have to implement the audit trail, and optimistic locking on my own, but, if I want to use another DAO implementation, I can. Larry On Fri, 28 Jan 2005 11:02:06 +0800, Nathan Coast [EMAIL PROTECTED] wrote: great, thanks for the quick response, its pretty much as I figured when I couldn't find anything covering these in the docs. Do you know if there is any support for these features planned in future releases? Larry Meadors wrote: 1) No. 2) Yes, you can do that for either auto-generated or assigned keys. 3) No. 4) No. That said, 3 4 can be pretty simple to build, depending on how intricate you want to get with them. Larry On Fri, 28 Jan 2005 10:16:42 +0800, Nathan Coast [EMAIL PROTECTED] wrote: 1) ddl - is it possible to define your table structures in ibatis and have ibatis create the tables? 2) inserts - is the primary key retuned for all calls to the SqlMapExecutor.insert method. I have only seen examples in the docs for inserts that require a selectKey (oracle, ms sql server) what about auto generated keys (e.g. mysql)? 3) optimistic locking - is there any declarative support for optimistic locking - e.g. having a last_modified column in the db / object throw an error during the update method if the last mod in the value is different from the db. 4) audit trail - 4 cols: created_by, created_timestamp, modified_by, modified_timestamp. again updated by the framework during insert, update statements. would require some form of authentication (jaas) from the enclosing container. I'm guessing the majority of apps run in one container or another tomcat, jboss etc so should be possible. Just trying to get a feel for what we'd get for free, what we'd have to build and what may be coming soon. -- Nathan Coast Managing Director Codeczar Ltd mob : (852) 9049 5581 tel : (852) 2834 8733 fax : (852) 2834 8755 web : http://www.codeczar.com
Re: One DAO implementatino for multiple databases
You can only declare a single Dao implementation per DaoManager. This is due to the fact that the Dao interface name is used as the key to lookup the implementation. I think the easiest route for you may be to extend the interface of your Daos interfaces that are shared. That way you will have a unique interface for each yet they can have the same implementation class. Brandon On Thu, 27 Jan 2005 15:04:52 -0800, Tony Li [EMAIL PROTECTED] wrote: Hi all I'm new to iBatis, and I want to find out the best practice for this problem. I want to use the same DAO implementation on a number of different databases. I'm pulling in data from an MS Access database (yes, please feel sorry for me here) that's been deployed in several locations. I'm using the both the iBATIS DAO layer and SqlMaps. Since I declare a DAO implementation under a specific Transaction Manager, and only one DataSource can be specified per transaction, I'm not sure what to do. If I define the same DAO implementation under a different Transaction Manager and Context, how would iBatis know which one to use when I ask for the DAO from the DaoManager? Any suggestions? Thanks, Tony
Re: Two separate map configs
Nah, DAO is super simple to use.
http://reumann.net/struts/ibatisLesson1.do
This covers an older version, but the one and two are very similar.
Larry
On Fri, 28 Jan 2005 10:23:23 -0500, Akins, Greg [EMAIL PROTECTED] wrote:
To be honest I was intimidated by learning that.
I had used Hibernate; so apart from getting up to speed on Ibatis; at
least I was familiar with Persistence layers..
However, I haven't used a DAO framework and am concerned that I won't be
able to get up to speed quickly enough.
-Original Message-
From: Brandon Goodin [mailto:[EMAIL PROTECTED]
Sent: Friday, January 28, 2005 10:11 AM
To: ibatis-user-java@incubator.apache.org
Subject: Re: Two separate map configs
yeah... use the DAO framework. :D
On Thu, 27 Jan 2005 16:55:56 -0500, Akins, Greg [EMAIL PROTECTED]
wrote:
OK, I stripped out my DAO objects for the second mapping (and the old
tests; and the new tests work... ) leading me to believe that I have
not reset an instance of SqlMapClient.
I'll have to rewrite my DAO super classes... Can anyone point me to a
good pattern for constructing those (since mine obviously sucked:')
Thanks for the pointers.
