[JBoss Cache: Core Edition] - Re: on child insert parent, child collection updated in DB b
by dukehoops
OK, I wired up DualNodeJtaTransactionManagerImpl. The test now passes if UserCount is below 4 or less, at 5 or above the test fails. However, the failure I am seeing with this test is NOT the same as with the test in our environment. It is this:
anonymous wrote : -------------------------------------------------------------------------------
| Test set: org.hibernate.test.cache.jbc2.functional.MVCCConcurrentWriteTest
| -------------------------------------------------------------------------------
| Tests run: 6, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 182.113 sec <<< FAILURE!
| testManyUsers(org.hibernate.test.cache.jbc2.functional.MVCCConcurrentWriteTest) Time elapsed: 165.831 sec <<< FAILURE!
| junit.framework.AssertionFailedError: Timed out waiting for user threads to finish. Their state at the time of forced shutdown: TEST CONFIG [userCount=5, iterationsPerUser=40, thinkTimeMillis=10] STATE of UserRunners: org.hibernate.test.cache.jbc2.functional.MVCCConcurrentWriteTest$UserRunner@ffc3eb[customerId=4 iterationsCompleted=30 completedAll=false causeOfFailure=org.hibernate.cache.CacheException: org.hibernate.cache.CacheException: org.jboss.cache.lock.TimeoutException: Unable to acquire lock on Fqn [/TS/test/org/hibernate/cache/UpdateTimestampsCache/Contacts] after [15000] milliseconds for requestor [Thread[UserRunnerThread-4,5,main]]! Lock held by [GlobalTransaction::903]
| at org.hibernate.cache.jbc2.timestamp.TimestampsRegionImpl.put(TimestampsRegionImpl.java:130)
| at org.hibernate.cache.UpdateTimestampsCache.preinvalidate(UpdateTimestampsCache.java:70)
| at org.hibernate.engine.ActionQueue.execute(ActionQueue.java:275)
| at org.hibernate.engine.ActionQueue.executeActions(ActionQueue.java:263)
| at org.hibernate.engine.ActionQueue.executeActions(ActionQueue.java:167)
| at org.hibernate.event.def.AbstractFlushingEventListener.performExecutions(AbstractFlushingEventListener.java:321)
| at org.hibernate.event.def.DefaultFlushEventListener.onFlush(DefaultFlushEventListener.java:50)
| at org.hibernate.impl.SessionImpl.flush(SessionImpl.java:1027)
| at org.hibernate.impl.SessionImpl.managedFlush(SessionImpl.java:365)
| at org.hibernate.transaction.CacheSynchronization.beforeCompletion(CacheSynchronization.java:88)
| at org.hibernate.test.cache.jbc2.functional.util.DualNodeJtaTransactionImpl.commit(DualNodeJtaTransactionImpl.java:76)
| at org.hibernate.test.cache.jbc2.functional.util.DualNodeJtaTransactionManagerImpl.commit(DualNodeJtaTransactionManagerImpl.java:123)
| at org.hibernate.test.cache.jbc2.functional.MVCCConcurrentWriteTest.commitTx(MVCCConcurrentWriteTest.java:255)
| at org.hibernate.test.cache.jbc2.functional.MVCCConcurrentWriteTest.addContact(MVCCConcurrentWriteTest.java:333)
| at org.hibernate.test.cache.jbc2.functional.MVCCConcurrentWriteTest.access$100(MVCCConcurrentWriteTest.java:29)
| at org.hibernate.test.cache.jbc2.functional.MVCCConcurrentWriteTest$UserRunner.run(MVCCConcurrentWriteTest.java:423)
| at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
| at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
| at java.lang.Thread.run(Thread.java:619)
