Keep Block handling in BlockwiseLayer #41
Assignees
Labels
Comments
sophokles73
pushed a commit
that referenced
this issue
May 20, 2016
The original fix for #36 was done in the Reliability layer, breaking separation of concerns between the layers of the stack. Instead of determining and notifying the original request about a timeout from within the Reliability layer, we now copy all MessageObservers of the original request to the individual blockwise requests in the Blockwise layer. This way, the Reliability layer can fall back to its original behavior of simply timing out the request at hand without checking if it is a blockwise request.
sophokles73
pushed a commit
that referenced
this issue
May 20, 2016
The original fix for #36 was done in the Reliability layer, breaking separation of concerns between the layers of the stack. Instead of determining and notifying the original request about a timeout from within the Reliability layer, we now copy all MessageObservers of the original request to the individual blockwise requests in the Blockwise layer. This way, the Reliability layer can fall back to its original behavior of simply timing out the request at hand without checking if it is a blockwise request. Conflicts: californium-core/src/main/java/org/eclipse/californium/core/network/stack/BlockwiseLayer.java californium-core/src/main/java/org/eclipse/californium/core/network/stack/ReliabilityLayer.java
|
Fixed in b266afd |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
#36 fixed a missing onError() call when an intermediary block times out.
To push a bit back on cross-layer concerns, I would like to solve this purely from the BlockwiseLayer, for instance by attaching a MessageObserver on the intermediary blocks that then calls the onError() of the original request.
This is more "nice to have" than urgent...
The text was updated successfully, but these errors were encountered: