locked
EventData Properties in Stream Analytics RRS feed

  • Question

  • When Input for Stream Analytics job is set to EventHub Data Stream, only the message payload is accessible for processing. The other fields of the EventData object like - Properties Dictionary, PartitionKey, SequenceNumber, Offset etc are not available. Is there any possibilt that these information are being exposed to stream analytics?
    Monday, February 8, 2016 8:45 AM

Answers

  • Currently only the message payload is accessible, not the properties which have been set on the EventData. If you need to access these you will have to create your own processor using the Event Processor Host, or add them to your data model and fill them upon creating the EventData object where possible.
    Monday, February 8, 2016 9:08 AM

All replies

  • Currently only the message payload is accessible, not the properties which have been set on the EventData. If you need to access these you will have to create your own processor using the Event Processor Host, or add them to your data model and fill them upon creating the EventData object where possible.
    Monday, February 8, 2016 9:08 AM
  • Hi Arun

    Currently they're not accessible, but we are considering to enable some scenarios in this direction in near future that may help you.

    cheers !


    [Disclaimer] This posting is provided "AS IS" with no warranties, and confers no rights. User assumes all.

    Thursday, February 11, 2016 8:04 PM
  • create your own processor using the Event Processor Host, or add them to your data model and fill them upon creating the EventData object where possible.

    do you have an example / sample that shows how to do this?

    Friday, April 15, 2016 6:53 PM
  • Any update on accessing AMQP proprieties of messages when ingested using Streaming Analytics ? We are using IoT AMQP and Streaming Analytics?

    Thanks,

    Mike

     
    Thursday, December 1, 2016 10:16 PM
  • You can use some metadata properties from EH messages using GetMetadataPropertyValue function 


    [Disclaimer] This posting is provided "AS IS" with no warranties, and confers no rights. User assumes all.

    Saturday, March 3, 2018 3:44 PM