| Caused by: org.hibernate.cache.CacheException: org.jboss.cache.lock.TimeoutException: Unable to acquire lock on Fqn [/TS/test/org/hibernate/cache/UpdateTimestampsCache/Contacts] after [15000] milliseconds for requestor [Thread[UserRunnerThread-4,5,main]]! Lock held by [GlobalTransaction::903]
| at org.hibernate.cache.jbc2.util.CacheHelper.put(CacheHelper.java:214)
| at org.hibernate.cache.jbc2.timestamp.TimestampsRegionImpl.put(TimestampsRegionImpl.java:128)
| ... 18 more]
| org.hibernate.test.cache.jbc2.functional.MVCCConcurrentWriteTest$UserRunner@18bbb61[customerId=6 iterationsCompleted=28 completedAll=false causeOfFailure=]
| org.hibernate.test.cache.jbc2.functional.MVCCConcurrentWriteTest$UserRunner@1cab18[customerId=2 iterationsCompleted=28 completedAll=false causeOfFailure=]
| org.hibernate.test.cache.jbc2.functional.MVCCConcurrentWriteTest$UserRunner@9cd8db[customerId=3 iterationsCompleted=30 completedAll=false causeOfFailure=]
|
|
To repro:
-set userCount to 5 or more
-run the test and wait patiently for up to 5 minutes (not sure why, as with < 5 User threads this test finished in 3 seconds).
The fact that this test isn't using a real JTA tx manager is a problem IMO - as the above exception seems to do more with TX cleanup (though I'd be happy to be proven wrong) . Even if the above bug is worked around and the test starts passing, it won't prove there isn't a problem since a crucial piece of the stack is replaced with a simpler impl.
Here's the updated source:
package org.hibernate.test.cache.jbc2.functional;
|
| import java.util.HashSet;
| import java.util.Random;
| import java.util.Set;
| import java.util.concurrent.ExecutorService;
| import java.util.concurrent.Executors;
| import java.util.concurrent.TimeUnit;
| import javax.transaction.SystemException;
| import junit.framework.Test;
| import org.hibernate.FlushMode;
| import org.hibernate.Session;
| import org.hibernate.cfg.Configuration;
| import org.hibernate.exception.ExceptionUtils;
| import org.hibernate.junit.functional.FunctionalTestClassTestSuite;
| import org.hibernate.stat.SecondLevelCacheStatistics;
| import org.hibernate.test.cache.jbc2.functional.util.DualNodeConnectionProviderImpl;
| import org.hibernate.test.cache.jbc2.functional.util.DualNodeJtaTransactionManagerImpl;
| import org.hibernate.test.cache.jbc2.functional.util.DualNodeTestUtil;
| import org.hibernate.test.cache.jbc2.functional.util.DualNodeTransactionManagerLookup;
| import org.hibernate.transaction.CMTTransactionFactory;
| import org.slf4j.Logger;
| import org.slf4j.LoggerFactory;
|
| /**
| *
| * @author nikita_tovstoles(a)mba.berkeley.edu
| */
| public class MVCCConcurrentWriteTest extends MVCCJBossCacheTest {
|
| private static final Logger LOG = LoggerFactory.getLogger(MVCCConcurrentWriteTest.class);
| /**
| * when USER_COUNT==1, tests pass, when >4 tests fail
| */
| final private int USER_COUNT = 5;
| final private int ITERATION_COUNT = 40;
| final private int THINK_TIME_MILLIS = 10;
| final private long LAUNCH_INTERVAL_MILLIS = 10;
| final private Random random = new Random();
| /**
| * kill switch used to stop all users when one fails
| */
| private static boolean TERMINATE_ALL_USERS = false;
| /**
