Details

    • Type: Sub-task Sub-task
    • Status: Open
    • Priority: Major Major
    • Resolution: Unresolved
    • Affects Version/s: None
    • Fix Version/s: None
    • Component/s: None
    • Labels:
      None

      Description

      The remaining JMS specified identifiers are:
      amqp.correlation_id
      amqp.message_id
      amqp.creation_time
      amqp.jms_type

      The ones remaining from the AMQP Apache filters spec are:
      amqp.to
      amqp.reply_to
      amqp.absolute_expiration_time

      • amqp.durable (it's not clear this is really there as distinct from amqp.delivery_mode)
      • amqp.delivery_count (it's not clear this is really there as distinct from amqp.redelivered)

      In order to implement these special identifiers there will need to be extra work to abstract some of these values from the 0-10 or 1-0 message code.

        Activity

        Hide
        Gordon Sim added a comment -

        Re: creation_time property, as mentioned on the dev list I think this is the wrong solution. For 1.0 the creation-time should be used (its a matter of two lines or so to record it during parsing). On 0-10 I would use the timestamp as set on the delivery-properties of the received message (not the timestamp as set by the broker). That way over AMQP 1.0 the solution isboth compliant with the selector filter as specified and with JMS. On 0-10 behaviour will still be compliant with JMS without doing anything worse than is already being done as regards the 0-10 spec.

        Show
        Gordon Sim added a comment - Re: creation_time property, as mentioned on the dev list I think this is the wrong solution. For 1.0 the creation-time should be used (its a matter of two lines or so to record it during parsing). On 0-10 I would use the timestamp as set on the delivery-properties of the received message (not the timestamp as set by the broker). That way over AMQP 1.0 the solution isboth compliant with the selector filter as specified and with JMS. On 0-10 behaviour will still be compliant with JMS without doing anything worse than is already being done as regards the 0-10 spec.
        Hide
        ASF subversion and git services added a comment -

        Commit 1500613 from Gordon Sim
        [ https://svn.apache.org/r1500613 ]

        QPID-4627: fix faulty switch statement

        Show
        ASF subversion and git services added a comment - Commit 1500613 from Gordon Sim [ https://svn.apache.org/r1500613 ] QPID-4627 : fix faulty switch statement
        Hide
        ASF subversion and git services added a comment -

        Commit 1500052 from Andrew Stitcher
        [ https://svn.apache.org/r1500052 ]

        QPID-4627: Implement most of the remaining selector special identifiers
        Implemented:
        message_id, correlation_id,
        jms_type, creation_time, absolute_expiry_time

        There are a couple of caveats: The easily available way to get
        jms_type doesn't distinguish between an empty string and the
        property not being sent at all. So we treat this case as property
        not set as that seems like it will get most cases correct (why bother
        to send an empty jms_type?). The creation_time property is currently
        implemented as the time the message was put on the queue (if enabled
        in the broker) as amqp 0_10 has no standard way to indicate the
        creation time and we're not currently holding the creation time for amqp 1.0
        messages.

        Show
        ASF subversion and git services added a comment - Commit 1500052 from Andrew Stitcher [ https://svn.apache.org/r1500052 ] QPID-4627 : Implement most of the remaining selector special identifiers Implemented: message_id, correlation_id, jms_type, creation_time, absolute_expiry_time There are a couple of caveats: The easily available way to get jms_type doesn't distinguish between an empty string and the property not being sent at all. So we treat this case as property not set as that seems like it will get most cases correct (why bother to send an empty jms_type?). The creation_time property is currently implemented as the time the message was put on the queue (if enabled in the broker) as amqp 0_10 has no standard way to indicate the creation time and we're not currently holding the creation time for amqp 1.0 messages.
        Hide
        Gordon Sim added a comment -

        correlation-id and message-id are quite common in selectors, is this planned for 0.24?

        Show
        Gordon Sim added a comment - correlation-id and message-id are quite common in selectors, is this planned for 0.24?

          People

          • Assignee:
            Andrew Stitcher
            Reporter:
            Andrew Stitcher
          • Votes:
            0 Vote for this issue
            Watchers:
            4 Start watching this issue

            Dates

            • Created:
              Updated:

              Development