-Original Message-
From: Clinton Begin [mailto:[EMAIL PROTECTED]
Sent: Thursday, January 27, 2005 2:12 PM
To: Akins, Greg
Cc: ibatis-user-java@incubator.apache.org
Subject: Re: Two separate map configs
Greg,
Do you have two SqlMapClient instances? (you should)
Cheers,
Clinton
On Thu, 27 Jan 2005 13:55:53 -0500, Akins, Greg [EMAIL PROTECTED]
wrote:
Thanks... I'm not getting any other errors when I run this. I can
post the rest of the code, but basically I'm Querying by calling
Object prod = sqlMap.queryForObject(getWare, srchObj.getName()) ;
And am getting this stacktrace from my tests.
com.ibatis.sqlmap.client.SqlMapException: There is no statement
named getWare in this SqlMap.
at
com.ibatis.sqlmap.engine.impl.SqlMapExecutorDelegate.getMappedStatem
en
t(
SqlMapExecutorDelegate.java:288)
at
com.ibatis.sqlmap.engine.impl.SqlMapExecutorDelegate.queryForObject(Sq
lM
apExecutorDelegate.java:552)
at
com.ibatis.sqlmap.engine.impl.SqlMapExecutorDelegate.queryForObject(Sq
lM
apExecutorDelegate.java:536)
at
com.ibatis.sqlmap.engine.impl.SqlMapSessionImpl.queryForObject(SqlMapS
es
sionImpl.java:97)
at
com.ibatis.sqlmap.engine.impl.SqlMapClientImpl.queryForObject(SqlMapCl
ie
ntImpl.java:69)
at
com.amervideo.ehold.dao.MesBaseDAO.getObject(MesBaseDAO.java:68)
at
com.amervideo.ehold.dao.WareDAO.getObjectByName(WareDAO.java:22)
at eHoldTest.TestWareDAO.testGetWare(TestWareDAO.java:39)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native
Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(Unknown
Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(Unknown
Source)
at java.lang.reflect.Method.invoke(Unknown Source)
at junit.framework.TestCase.runTest(TestCase.java:154)
at junit.framework.TestCase.runBare(TestCase.java:127)
at junit.framework.TestResult$1.protect(TestResult.java:106)
at
junit.framework.TestResult.runProtected(TestResult.java:124)
at junit.framework.TestResult.run(TestResult.java:109)
at junit.framework.TestCase.run(TestCase.java:118)
at junit.framework.TestSuite.runTest(TestSuite.java:208)
at junit.framework.TestSuite.run(TestSuite.java:203)
at
org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.runTests(Remote
Te
stRunner.java:421)
at
org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.run(RemoteTestR
un
ner.java:305)
at
org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.main(RemoteTest
Ru
nner.java:186)
SqlMapsMES.xml
?xml version=1.0 encoding=UTF-8?
!DOCTYPE sqlMapConfig
PUBLIC -//iBATIS.com//DTD SQL Map Config 2.0//EN
http://www.ibatis.com/dtd/sql-map-config-2.dtd;
sqlMapConfig
properties resource=mesDb.properties /
transactionManager type=JDBC
dataSource type=SIMPLE
property name=JDBC.Driver
value=${driver}/
property name=JDBC.ConnectionURL
value=${url}/
property name=JDBC.Username
value=${username}/
property name=JDBC.Password
value=${password}/
/dataSource
/transactionManager
sqlMap resource=com/amervideo/ehold/sql/MoldSQL.xml /
sqlMap resource=com/amervideo/ehold/sql/WareSQL.xml /
/sqlMapConfig
WareSQL.xml
?xml version=1.0 encoding=UTF-8 ?