| * collection of IDs of all customers participating in this test
| */
| private Set<Integer> customerIDs = new HashSet<Integer>();
|
| public MVCCConcurrentWriteTest(String x) {
| super(x);
| }
|
| /**
| * test that DB can be queried
| * @throws java.lang.Exception
| */
| public void testPingDb() throws Exception {
| try {
| beginTx();
| getEnvironment().getSessionFactory().getCurrentSession().createQuery("from " + Customer.class.getName()).list();
| commitTx();
| } catch (Exception e) {
| rollbackTx();
| fail("failed to query DB; exception=" + e);
| }
| }
|
| @Override
| protected void prepareTest() throws Exception {
| super.prepareTest();
| }
|
| @Override
| protected void cleanupTest() throws Exception {
| try {
| super.cleanupTest();
|
| } finally {
| cleanup();
| //DualNodeJtaTransactionManagerImpl.cleanupTransactions();
| //DualNodeJtaTransactionManagerImpl.cleanupTransactionManagers();
| }
| }
|
| @Override
| public void configure(Configuration cfg) {
| super.configure(cfg);
| cfg.setProperty(DualNodeTestUtil.NODE_ID_PROP, DualNodeTestUtil.LOCAL);
| }
|
| @Override
| protected Class getConnectionProviderClass() {
| return DualNodeConnectionProviderImpl.class;
| }
|
| @Override
| protected Class getTransactionManagerLookupClass() {
| return DualNodeTransactionManagerLookup.class;
| }
|
| @Override
| protected Class getTransactionFactoryClass() {
| return CMTTransactionFactory.class;
| }
|
| @Override
| public void testEmptySecondLevelCacheEntry() throws Exception {
| //do nothing
| }
|
| @Override
| public void testQueryCacheInvalidation() {
| //do nothing
| }
|
| @Override
| public void testStaleWritesLeaveCacheConsistent() {
| //do nothing
| }
|
| public void testSingleUser() throws Exception {
| //setup
| Customer customer = createCustomer(0);
| final Integer customerId = customer.getId();
| getCustomerIDs().add(customerId);
|
| assertNull("contact exists despite not being added", getFirstContact(customerId));
|
| //check that cache was hit
| SecondLevelCacheStatistics customerSlcs = getEnvironment().getSessionFactory().getStatistics().getSecondLevelCacheStatistics(
| getPrefixedRegionName(Customer.class.getName()));
| assertEquals(customerSlcs.getPutCount(), 1);
| assertEquals(customerSlcs.getElementCountInMemory(), 1);
| assertEquals(customerSlcs.getEntries().size(), 1);
|
| SecondLevelCacheStatistics contactsCollectionSlcs = getEnvironment().getSessionFactory().getStatistics().getSecondLevelCacheStatistics(
| getPrefixedRegionName(Customer.class.getName() + ".contacts"));
| assertEquals(1, contactsCollectionSlcs.getPutCount());
| assertEquals(1, contactsCollectionSlcs.getElementCountInMemory());
| assertEquals(1, contactsCollectionSlcs.getEntries().size());
|
| final Contact contact = addContact(customerId);
| assertNotNull("contact returned by addContact is null", contact);
| assertEquals("Customer.contacts cache was not invalidated after addContact",
| 0, contactsCollectionSlcs.getElementCountInMemory());
|
| assertNotNull("Contact missing after successful add call", getFirstContact(customerId));
|
|
| //read everyone's contacts
| readEveryonesFirstContact();
|
| removeContact(customerId);
| assertNull("contact still exists after successful remove call", getFirstContact(customerId));
|
| }
