Hi David... I have attached a patch with my changes http://issues.apache.org/jira/browse/OPENEJB-438 . Meanwhile I will play with the subject of annotated test clients.
On 2/13/07, David Blevins <[EMAIL PROTECTED]> wrote:
On Feb 12, 2007, at 11:08 AM, Mohammad Nour El-Din wrote: > Hi Dave... > > On 1/31/07, Mohammad Nour El-Din <[EMAIL PROTECTED]> wrote: >> >> Hi David... >> >> On 1/31/07, David Blevins <[EMAIL PROTECTED]> wrote: >> > >> > >> > On Jan 30, 2007, at 2:01 AM, Mohammad Nour El-Din wrote: >> > >> > > Hi David... >> > > >> > > On 1/30/07, David Blevins < [EMAIL PROTECTED]> wrote: >> > >> >> > >> Hi Mohammad, >> > >> >> > >> On Jan 28, 2007, at 7:21 AM, Mohammad Nour El-Din wrote: >> > >> >> > >> > I am facing a new situation here, while tests run during >> build - as >> > >> > >> > I wanted >> > >> > to test the impl of OPENEJB-438 - all tests related to >> > >> RemoteiTest are >> > >> > causing errors because of not being able to find the JNDI >> name " >> > >> > client/tools/DatabaseHome", this has never happened to me >> before - >> > >> > except >> > >> > while trying to run tests from within eclipse but not while >> > >> building >> > >> >> > >> The "client/tools/DatabaseHome" is just the first bean that the >> > >> itests try to lookup. The error is really that the app >> doesn't exist >> > >> in the system and wasn't successfully deployed. My guess is >> one of >> > >> your changes for OPENEJB-438 broke something in deployment. >> > >> >> > >> Definitely read the maven output for openejb log messages and >> > >> definitely check out the target/surefire dir for any output >> that came >> > >> out of the test. Those are the sources I primarily use when >> I break >> > >> things (which is all the time). When that doesn't work I run >> the >> > >> tests in my IDE with debug. And for those rare occasions >> something >> > >> works in my IDE but not via maven, I have maven run the tests >> with >> > >> remote debug turned on and connect via my IDE. >> > >> >> > >> If you get frustrated after some time debugging I'll be happy >> to give >> > >> your changes a try. >> > >> >> > >> -David >> > > >> > > >> > > Thanks for your feedback, I already use the sources you >> mentioned, >> > > and I >> > > already started debugging, but I was hoping to have an answer >> before >> > > starting :) >> > >> > :) I only wish I was that talented :) >> >> >> Sure you are :) >> >> > , anyway it seems that I broke the code with my latest changes >> > > and I am fixing it :), thanks for your feedback again. >> > >> > No problem. Offer still open if you get stumped! >> > >> > -David >> >> >> Well, the whole problem was related to >> AnnotatedFieldInjectionStateless\StatefulBean. First the stateful >> bean was >> assigned the EncStatelessObject as its remote interface. And I >> should add >> the new methods for looking-up the extended Env-Entries to the >> stateless >> bean. I have a question but I will dig for and when I fail to get >> the answer >> myself I will ask you talented man ;) >> > > Well I think I need some help here, I've been debugging this for a > while, > but seems I have problems representing newlines in the ejb-jar.xml > file, the > following env-entry is not parsed right by the PropertiesEditor > > <env-entry> > > <env-entry-name>stateless/references/Properties</env- > entry-name> > > <env-entry-type>java.util.Properties</env-entry-type> > > <env-entry-value>key1=value1\r\nkey2=value2</env-entry- > value> > > </env-entry> > > the editor creates a Properties object instance but with one entry, > having > *key1* as the key and *value1\r\nkey2=value2* as the value, instead of > making two entries for both keys and values, I tried everything but > no use, > I need help here. I'd be happy to look at it. Attach a patch to the issue and I'll apply it and take a look. -David > > > >> > > >> > > >> > > >> > > >> > > -- >> > > Thanks >> > > - Mohammad Nour >> > >> > >> >> >> -- >> Thanks >> - Mohammad Nour > > > > > -- > Thanks > - Mohammad Nour
-- Thanks - Mohammad Nour