!DOCTYPE sqlMap PUBLIC -//iBatis.com//DTD SQL Map 2.0//EN
Re: Feedback
should we put these feedback on the wiki. It would be great to record successful experiences with IBatis. On Fri, 28 Jan 2005 10:50:22 +0100, Fabrizio Gianneschi [EMAIL PROTECTED] wrote: Hello all, I write this mail to share with you all my satisfaction in using iBATIS. The project I was working on until yesterday has been completed and delivered perfectly in time. Since it was a db-intensive web application (not much business logic, a lot of queries, a not usual db configuration) I consider iBATIS as an important factor that contributed to our success. Its use permitted us to be better focused on other application aspects, with simple XML changes and without recompiling the Java code. THIS, is the way J2EE should take. Thanks a lot, Clinton. Thanks to all of the people that work on this great project. Fabrizio Gianneschi Italy
Re: Mapping an Oracle CHAR(1) column to java char property
Chars haven't been supported to date, because JDBC does not support them. However, I suppose there's no reason why you couldn't build a custom type handler for that Clinton On Fri, 28 Jan 2005 11:07:08 +1300, Stephen Chell (DSL AK) [EMAIL PROTECTED] wrote: Hi, I'm having trouble mapping an oracle CHAR(1) column to a javabeans property that is of type char. I'm using SQL Maps 2.0.9 with Spring 1.1.3. Here is my resultMap (defined in the mapping file ChargeType.xml): resultMap id=result class=nz.ac.otago.its.chargeadmin.domain.ChargeType result property=code column=CHARGE_TYPE_CODE columnIndex=1/ result property=narration column=NARRATION columnIndex=2/ result property=status column=STATUS columnIndex=3/ result property=versionNo column=VERSION_NO columnIndex=4/ result property=versionDateTime column=VERSION_DATE_TIME columnIndex=5/ result property=versionUser column=VERSION_USER columnIndex=6/ /resultMap The offending column is the status column, which is a CHAR(1) in Oracle and a char property in the ChargeType class. When I try to execute a query that uses this result map, the following exception occurs: org.springframework.jdbc.UncategorizedSQLException: (SqlMapClient operation): encountered SQLException [ --- The error occurred in nz/ac/otago/its/chargeadmin/dao/ibatis/maps/ChargeType.xml. --- The error occurred while applying a result map. --- Check the ChargeType.result. --- Check the result mapping for the 'status' property. --- Cause: com.ibatis.sqlmap.client.SqlMapException: No type handler could be found to map the property 'status' to the column 'STATUS'. One or both of the types, or the combination of types is not supported.]; nested exception is com.ibatis.common.jdbc.exception.NestedSQLException: --- The error occurred in nz/ac/otago/its/chargeadmin/dao/ibatis/maps/ChargeType.xml. --- The error occurred while applying a result map. --- Check the ChargeType.result. --- Check the result mapping for the 'status' property. --- Cause: com.ibatis.sqlmap.client.SqlMapException: No type handler could be found to map the property 'status' to the column 'STATUS'. One or both of the types, or the combination of types is not supported. com.ibatis.common.jdbc.exception.NestedSQLException: --- The error occurred in nz/ac/otago/its/chargeadmin/dao/ibatis/maps/ChargeType.xml. --- The error occurred while applying a result map. --- Check the ChargeType.result. --- Check the result mapping for the 'status' property. --- Cause: com.ibatis.sqlmap.client.SqlMapException: No type handler could be found to map the property 'status' to the column 'STATUS'. One or both of the types, or the combination of types is not supported.Caused by: com.ibatis.sqlmap.client.SqlMapException: No type handler could be found to map the property 'status' to the column 'STATUS'. One or both of the types, or the combination of types is not supported. at com.ibatis.sqlmap.engine.mapping.statement.GeneralStatement.executeQuery WithCallback(GeneralStatement.java:183) at com.ibatis.sqlmap.engine.mapping.statement.GeneralStatement.executeQuery ForList(GeneralStatement.java:118) at com.ibatis.sqlmap.engine.impl.SqlMapExecutorDelegate.queryForList(SqlMap ExecutorDelegate.java:610) at com.ibatis.sqlmap.engine.impl.SqlMapExecutorDelegate.queryForList(SqlMap ExecutorDelegate.java:584) at com.ibatis.sqlmap.engine.impl.SqlMapSessionImpl.queryForList(SqlMapSessi onImpl.java:105) at org.springframework.orm.ibatis.SqlMapClientTemplate$3.doInSqlMapClient(S qlMapClientTemplate.java:202) at org.springframework.orm.ibatis.SqlMapClientTemplate.execute(SqlMapClient Template.java:142) at org.springframework.orm.ibatis.SqlMapClientTemplate.executeWithListResul t(SqlMapClientTemplate.java:164) at org.springframework.orm.ibatis.SqlMapClientTemplate.queryForList(SqlMapC lientTemplate.java:200) at nz.ac.otago.its.chargeadmin.dao.ibatis.SqlMapChargeTypeDao.getChargeType List(SqlMapChargeTypeDao.java:22) [etc] Explicity specifing the type of the status property in the result element using the javaType and/or jdbcType attributes doesn't help. I tried all of the following combinations: result property=status column=STATUS columnIndex=3 javaType=char/ result property=status column=STATUS columnIndex=3 javaType=java.lang.Character/ result property=status column=STATUS columnIndex=3 jdbcType=CHAR/ result property=status column=STATUS columnIndex=3 javaType=char jdbcType=CHAR/ result property=status column=STATUS columnIndex=3 javaType=java.lang.Character jdbcType=CHAR/ None of these work, although the exception when specifying javaType=char is different, as shown below. What am I doing wrong? Thanks in advance Steve
A question about nullValue in resultMap.