|
| public void testManyUsers() throws Exception {
|
| //setup - create users
| for (int i = 0; i < USER_COUNT; i++) {
| Customer customer = createCustomer(0);
| getCustomerIDs().add(customer.getId());
| }
|
| assertEquals("failed to create enough Customers", USER_COUNT, getCustomerIDs().size());
|
| final ExecutorService pool = Executors.newFixedThreadPool(USER_COUNT);
|
| Set<UserRunner> runners = new HashSet<UserRunner>();
| for (Integer customerId : getCustomerIDs()) {
| UserRunner r = new UserRunner(customerId);
| runners.add(r);
| pool.execute(r);
| LOG.info("launched " + r);
| Thread.sleep(LAUNCH_INTERVAL_MILLIS); //rampup
| }
|
| assertEquals("not all user threads launched", USER_COUNT, runners.size());
|
| pool.shutdown();
| boolean finishedInTime = pool.awaitTermination(120, TimeUnit.SECONDS);
|
| if (!finishedInTime) { //timed out waiting for users to finish
| fail("Timed out waiting for user threads to finish. Their state at the time of forced shutdown: " + statusOfRunnersToString(runners));
| } else {
| //if here -> pool finished before timing out
| //check whether all runners suceeded
| boolean success = true;
| for (UserRunner r : runners) {
| if (!r.isSuccess()) {
| success = false;
| break;
| }
| }
| assertTrue("at least one UserRunner failed: " + statusOfRunnersToString(runners), success);
| }
| }
|
| public void cleanup() throws Exception {
|
| getCustomerIDs().clear();
|
| String deleteContactHQL = "delete from Contact";
| String deleteCustomerHQL = "delete from Customer";
|
| beginTx();
| try {
|
| //Session session = getSessions().getCurrentSession();
| Session session = getEnvironment().getSessionFactory().getCurrentSession();
| session.createQuery(deleteContactHQL).setFlushMode(FlushMode.AUTO).executeUpdate();
| session.createQuery(deleteCustomerHQL).setFlushMode(FlushMode.AUTO).executeUpdate();
| commitTx();
| } catch (Exception e) {
| rollbackTx();
| throw e;
| }
|
| }
|
| private Customer createCustomer(int nameSuffix) throws Exception {
| beginTx();
| try {
| Customer customer = new Customer();
| customer.setName("customer_" + nameSuffix);
| customer.setContacts(new HashSet<Contact>());
|
| getEnvironment().getSessionFactory().getCurrentSession().persist(customer);
| commitTx();
| return customer;
| } catch (Exception e) {
| rollbackTx();
| throw e;
| }
| }
|
| /**
| * delegate method since I'm trying to figure out which txManager to use
| * given that this test runs multiple threads (SimpleJtaTxMgrImpl isn't suited for that).
| *
| * What is needed is a thread-safe JTATransactionManager impl that can handle concurrent TXs
| *
| * @throws java.lang.Exception
| */
| private void beginTx() throws Exception {
| DualNodeJtaTransactionManagerImpl.getInstance(DualNodeTestUtil.LOCAL).begin();
| }
|
| /**
| * @see #beginTx()
| * @throws java.lang.Exception
| */
| private void commitTx() throws Exception {
| DualNodeJtaTransactionManagerImpl.getInstance(DualNodeTestUtil.LOCAL).commit();
| }
|
| /**
| * @see #beginTx()
| * @throws java.lang.Exception
| */
| private void rollbackTx() throws Exception {
| DualNodeJtaTransactionManagerImpl.getInstance(DualNodeTestUtil.LOCAL).rollback();
| }
|
| /**
| * read first contact of every Customer participating in this test.
