[
https://issues.apache.org/jira/browse/ARTEMIS-3369?focusedWorklogId=618425&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-618425
]
ASF GitHub Bot logged work on ARTEMIS-3369:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 04/Jul/21 12:11
Start Date: 04/Jul/21 12:11
Worklog Time Spent: 10m
Work Description: erwindon commented on pull request #3638:
URL: https://github.com/apache/activemq-artemis/pull/3638#issuecomment-873577096
@jbertram I'll add the regression tests
On your additional thoughts:
> The anycastPrefix and multicastPrefix are useful mainly for protocols that
have no semantic distinctions for their destinations (e.g. STOMP). The CLI uses
JMS which does have semantic distinctions for its destinations. That is why it
has the built-in queue:// and topic:// prefixes for determining semantics.
Therefore, I'm not sure it makes a lot of sense to use custom prefixes on the
CLI when built-in prefixes already exist to do the same work.
I was setting up some basic tests using the `artemis producer` and `artemis
consumer` to show how a client application can use the prefixes to get what it
wants. In this case these are low-level amqp clients that do not (all) have the
opportunity to set a capabilities value "queue" or "topic"; thus they rely on
using prefixes.
> The broker will strip any configured prefix. So, for example, if you had
anycastPrefix=foo:// and you used foo://myQueue from a client then the broker
would strip foo:// from the name and just use myQueue with anycast semantics.
In your example, if you actually had anycastPrefix=xyz:// the second screenshot
would look exactly like the first.
I used "xyz://" specifically in the above tests to have a prefix that was
_not_ recognized by the broker. That way I could easily verify whether the
prefix reaches the server, because it would show up in the gui. In reality of
course the true anycastPrefix value or multicastPrefix value would be used.
Sorry for the confusion. I did not to want to use addresses/queues that really
start with anything that looks like a prefix :-)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 618425)
Time Spent: 2h 40m (was: 2.5h)
> CLI removes server-side prefixes from a given address
> -----------------------------------------------------
>
> Key: ARTEMIS-3369
> URL: https://issues.apache.org/jira/browse/ARTEMIS-3369
> Project: ActiveMQ Artemis
> Issue Type: Bug
> Affects Versions: 2.17.0
> Reporter: Erwin Dondorp
> Priority: Major
> Time Spent: 2h 40m
> Remaining Estimate: 0h
>
> The Artemis CLI variations {{producer}}, {{consumer}}, {{browse}} and
> {{transfer}}, all prepare a destination using function {{getDestination}}.
> However, this function also removes a prefix that it was not looking for. It
> tests for a topic prefix ({{topic://}}) and then removes any prefix that was
> present.
> It should explicitly test for the 2 prefixes it cares about ({{topic://}} and
> {{queue://}}) and then remove only a recognized prefix when it was present.
> PR is present.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