Hi, So far in iBbatis, when the parameter of the result Bean's setter is primitive type (for example, setParentCategoryID(long id)), if the corresponding column might get a database NULL value during execution, we have at least three solutions: 1. Change the Bean's setter to use Object type parameter instead of primitive type. (for example, use setParentCategoryID(Long id_object) instead). 2. Use ResultMap and specify nullValue of the nullable property. 3. Use ResultMap and specify TypeHandler to take care of the returned Null value. Obviously solution 1 is not a good one while solution 2 and 3 require using ResultMap instead of using ResultClass directly. Now the question is, I want to keep the sqlmap xml file as simple as possible, so I want to use as much implicit result mapping as possible. (Because using resultMap will make the xml file a little bit harder to read, understand, debug and maintain.) I want to avoid ResultMap even there is nullable column in the SQL statement. So I hope iBatis could take care of the null values for me. I am wondering whether the iBatis team will consider adding one more property to the iBatis configuration file like SkipSettingNullResult. If SkipSettingNullResult=true, while iBatis sees a Null value returned from the resultSet, if the corresponding setter is taking a primitive type parameter, iBatis simply skip this setter and the Bean will use the default value instead of executing the setter. I will sincerely appreciate it if the iBatis team could make it happen in the near future. Thanks a lot. Bing
Re: A question about nullValue in resultMap.
Just curious: If you can have null values, why are you using primitive
values. You do realize that even if we do not call the setter, the
property holder will be set to 0 anyway, right?
public class MyBean{
int myProp; // -- since this MUST be initialized, the JVM sets it to 0
public int getMyProp(){
return myProp;
}
public void setMyProp(int val){
myProp = val;
}
public static void main(String[] a){
MyBean myBean = new MyBean();
System.out.println(myBean.getMyProp()); // -- will be 0
}
}
So, do you have some other logig in your setter or something that
detects if it was called, or is there some other compelling reason for
this? IMO, it seems silly.
Larry
On Fri, 28 Jan 2005 15:10:19 -0500, Bing Zou [EMAIL PROTECTED] wrote:
Hi,
So far in iBbatis, when the parameter of the result Bean's setter is
primitive type (for example, setParentCategoryID(long id)), if the
corresponding column might get a database NULL value during execution,
we have at least three solutions:
1. Change the Bean's setter to use Object type parameter instead of
primitive type. (for example, use setParentCategoryID(Long id_object)
instead).
2. Use ResultMap and specify nullValue of the nullable property.
3. Use ResultMap and specify TypeHandler to take care of the returned
Null value.
Obviously solution 1 is not a good one while solution 2 and 3 require
using ResultMap instead of using ResultClass directly.
Now the question is, I want to keep the sqlmap xml file as simple as
possible, so I want to use as much implicit result mapping as
possible. (Because using resultMap will make the xml file a little bit
harder to read, understand, debug and maintain.) I want to avoid
ResultMap even there is nullable column in the SQL statement. So I
hope iBatis could take care of the null values for me.
I am wondering whether the iBatis team will consider adding one more
property to the iBatis configuration file like SkipSettingNullResult.
If SkipSettingNullResult=true, while iBatis sees a Null value returned
from the resultSet, if the corresponding setter is taking a primitive
type parameter, iBatis simply skip this setter and the Bean will use
the default value instead of executing the setter.
I will sincerely appreciate it if the iBatis team could make it happen
in the near future.
Thanks a lot.
Bing
jpetstore: port to UNIX?
Hi - has anyone ported JPetStore 4 to MySQL 4.1.x and Tomcat 5.5.4 under UNIX - from whence it came? It appears to be a question of using the case on the SQL commands consistently. For instance, the MySQL schema creates a database 'JPETSTORE' with upper case, and the 'account' table and columns with lower case create table account ( userid varchar(80) not null, email varchar(80) not null, firstname varchar(80) not null, lastname varchar(80) not null, status varchar(2) null, addr1 varchar(80) not null, addr2 varchar(40) null, city varchar(80) not null, state varchar(80) not null, zip varchar(20) not null, country varchar(20) not null, phone varchar(80) not null, constraint pk_account primary key (userid) ); but the sqlmapdao generates a mapping for upper case column names sqlMap namespace=Account typeAlias alias=account type=com.ibatis.jpetstore.domain.Account/ resultMap id=accountResult class=account result property=username column=USERID/ result property=email column=EMAIL/ result property=firstName column=FIRSTNAME/ result property=lastName column=LASTNAME/ result property=status column=STATUS/ result property=address1 column=ADDR1/ result property=address2 column=ADDR2/ result property=city column=CITY/ result property=state column=STATE/ result property=zip column=ZIP/ result property=country column=COUNTRY/ result property=phone column=PHONE/ result property=languagePreference column=LANGPREF/ result property=favouriteCategoryId column=FAVCATEGORY / result property=listOption column=MYLISTOPT / result property=bannerOption column=BANNEROPT / result property=bannerName column=BANNERNAME / /resultMap [snip] -- Ken
WebSphere 5.1.X -- LocalTransaction rolled-back due to setRollbackOnly
Hello.