| * this forces concurrent cache writes of Customer.contacts Collection cache node
| *
| * @return who cares
| * @throws java.lang.Exception
| */
| private void readEveryonesFirstContact() throws Exception {
| beginTx();
| try {
| for (Integer customerId : getCustomerIDs()) {
| final Customer customer = (Customer) getEnvironment().getSessionFactory().getCurrentSession().load(Customer.class, customerId);
| Set<Contact> contacts = customer.getContacts();
| Contact firstContact = contacts.isEmpty() ? null : contacts.iterator().next();
| }
| commitTx();
| } catch (Exception e) {
| rollbackTx();
| throw e;
| }
| }
|
| /**
| * -load existing Customer
| * -get customer's contacts; return 1st one
| *
| * @param customerId
| * @return first Contact or null if customer has none
| */
| private Contact getFirstContact(Integer customerId) throws Exception {
| assert customerId != null;
|
| beginTx();
| try {
| final Customer customer = (Customer) getEnvironment().getSessionFactory().getCurrentSession().load(Customer.class, customerId);
| Set<Contact> contacts = customer.getContacts();
| Contact firstContact = contacts.isEmpty() ? null : contacts.iterator().next();
| commitTx();
| return firstContact;
| } catch (Exception e) {
| rollbackTx();
| throw e;
| }
| }
|
| /**
| * -load existing Customer
| * -create a new Contact and add to customer's contacts
| *
| * @param customerId
| * @return added Contact
| */
| private Contact addContact(Integer customerId) throws Exception {
| assert customerId != null;
|
| beginTx();
| try {
| final Customer customer = (Customer) getEnvironment().getSessionFactory().getCurrentSession().load(Customer.class, customerId);
|
| Contact contact = new Contact();
| contact.setName("contact name");
| contact.setTlf("wtf is tlf?");
|
| contact.setCustomer(customer);
| customer.getContacts().add(contact);
|
| //assuming contact is persisted via cascade from customer
| commitTx();
|
| return contact;
| } catch (Exception e) {
| rollbackTx();
| throw e;
| }
| }
|
| /**
| * remove existing 'contact' from customer's list of contacts
| *
| * @param contact contact to remove from customer's contacts
| * @param customerId
| * @throws IllegalStateException if customer does not own a contact
| */
| private void removeContact(Integer customerId) throws Exception {
| assert customerId != null;
|
| beginTx();
| try {
| Customer customer = (Customer) getEnvironment().getSessionFactory().getCurrentSession().load(Customer.class, customerId);
| Set<Contact> contacts = customer.getContacts();
| if (contacts.size() != 1) {
| throw new IllegalStateException("can't remove contact: customer id=" + customerId + " expected exactly 1 contact, " +
| "actual count=" + contacts.size());
| }
|
| Contact contact = contacts.iterator().next();
| contacts.remove(contact);
| contact.setCustomer(null);
|
| //explicitly delete Contact because hbm has no 'DELETE_ORPHAN' cascade?
| //getEnvironment().getSessionFactory().getCurrentSession().delete(contact); //appears to not be needed
|
| //assuming contact is persisted via cascade from customer
| commitTx();
| } catch (Exception e) {
| rollbackTx();
| throw e;
| }
| }
|
| /**
| * @return the customerIDs
| */
| public Set<Integer> getCustomerIDs() {
| return customerIDs;
| }
|
| private String statusOfRunnersToString(Set<UserRunner> runners) {
| assert runners != null;
|
| StringBuilder sb = new StringBuilder("TEST CONFIG [userCount=" + USER_COUNT +
| ", iterationsPerUser=" + ITERATION_COUNT +
| ", thinkTimeMillis=" + THINK_TIME_MILLIS + "] " +
| " STATE of UserRunners: ");
|
| for (UserRunner r : runners) {
| sb.append(r.toString() + System.getProperty("line.separator"));
| }
| return sb.toString();
| }
|
| class UserRunner implements Runnable {
|
| final private Integer customerId;
| private int completedIterations = 0;
| private Throwable causeOfFailure;
|
| public UserRunner(final Integer cId) {
| assert cId != null;
| this.customerId = cId;
| }
|
| private boolean contactExists() throws Exception {
| return getFirstContact(customerId) != null;
| }
|
| public void run() {
|
| //name this thread for easier log tracing
