INT-3661: Fix the eager BF access from BPPs (P I) #1393
Conversation
|
Also fixes https://jira.spring.io/browse/INT-3662 |
There was a problem hiding this comment.
This class has a bug in the afterPropertiesSet() will be fixed tomorrow
|
Pushed |
JIRA: https://jira.spring.io/browse/INT-3661 ##Status Quo Previously there were a lot of noise from the `PostProcessorRegistrationDelegate$BeanPostProcessorChecker` for early access for beans. That may produce some side-effects when some of `BeanFactoryPostProcessor`s won't adjust those beans. The issue is based on two facts: 1. Loading beans from `BPP`, e.g. `IntegrationEvaluationContextAwareBeanPostProcessor` (or `ChannelSecurityInterceptorBeanPostProcessor` - https://jira.spring.io/browse/INT-3663) 2. Loading beans from `setBeanFactory()/setApplicationContext()` container methods ## The fix * Move all stuff from `setBeanFactory()` with access to the `BeanFactory` (e.g. `this.messageBuilderFactory = IntegrationUtils.getMessageBuilderFactory(this.beanFactory);`) to some other lazy-load methods like `getMessageBuilderFactory()` * Fix parser tests to `assertNotSame` for `messageBuilderFactory` since there is no activity for target components to lazy-load the stuff * Polish some test according the new lazy-load logic * Rework `IntegrationEvaluationContextAwareBeanPostProcessor` to the `SmartInitializingSingleton` and make it `Ordered` * Populate `beanFactory` for the internal instance of `connectionFactory` in the `TcpSyslogReceivingChannelAdapter` * Populate `beanFactory` for the internal `UnicastReceivingChannelAdapter` in the `UdpSyslogReceivingChannelAdapter` * Add `log.info` that `UdpSyslogReceivingChannelAdapter` overrides `outputChannel` for the provided `UnicastReceivingChannelAdapter` * Change the internal `MessageChannel` in the `UdpSyslogReceivingChannelAdapter` to the `FixedSubscriberChannel` for better performance
* Add JavaDocs for the `IntegrationEvaluationContextAware`
|
Pushed fix for |
There was a problem hiding this comment.
This makes no sense to me; the goal was to use the global MBF for all message builds.
There was a problem hiding this comment.
Right. I understand your original goal. But I wanted to pay your attention here that we now is in the lazy-load, hence these tests really don't make sense because there is no anymore any interaction.
Are you OK if I remove these assertNotSame with the next commit/polishing to this PR ?
There was a problem hiding this comment.
Oh, I see; you're not saying that at runtime we're using a different builder just that, since these are parser only tests, we won' have it yet.
I agree; they should just be removed.
However, we might want to consider adding verification to runtime tests instead 😄
(That can be a separate PR if you prefer)
There was a problem hiding this comment.
Yes, I also intend to remove them.
Re. separate JIRA: no need, really. See PublishSubscribeChannelParserTests changes. I hope that should prove that the fix works as promised.
|
Wow - a lot of repetitive, thankless work (but I thank you !!). |
|
Nothing more from me. |
|
Pushed |
|
Merged as 2bde14b |
2 participants
JIRA: https://jira.spring.io/browse/INT-3661
Status Quo
Previously there were a lot of noise from the
PostProcessorRegistrationDelegate$BeanPostProcessorCheckerfor early access for beans.That may produce some side-effects when some of
BeanFactoryPostProcessors won't adjust those beans.The issue is based on two facts:
BPP, e.g.IntegrationEvaluationContextAwareBeanPostProcessor(orChannelSecurityInterceptorBeanPostProcessor- https://jira.spring.io/browse/INT-3663)setBeanFactory()/setApplicationContext()container methodsThe fix
setBeanFactory()with access to theBeanFactory(e.g.this.messageBuilderFactory = IntegrationUtils.getMessageBuilderFactory(this.beanFactory);)to some other lazy-load methods like
getMessageBuilderFactory()assertNotSameformessageBuilderFactorysince there is no activity for target components to lazy-load the stuffIntegrationEvaluationContextAwareBeanPostProcessorto theSmartInitializingSingletonand make itOrderedbeanFactoryfor the internal instance ofconnectionFactoryin theTcpSyslogReceivingChannelAdapterbeanFactoryfor the internalUnicastReceivingChannelAdapterin theUdpSyslogReceivingChannelAdapterlog.infothatUdpSyslogReceivingChannelAdapteroverridesoutputChannelfor the providedUnicastReceivingChannelAdapterMessageChannelin theUdpSyslogReceivingChannelAdapterto theFixedSubscriberChannelfor better performance