We are currently developing our first application with iBATIS, and we are extremely impressed by its design and capabilities. Everything was going great until we deployed our application to WAS 5.1 utilizing a JNDI datasource.
We end up receiving the following exception:
[1/28/05 20:31:45:954 CST] 796d3dfd WebGroup E SRVE0026E: [Servlet Error]-[LocalTransaction rolled-back due to setRollbackOnly]: com.ibm.ws.LocalTransaction.RolledbackException
I've read the following post concerning issues with WebSphere:
http://wiki.apache.org/ibatis/Database_20Specific_20Information
But I'm afraid that we cannot implement the suggestions since we externally acquire a connection and pass it to the SqlMapClient. Therefore, I'm not sure how or if we can set atransactionManager element within our configuration file since it requires a dataSource element as a child. Currently, we do not have a transactionManager element at all.
Here is the code we use to pass the connection (sorry about the formatting -- I'm using a web-based e-mail editor):
/**
* Returns the cached code[EMAIL PROTECTED] SqlMapClient}/code instance for the DataSource alias.
*/
protected SqlMapClient getSqlMapClient(String dataSourceAlias)
throws SQLException {
try {
SqlMapClient sqlMapClient =
IBatisSqlMapService.getInstance().getSqlMapClient(
dataSourceAlias);
if (sqlMapClient.getCurrentConnection() == null) {
//try to acquire a Connection from the abstract broker
try {
Connection connection =
getDatabaseConnection(dataSourceAlias);
if (connection != null) {
sqlMapClient.setUserConnection(connection);
logger.debug(
"Connection set via Common Services for \""
+ dataSourceAlias
+ "\".");
} else {
logger.debug(
"Connection not configured via Common Services for \""
+ dataSourceAlias
+ "\". Assuming iBatis will supply connection.");
}
} catch (SQLException e) {
logger.debug(
"Connection not configured via Common Services for \""
+ dataSourceAlias
+ "\". Assuming iBatis will supply connection.");
}
}
return sqlMapClient;
} catch (IOException e) {
throw new Error(e);
}
}
Also we "clean up" the application supplied user connection with the following code:
/**
* Clear out the code[EMAIL PROTECTED] Connection}/code instanced stored by this thread's
* code[EMAIL PROTECTED] SqlMapClient}/code instance keyed by codedataSourceName/code.
*/
public void closeConnection(String dataSourceName) throws SQLException {
try {
//clear out user connection on sql map
SqlMapClient sqlMapClient =
IBatisSqlMapService.getInstance().getSqlMapClient(dataSourceName);
if (sqlMapClient != null) {
//explicitly commit transaction prior to close
Connection connection = sqlMapClient.getCurrentConnection();
if (connection != null) {
//attempt to prevent exception -- didn't work
connection.commit();
}
sqlMapClient.setUserConnection(null);
logger.debug(
"The transaction for SqlMapClient \""
+ dataSourceName
+ "\" has been cleared for calling thread.");
}
//close connection
super.closeConnection(dataSourceName);
} catch (IOException e) {
throw new Error(e.getMessage());
}
}
The IBatisSqlMapService is a singleton that caches SQLMapClients.
I would appreciate any advice about how I could prevent the exception from occurring. Does iBATIS take part in transactions even if the connection is supplied externally? I'm also aware that the issue could be with our configuration of WebSphere, but I would like some ideas before I approach our WebSphere administrator.
Thanks,
Mark__Do You Yahoo!?Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
Re: WebSphere 5.1.X -- LocalTransaction rolled-back due to setRollbackOnly
Please set transactionManagercommitRequired=true
More info is available in the following poorly titled wiki (my fault):
http://wiki.apache.org/ibatis/Database_20Specific_20Information
Clinton
On Fri, 28 Jan 2005 19:09:29 -0800 (PST), Mark Nabours
[EMAIL PROTECTED] wrote:
Hello.