| Thread.currentThread().setName("UserRunnerThread-" + getCustomerId());
| try {
| for (int i = 0; i < ITERATION_COUNT; i++) {
|
| if (contactExists()) {
| throw new IllegalStateException("contact already exists before add, customerId=" + customerId);
| }
|
| addContact(customerId);
|
| if (!contactExists()) {
| throw new IllegalStateException("contact missing after successful add, customerId=" + customerId);
| }
|
| thinkRandomTime();
|
| //read everyone's contacts
| readEveryonesFirstContact();
|
| thinkRandomTime();
|
| removeContact(customerId);
|
| if (contactExists()) {
| throw new IllegalStateException("contact still exists after successful remove call, customerId=" + customerId);
| }
|
| thinkRandomTime();
|
| ++completedIterations;
| }
|
| } catch (Throwable t) {
|
| this.causeOfFailure = t;
| TERMINATE_ALL_USERS = true;
|
| //rollback current transaction if any
| //really should not happen since above methods all follow begin-commit-rollback pattern
| try {
| if (DualNodeJtaTransactionManagerImpl.getInstance(DualNodeTestUtil.LOCAL).getTransaction() != null) {
| DualNodeJtaTransactionManagerImpl.getInstance(DualNodeTestUtil.LOCAL).rollback();
| }
| } catch (SystemException ex) {
| throw new RuntimeException("failed to rollback tx", ex);
| }
| }
| }
|
| public boolean isSuccess() {
| return ITERATION_COUNT == getCompletedIterations();
| }
|
| public int getCompletedIterations() {
| return completedIterations;
| }
|
| public Throwable getCauseOfFailure() {
| return causeOfFailure;
| }
|
| public Integer getCustomerId() {
| return customerId;
| }
|
| @Override
| public String toString() {
| return super.toString() +
| "[customerId=" + getCustomerId() +
| " iterationsCompleted=" + getCompletedIterations() +
| " completedAll=" + isSuccess() +
| " causeOfFailure=" + (this.causeOfFailure != null ? ExceptionUtils.getStackTrace(causeOfFailure) : "") + "] ";
| }
| }
|
| /**
| * sleep between 0 and THINK_TIME_MILLIS.
| * @throws RuntimeException if sleep is interruped or TERMINATE_ALL_USERS flag was set to true in the meantime
| */
| private void thinkRandomTime() {
| try {
| Thread.sleep(random.nextInt(THINK_TIME_MILLIS));
| } catch (InterruptedException ex) {
| throw new RuntimeException("sleep interrupted", ex);
| }
|
| if (TERMINATE_ALL_USERS) {
| throw new RuntimeException("told to terminate (because a UserRunner had failed)");
| }
| }
|
| public static Test suite() {
| return new FunctionalTestClassTestSuite(MVCCConcurrentWriteTest.class);
| }
| }
|
View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4215956#4215956
Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4215956
17 years, 1 month
[JBoss Portal] - Accessing JBossPortletRequest/HTTPServletRequest in Portal 2
by tschaible
Hello everyone,
We have a use case where we need a portlet to present an iframe to an existing home-grown PHP application and do SSO between the two applications. We're on a short timeline, so changes the existing PHP app to support some more robust form of SSO are not an option at the moment.
I have been able to accomplish this outside the portal context by populating the PHP session db tables in the background and setting an appropriate cookie. (apps are on different sub-domains of the same domain, so this should be fine)
Inside the portal however, I cannot access the HTTPServletRequest/Response to grab IP information and to set cookies.
I have seen several posts recommendeding that i grab the JBossPortletRequest and do the following
jbRequest.getControllerContext().getServerInvocation().getServerContext().getClientRequest()
This seems to work fine in Portal 2.6.x, but on the 2.7.x releases, the request comes back as an org.jboss.portal.portlet.impl.jsr168.api.RenderRequestImpl which cannot be cast to a JBossPortletRequest.
Understandably the 168 spec does not allow access to the base HTTP Request/Response, but is there a way in Portal 2.7.x to still access these objects outside of the 168 spec? It was available in 2.6 and would be a nice to have in 2.7 with the understanding that any portlets using those specific api calls would not be portable.
