open-zwave/config/everspring/st814.xml

58 lines
2.4 KiB
XML
Raw Permalink Normal View History

<?xml version="1.0" encoding="utf-8"?>
<Product xmlns='http://code.google.com/p/open-zwave/'>
2017-11-07 08:32:45 +01:00
<!-- https://products.z-wavealliance.org/products/271 -->
<!-- Configuration Parameters -->
<CommandClass id="112">
<Value type="byte" index="1" genre="config" label="Basic Set Level" units="%" min="0" max="99" value="99">
<Help>
Defines the level in the Basic Set event sent to group 2 when the sensor is triggered. Default is 99 (full brightness for a Z-Wave dimmer). 0 disables.
</Help>
</Value>
<Value type="byte" index="2" genre="config" label="Temperature On Trigger" units="degrees" min="-20" max="99" value="30">
<Help>
Temperature level to trigger when a ON command is sent out. 99 will clear the value.
</Help>
</Value>
<Value type="byte" index="3" genre="config" label="Temperature Off Trigger" units="degrees" min="-20" max="99" value="20">
<Help>
Temperature level to trigger when an OFF command is sent out. 99 will clear the value.
</Help>
</Value>
<Value type="byte" index="4" genre="config" label="Humidity On Trigger" units="%" min="20" max="99" value="50">
<Help>
Humidity level to trigger when a ON command is sent out. 99 will clear the value.
</Help>
</Value>
<Value type="byte" index="5" genre="config" label="Humidity Off Trigger" units="%" min="20" max="99" value="40">
<Help>
Humidity level to trigger when a OFF command is sent out. 99 will clear the value.
</Help>
</Value>
<Value type="short" index="6" genre="config" label="Auto Report Time" units="minutes" min="0" max="1439" value="0">
<Help>
Sets the auto report time interval. 0 disables.
</Help>
</Value>
<Value type="byte" index="7" genre="config" label="Auto Report Temperature" units="degrees" min="0" max="70" value="0">
<Help>
Sets the auto report temperature trigger interval. 0 disables.
</Help>
</Value>
2017-11-07 08:32:45 +01:00
<Value type="byte" index="8" genre="config" label="Auto Report Humidity" units="%" min="0" max="70" value="0">
<Help>
Sets the auto report humidity trigger interval. 0 disables.
</Help>
</Value>
</CommandClass>
Put back the classgetsupported xml option under the Version class as all the Leviton devices appear to not support the class get operation. Add Everspring ST815 device. Add a configuration file for the Leviton VRCPG handheld controller. Manager doesn't use m_exitEvent so it has been removed. Add Manager::GetSUCNodeId method. Add Manager::HealNetworkNode and Manager::HealNetwork method. These routines will send a request node neighbor update to a specific node or every node in the network. They take an optional boolean argument to also refresh the return routes as well. Improve UserCode class a bit. For devices that support 255 user codes, we only handle 10 for now. Still need a better way to discover lots of data. Address some issues in the WakeUp class. Put controller requests on the WakeUp message queue in an effort to speed them up. Send a wake up notification to user programs. Handle some multiple step controller commands on the wake up queue. Fix bug in the create_vars xml configuration. For classes that already created the variables, create_vars=false will cause them to be removed. Also output create_vars state correctly. Add an interface to the PowerLevel class. This gives a UI to nodes that support it. Try to improve the ControllerReplication class. Still a work in progress. Not sure how useful it is. Give the NoOperation its own message queue that has higher priority then normal message traffic. This let's message diagnositics run before normal messages. Address some issues in the SwitchMultiLevel class. Only the Level can be queried. Other internal values weren't being set properly since they weren't conveyed from the temporary values to the permanent values. Configuration class now supports button values. Big update is the change in how controller commands are handled. Redefined a number of controller commands. Now there is only AddDevice and RemoveDevice. The controller add and remove were redundant. Controller commands can now be queued and will be executed from a seperate message queue just above commands. Thus controller commands have a higher priority them normal message traffic. They can also be stacked. This is a big improvement over the global variables we were using before. Some controller commands can be queued onto the wake up queue for sleeping devices. This area will still need some work. Added error returns through the callback function so initial start up of the controller commands can return some useful information. Added some new controller states to pass back to the callback routines. Two new controller commands were added: SendNodeInformation and ReplicationSend. Confirmed most of the controller commands are working properly. Changed the name of the round trip counters in the node object. The current round trip times measure the controller's initial request. The name changes reflect the difference between the initial request and the not yet measured response (if there is one). Discovery will query controllers for Node Information as it only tries once. Initialize controller with a node info message so other devices can get it. May need to revisit what is in this message. Remove left over code that checked for controller commands as it is replaced by the message queues. Have MoveMessagesToWakeUpQueue mark nodes as sleeping when no ack is received for other message types, such as controller command responses. No longer exclude handheld controllers from no ack (sleeping) message queuing. Centralize no ack and other transmission error check into a single routine. Add more message types to Driver::IsExpectedReply. Especially ZW_SEND_DATA. Add support for request node neighbor update with options. Save the API version as some older controllers handle add/remove/stop device updates differently. After transferring primary, a driver reset notification is sent but it is deferred until reinitializing the controller since the controller state may have changed. Fix remove failed node not updating node list. Remove the race condition created by sending the finished controller callback from the RequestNodeNeighbors routine. Since RequestNodeNeighbors was called from discovery it could try invoking the global callback in the prior code or the outstanding current controller command in this newer code. The consequence may be that the callback may happen prior to the RequestNodeNeighbors response arrives and older routing tables are still in place. No longer call replication when adding nodes. Still need to figure out where replication is needed. Bridge button mapping code is untested with the new controller command interface. Have the Delete and Assign return route code use the new controller command interface. No retransmit on failures in current code.