We are currently developing our first application with iBATIS, and we are
extremely impressed by its design and capabilities. Everything was going
great until we deployed our application to WAS 5.1 utilizing a JNDI
datasource.
We end up receiving the following exception:
[1/28/05 20:31:45:954 CST] 796d3dfd WebGroup E SRVE0026E: [Servlet
Error]-[LocalTransaction rolled-back due to setRollbackOnly]:
com.ibm.ws.LocalTransaction.RolledbackException
I've read the following post concerning issues with WebSphere:
http://wiki.apache.org/ibatis/Database_20Specific_20Information
But I'm afraid that we cannot implement the suggestions since we externally
acquire a connection and pass it to the SqlMapClient. Therefore, I'm not
sure how or if we can set a transactionManager element within our
configuration file since it requires a dataSource element as a child.
Currently, we do not have a transactionManager element at all.
Here is the code we use to pass the connection (sorry about the formatting
-- I'm using a web-based e-mail editor) :
/**
* Returns the cached code[EMAIL PROTECTED] SqlMapClient}/code instance
for the
DataSource alias.
*/
protected SqlMapClient getSqlMapClient(String dataSourceAlias)
throws SQLException {
try {
SqlMapClient sqlMapClient =
IBatisSqlMapService.getInstance().getSqlMapClient(
dataSourceAlias);
if (sqlMapClient.getCurrentConnection() == null) {
//try to acquire a Connection from the abstract broker
try {
Connection connection =
getDatabaseConnection(dataSourceAlias);
if (connection != null) {
sqlMapClient.setUserConnection(connection);
logger.debug(
Connection set via Common Services for \
+ dataSourceAlias
+ \.);
} else {
logger.debug(
Connection not configured via Common Services for \
+ dataSourceAlias
+ \. Assuming iBatis will supply connection.);
}
} catch (SQLException e) {
logger.debug(
Connection not configured via Common Services for \
+ dataSourceAlias
+ \. Assuming iBatis will supply connection.);
}
}
return sqlMapClient;
} catch (IOException e) {
throw new Error(e);
}
}
Also we clean up the application supplied user connection with the
following code:
/**
* Clear out the code[EMAIL PROTECTED] Connection}/code instanced stored
by this
thread's
* code[EMAIL PROTECTED] SqlMapClient}/code instance keyed by
codedataSourceName/code.
*/
public void closeConnection(String dataSourceName) throws SQLException {
try {
//clear out user connection on sql map
SqlMapClient sqlMapClient =
IBatisSqlMapService.getInstance().getSqlMapClient(dataSourceName);
if (sqlMapClient != null) {
//explicitly commit transaction prior to close
Connection connection = sqlMapClient.getCurrentConnection();
if (connection != null) {
//attempt to prevent exception -- didn't work
connection.commit();
}
sqlMapClient.setUserConnection(null);
logger.debug(
The transaction for SqlMapClient \
+ dataSourceName
+ \ has been cleared for calling thread.);
}
//close connection
super.closeConnection(dataSourceName);
} catch (IOException e) {
throw new Error(e.getMessage());
}
}
The IBatisSqlMapService is a singleton that caches SQLMapClients.
I would appreciate any advice about how I could prevent the exception from
occurring. Does iBATIS take part in transactions even if the connection is
supplied externally? I'm also aware that the issue could be with our
configuration of WebSphere, but I would like some ideas before I approach
our WebSphere administrator.
Thanks,
Mark
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
Re: Feedback
That's a SUPER idea Brandon. Because we often get requests for feedback from other users. Go for it bud! Clinton On Fri, 28 Jan 2005 08:31:01 -0700, Brandon Goodin [EMAIL PROTECTED] wrote: should we put these feedback on the wiki. It would be great to record successful experiences with IBatis. On Fri, 28 Jan 2005 10:50:22 +0100, Fabrizio Gianneschi [EMAIL PROTECTED] wrote: Hello all, I write this mail to share with you all my satisfaction in using iBATIS. The project I was working on until yesterday has been completed and delivered perfectly in time. Since it was a db-intensive web application (not much business logic, a lot of queries, a not usual db configuration) I consider iBATIS as an important factor that contributed to our success. Its use permitted us to be better focused on other application aspects, with simple XML changes and without recompiling the Java code. THIS, is the way J2EE should take. Thanks a lot, Clinton. Thanks to all of the people that work on this great project. Fabrizio Gianneschi Italy