Thanks,
--Tom
View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4215940#4215940
Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4215940
17 years, 1 month
[JBoss Tools (users)] - Re: Hibernate reverse engineering errors
by max.andersen＠jboss.com
"bdlink" wrote :
| With respect to some of my questions that were unclear:
| 1. JoinTables can map 1-1 and 1-n as well as n-n. I was not clear on how to reverse engineer this situation.
|
It is not supported at this time.
anonymous wrote :
| 2. I can use the same primary key in both tables to make a 1-1 relation. This wants a @PrimaryKeyJoinColumn(s) annotation instead of @JoinColumn. I was not clear how to reverse engineer this situation.
|
Isn't this not what is happening for you now ? (It should if I remember correctly)
anonymous wrote :
| I understand that it would be a lot of work to add everything! I had understood from the original Hibernate books years ago that the philosophy was that you could use three strategies with Hibernate:
| 1. Annotate the Java
| 2. write an hbm.xml
| 3. reverse engineer the db metadata
| choosing which ever one suited the situation (and whichever one you write, the other two will be generated). Since the database metadata does not contain enough information to generate the other two, you need to help that case with reveng.xml.
|
That is true, but reveng has limitations - but naturally and others because it was simply not requested or implemented yet. (I give commit access to anyone contributing features 90% properly done and with testcases - hint hint ;)
anonymous wrote :
| I understood the ultimate goal would be that there are equal capabilities with all three strategies. If not, we should say that one of these is the best practice, but the others can be useful at times (since the world seems to be moving away from xml, I guess that the Java annotations would have to be primary?)
|
That is correct - I don't believe we ever said otherwise.
anonymous wrote :
| Java Persistence with Hibernate, Bauer and King, indicate that the above three strategies are feasible (and recommend 2 only for the seasoned Hibernate expert). Also Dan Allen's Seam in Action book gives the three strategies.
|
I will need to check but I believe they write that reverse engineering has limitations and should be used as a "first-shot"/inspiration.
anonymous wrote :
| So I would not be put out of reveng.xml became as big as hbm.xml :)
|
Sounds like you just volunteered ?
anonymous wrote :
| Further information on the error (this is H11 update installed) By the way, where do I look to find the hibernate-reverse-engineering-3.0.dtd?:
|
in hibernate-tools.jar
View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4215924#4215924
Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4215924
17 years, 1 month
[JBoss Tools (users)] - Re: Hibernate reverse engineering errors
by max.andersen＠jboss.com
"bdlink" wrote : OK, I edited the reveng.xml file by hand, eliminating all of the empty string attribures, and that eliminates the exception (cheers!)
|
So two bugs found: the editor is leaving empty quotes (bad style) and the reveng.xml parser chokes on empty quotes (which it shouldn't).
Did you say you saw jira's for this already ? (need to create them if they aren't there)
anonymous wrote :
| I looked at my DB to see why extra associations were showing up and found duplicate foreign key constraints that mysql added at some time, and when I removed these, I get the associations I would expect.
|
Good - i'm not insane then.
anonymous wrote :
| So things are much improved! There are some remaining issues I am looking at including :
| 1. Even though I annotate a foreign key with one-to-one, the property is marked with @ManyToOne and the inverse side uses a Set (when it is not excluded).
|
hmmm - the one-to-one detection is rather new; looks like it requires the underlying db to actually have a one-to-one foreign key (i.e. it only listens to reveng.xml about one-to-one attributes not its definition plus the one-to-one between pk and a non pk is not "allowed").
Needs a bug report. Unfortunately I can't fix this for GA.
anonymous wrote :
| 2. I have one case where there are two associations between Student and Review (both one Student to many Reviews). In the ReviewHome class it injects two studentHome components (duplicate names), where one of them should be a different @Role. (I think this is in JIRA from long ago, but I will check later)
|
This is in context of Seam gen, rigth ? This is something to report against seam-gen in Seam - nothing hibernatetools specific about this afaik. If you got a jira ref let me know ;)
anonymous wrote :
| The one-to-one issue is obviously more important.
|
I can't do it for GA but will look into it straight after.
Until then you can try and implement the oneToone related methods in reverse engineering strategy to think there is a one-to-one.
View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4215923#4215923
Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4215923
17 years, 1 month