[
https://issues.apache.org/jira/browse/ARTEMIS-4532?focusedWorklogId=901417&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-901417
]
ASF GitHub Bot logged work on ARTEMIS-4532:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 24/Jan/24 11:42
Start Date: 24/Jan/24 11:42
Worklog Time Spent: 10m
Work Description: gemmellr commented on code in PR #4710:
URL: https://github.com/apache/activemq-artemis/pull/4710#discussion_r1464750566
##########
tests/integration-tests/src/test/java/org/apache/activemq/artemis/tests/integration/mqtt5/spec/controlpackets/PublishTests.java:
##########
@@ -653,6 +654,10 @@ public void messageArrived(String topic, MqttMessage
message) throws Exception {
consumer.close();
}
+ private Queue getRetainedMessageQueue(String TOPIC) {
+ return
server.locateQueue(MQTTUtil.convertMqttTopicFilterToCore(MQTTUtil.MQTT_RETAIN_ADDRESS_PREFIX,
TOPIC, server.getConfiguration().getWildcardConfiguration()));
+ }
Review Comment:
The MQTTSubscriptionManager.getQueueNameForTopic method (which delegates to
this util method now) seems to have been the source of multiple breaks, or
changes in behaviour, over the last several releases. Obviously bugs happen,
but it feels like some of these would have been caught by comprehensive unit
testing of the method. The broker calls this method to name its queues, but the
system tests also call the method to get the queue name the broker used. If for
example the method returns "wrong-value" as it effectively has been doing, then
broker names things in that way, and the system tests also just use that as the
expected value, and so in those cases theyll always match and wont notice if
anything is wrong, e.g I go in there later and mess something up with the
naming.
If on the other hand there is a direct unit testing of the method
comprehensively checking the output, ensuring that the exact value expected is
returned, then the behaviour of that method is then more directly tested (and a
few ms should tell if it is later broken...vs running minutes of system tests
that mostly wont notice because they use the same method to establish what is
expected), and likely more understandable because there are now specific tests
for it showing what its expected to do, and quickly pointing out if changed
inappropriately.
Issue Time Tracking
-------------------
Worklog Id: (was: 901417)
Time Spent: 2h 20m (was: 2h 10m)
> MQTT-to-core wildcard conversion is broken
> ------------------------------------------
>
> Key: ARTEMIS-4532
> URL: https://issues.apache.org/jira/browse/ARTEMIS-4532
> Project: ActiveMQ Artemis
> Issue Type: Bug
> Reporter: Justin Bertram
> Assignee: Justin Bertram
> Priority: Major
> Time Spent: 2h 20m
> Remaining Estimate: 0h
>
> Currently when an MQTT topic filter contains characters from the configured
> [wildcard
> syntax|https://activemq.apache.org/components/artemis/documentation/latest/wildcard-syntax.html#wildcard-syntax]
> the conversion to/from this syntax breaks.
> For example, when using the default wildcard syntax if an MQTT topic filter
> contains a {{.}} the conversion from the MQTT wildcard syntax to the core
> wildcard syntax and back will result in the {{.}} being replaced with a {{/}}.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