2012-12-23 06:01:42 +00:00
<!-- COMMAND_CLASS_ASSOCIATION. Groups -->
<CommandClass id="133">
<Associations num_groups="2">
<Group index="1" max_associations="1" label="Reports" />
<Group index="2" max_associations="3" label="Basic" />
Put back the classgetsupported xml option under the Version class as all the Leviton devices appear to not support the class get operation. Add Everspring ST815 device. Add a configuration file for the Leviton VRCPG handheld controller. Manager doesn't use m_exitEvent so it has been removed. Add Manager::GetSUCNodeId method. Add Manager::HealNetworkNode and Manager::HealNetwork method. These routines will send a request node neighbor update to a specific node or every node in the network. They take an optional boolean argument to also refresh the return routes as well. Improve UserCode class a bit. For devices that support 255 user codes, we only handle 10 for now. Still need a better way to discover lots of data. Address some issues in the WakeUp class. Put controller requests on the WakeUp message queue in an effort to speed them up. Send a wake up notification to user programs. Handle some multiple step controller commands on the wake up queue. Fix bug in the create_vars xml configuration. For classes that already created the variables, create_vars=false will cause them to be removed. Also output create_vars state correctly. Add an interface to the PowerLevel class. This gives a UI to nodes that support it. Try to improve the ControllerReplication class. Still a work in progress. Not sure how useful it is. Give the NoOperation its own message queue that has higher priority then normal message traffic. This let's message diagnositics run before normal messages. Address some issues in the SwitchMultiLevel class. Only the Level can be queried. Other internal values weren't being set properly since they weren't conveyed from the temporary values to the permanent values. Configuration class now supports button values. Big update is the change in how controller commands are handled. Redefined a number of controller commands. Now there is only AddDevice and RemoveDevice. The controller add and remove were redundant. Controller commands can now be queued and will be executed from a seperate message queue just above commands. Thus controller commands have a higher priority them normal message traffic. They can also be stacked. This is a big improvement over the global variables we were using before. Some controller commands can be queued onto the wake up queue for sleeping devices. This area will still need some work. Added error returns through the callback function so initial start up of the controller commands can return some useful information. Added some new controller states to pass back to the callback routines. Two new controller commands were added: SendNodeInformation and ReplicationSend. Confirmed most of the controller commands are working properly. Changed the name of the round trip counters in the node object. The current round trip times measure the controller's initial request. The name changes reflect the difference between the initial request and the not yet measured response (if there is one). Discovery will query controllers for Node Information as it only tries once. Initialize controller with a node info message so other devices can get it. May need to revisit what is in this message. Remove left over code that checked for controller commands as it is replaced by the message queues. Have MoveMessagesToWakeUpQueue mark nodes as sleeping when no ack is received for other message types, such as controller command responses. No longer exclude handheld controllers from no ack (sleeping) message queuing. Centralize no ack and other transmission error check into a single routine. Add more message types to Driver::IsExpectedReply. Especially ZW_SEND_DATA. Add support for request node neighbor update with options. Save the API version as some older controllers handle add/remove/stop device updates differently. After transferring primary, a driver reset notification is sent but it is deferred until reinitializing the controller since the controller state may have changed. Fix remove failed node not updating node list. Remove the race condition created by sending the finished controller callback from the RequestNodeNeighbors routine. Since RequestNodeNeighbors was called from discovery it could try invoking the global callback in the prior code or the outstanding current controller command in this newer code. The consequence may be that the callback may happen prior to the RequestNodeNeighbors response arrives and older routing tables are still in place. No longer call replication when adding nodes. Still need to figure out where replication is needed. Bridge button mapping code is untested with the new controller command interface. Have the Delete and Assign return route code use the new controller command interface. No retransmit on failures in current code.
2012-12-23 06:01:42 +00:00
</Associations>
</CommandClass>
<CommandClass id="96" mapping="endpoints" />
</Product>