thoughts on message queuee message
new interface. What if there are MQ messages with different sizes? -> generic interface furthermore, maybe command message should be refactored to operate on a mq message instead of implementing it
This commit is contained in:
@ -5,12 +5,14 @@
|
||||
#include <limits>
|
||||
|
||||
union sid_t {
|
||||
static constexpr uint64_t INVALID_ADDRESS = std::numeric_limits<uint64_t>::max();
|
||||
static constexpr uint64_t INVALID_ADDRESS =
|
||||
std::numeric_limits<uint64_t>::max();
|
||||
sid_t(): raw(INVALID_ADDRESS) {}
|
||||
|
||||
struct {
|
||||
object_id_t objectId ;
|
||||
// A generic 32 bit ID to identify unique HK packets for a single object.
|
||||
// A generic 32 bit ID to identify unique HK packets for a single
|
||||
// object.
|
||||
// For example, the DeviceCommandId_t is used for DeviceHandlers
|
||||
uint32_t ownerSetId;
|
||||
};
|
||||
@ -19,15 +21,17 @@ union sid_t {
|
||||
*/
|
||||
uint64_t raw;
|
||||
};
|
||||
class HousekeepingMessage {
|
||||
|
||||
|
||||
class HousekeepingMessage: public MessageQueueMessage {
|
||||
public:
|
||||
/**
|
||||
* No instances of a message shall be created, instead
|
||||
* a CommandMessage instance is manipulated.
|
||||
*/
|
||||
HousekeepingMessage() = delete;
|
||||
HousekeepingMessage(const HousekeepingMessage&) = delete;
|
||||
HousekeepingMessage operator=(const HousekeepingMessage &) = delete;
|
||||
HousekeepingMessage();
|
||||
// HousekeepingMessage(const HousekeepingMessage&) = delete;
|
||||
// HousekeepingMessage operator=(const HousekeepingMessage &) = delete;
|
||||
|
||||
static constexpr uint8_t MESSAGE_ID = MESSAGE_TYPE::HOUSEKEEPING;
|
||||
static constexpr Command_t ADD_HK_REPORT_STRUCT =
|
||||
@ -76,6 +80,7 @@ public:
|
||||
|
||||
// static void setAddHkReportStructMessage(CommandMessage* message,
|
||||
// DevisetId, store_address_t packet);
|
||||
static void setHkReportMessage();
|
||||
};
|
||||
|
||||
|
||||
|
Reference in New Issue
Block a user