You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@camel.apache.org by "Hadrian Zbarcea (JIRA)" <ji...@apache.org> on 2012/06/04 21:20:22 UTC

[jira] [Commented] (CAMEL-4857) Endpoint URI normalization: information in path is lost

    [ https://issues.apache.org/jira/browse/CAMEL-4857?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13288831#comment-13288831 ] 

Hadrian Zbarcea commented on CAMEL-4857:
----------------------------------------

I beg to differ.

The question is not if Camel is too restrictive, but rather if camel uses URIs or not. What a URI is and isn't is clearly defined by a spec. The changes for CAMEL-4256 were intended to still support the invalid uris we now use in Camel and provide a migration path for 3.0. If some scenarios were missed, that can be fixed.

However the question still remains, does Camel use URIs or not? Today, I wouldn't know how to define the concept of URI flexibility. I totally agree that that component writers should have complete control on how to handle URIs but those should be URIs to start with. That's not the case today, is it?
                
> Endpoint URI normalization: information in path is lost
> -------------------------------------------------------
>
>                 Key: CAMEL-4857
>                 URL: https://issues.apache.org/jira/browse/CAMEL-4857
>             Project: Camel
>          Issue Type: Bug
>          Components: camel-core
>    Affects Versions: 2.9.0
>            Reporter: Alexander Azarov
>            Assignee: Claus Ibsen
>             Fix For: Future
>
>
> A project with a test case is here: https://github.com/alaz/camel290_uri
> The test case with the problem description (the problem occured during migration of camel-beanstalk https://github.com/osinka/camel-beanstalk component onto 2.9.0):
> {code:title=UriTest.java}
> package camel290.uri;
> import java.util.Map;
> import org.apache.camel.Consumer;
> import org.apache.camel.Endpoint;
> import org.apache.camel.Processor;
> import org.apache.camel.Producer;
> import org.apache.camel.impl.DefaultComponent;
> import org.apache.camel.impl.DefaultEndpoint;
> import org.apache.camel.test.CamelTestSupport;
> import org.junit.Before;
> import org.junit.Test;
> import static org.junit.Assert.*;
> public class UriTest extends CamelTestSupport {
>   /**
>    * An URI of Camel Beanstalk component consists of a hostname, port and a list
>    * of tube names. Tube names are separated by "+" character (which is more or less
>    * usualy used on the Web to make lists), but every tube name may contain URI special
>    * characters like ? or +
>    */
>   class MyEndpoint extends DefaultEndpoint {
>     String uri = null;
>     String remaining = null;
>     public MyEndpoint(final String uri, final String remaining) {
>       this.uri = uri;
>       this.remaining = remaining;
>     }
>     public Producer createProducer() throws Exception {
>       throw new UnsupportedOperationException("Not supported yet.");
>     }
>     public Consumer createConsumer(Processor prcsr) throws Exception {
>       throw new UnsupportedOperationException("Not supported yet.");
>     }
>     public boolean isSingleton() {
>       return true;
>     }
>   }
>   class MyComponent extends DefaultComponent {
>     @Override
>     protected Endpoint createEndpoint(final String uri, final String remaining, final Map<String, Object> parameters) throws Exception {
>       return new MyEndpoint(uri, remaining);
>     }
>   }
>   @Before
>   @Override
>   public void setUp() throws Exception {
>     super.setUp();
>     context.addComponent("my", new MyComponent());
>   }
>   @Test
>   public void testExclamationInUri() {
>     /**
>      * %3F is not an ?, it's part of tube name.
>      */
>     MyEndpoint endpoint = context.getEndpoint("my:host:11303/tube1+tube%2B+tube%3F", MyEndpoint.class);
>     assertNotNull("endpoint", endpoint);
>   }
>   @Test
>   public void testPath() {
>     /**
>      * Here a tube name is "tube+" and written in URI as "tube%2B", but it gets
>      * normalized, so that an endpoint sees "tube1+tube+"
>      */
>     MyEndpoint endpoint = context.getEndpoint("my:host:11303/tube1+tube%2B", MyEndpoint.class);
>     assertEquals("Path contains several tube names, every tube name may have + or ? characters", "host:11303/tube1+tube%2B", endpoint.remaining);
>   }
> }
> {code} 

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira