• Type: Enhancement
• Resolution: Obsolete
• Priority: Major
• Fix Version/s: None
• Affects Version/s: 9.1.1.Final
• Component/s: Expiration
• Labels:

  None

[ISPN-8331] CacheEntryExpiredEvent.getValue() is always null when entry expired from cache store

Collapse comment: Tristan Tarrant added a comment - 2024/11/27 3:02 PM

Tristan Tarrant added a comment - 2024/11/27 3:02 PM

Infinispan issue tracking has been migrated to GitHub issues: https://github.com/infinispan/infinispan/issues
If you still want this issue to be worked on, create a new issue on GitHub and link this issue.

Expand comment: Tristan Tarrant added a comment - 2024/11/27 3:02 PM

Tristan Tarrant added a comment - 2024/11/27 3:02 PM Infinispan issue tracking has been migrated to GitHub issues: https://github.com/infinispan/infinispan/issues If you still want this issue to be worked on, create a new issue on GitHub and link this issue.

Collapse comment: Radoslav Husar added a comment - 2018/03/09 8:35 AM

Radoslav Husar added a comment - 2018/03/09 8:35 AM

The API for this is already implemented by wburns@redhat.com with @Experimental org.infinispan.persistence.spi.AdvancedCacheExpirationWriter.ExpirationPurgeListener#marshalledEntryPurged. The problem of course is that none of the stores currently implement this.

If we do, we may want to provide a way to optionally pull the value such as a `Supplier` to get the value.

Why would this be necessary? The implenters would return implementation of the org.infinispan.marshall.core.MarshalledEntry which would proxy the getValueBytes() and getValue() method calls.

Expand comment: Radoslav Husar added a comment - 2018/03/09 8:35 AM

Radoslav Husar added a comment - 2018/03/09 8:35 AM The API for this is already implemented by wburns@redhat.com with @Experimental org.infinispan.persistence.spi.AdvancedCacheExpirationWriter.ExpirationPurgeListener#marshalledEntryPurged . The problem of course is that none of the stores currently implement this. If we do, we may want to provide a way to optionally pull the value such as a `Supplier` to get the value. Why would this be necessary? The implenters would return implementation of the org.infinispan.marshall.core.MarshalledEntry which would proxy the getValueBytes() and getValue() method calls.

Collapse comment: Paul Ferraro added a comment - 2017/09/26 6:51 AM

Paul Ferraro added a comment - 2017/09/26 6:51 AM

wburns@redhat.com Why not PurgeListener.entryPurged(MarshalledEntry entry)?
That way both values and meta data are exposed, yet unmarshalled lazily.

Expand comment: Paul Ferraro added a comment - 2017/09/26 6:51 AM

Paul Ferraro added a comment - 2017/09/26 6:51 AM wburns@redhat.com Why not PurgeListener.entryPurged(MarshalledEntry entry)? That way both values and meta data are exposed, yet unmarshalled lazily.

Collapse comment: Will Burns added a comment - 2017/09/25 3:51 PM

Will Burns added a comment - 2017/09/25 3:51 PM

Unfortunately this is the way the loader spi was implemented. I would love to change this to support values as well. If we do, we may want to provide a way to optionally pull the value such as a `Supplier` to get the value. I am not sure if we have implementations that it may be very costly to pull the value or not.

Also cc/ remerson@redhat.com since he has been dealing with loaders recently.

Expand comment: Will Burns added a comment - 2017/09/25 3:51 PM

Will Burns added a comment - 2017/09/25 3:51 PM Unfortunately this is the way the loader spi was implemented. I would love to change this to support values as well. If we do, we may want to provide a way to optionally pull the value such as a `Supplier` to get the value. I am not sure if we have implementations that it may be very costly to pull the value or not. Also cc/ remerson@redhat.com since he has been dealing with loaders recently.