Add/remove/update comments and TODOs in OmnipodPumpPlugin and rename a method
This commit is contained in:
parent
afd890a519
commit
aede5648a9
|
@ -113,7 +113,6 @@ public class OmnipodPumpPlugin extends PumpPluginBase implements PumpInterface,
|
|||
private final PumpType pumpType = PumpType.Insulet_Omnipod;
|
||||
|
||||
private final List<CustomAction> customActions = new ArrayList<>();
|
||||
// TODO: BS: Not really sure what this is all about, have a closer look at it some time
|
||||
private final List<OmnipodStatusRequest> omnipodStatusRequestList = new ArrayList<>();
|
||||
private final CompositeDisposable disposables = new CompositeDisposable();
|
||||
|
||||
|
@ -127,7 +126,8 @@ public class OmnipodPumpPlugin extends PumpPluginBase implements PumpInterface,
|
|||
private int timeChangeRetries;
|
||||
private long nextPodCheck;
|
||||
private boolean sentIdToFirebase;
|
||||
private long lastConnectionTimeMillis;
|
||||
private long lastConnectionTimeMillis;// 3 minutes
|
||||
private static final long RILEY_LINK_CONNECT_TIMEOUT = 3 * 60 * 1000L;
|
||||
|
||||
@Inject
|
||||
public OmnipodPumpPlugin(
|
||||
|
@ -387,12 +387,15 @@ public class OmnipodPumpPlugin extends PumpPluginBase implements PumpInterface,
|
|||
lastConnectionTimeMillis = System.currentTimeMillis();
|
||||
}
|
||||
|
||||
// TODO seems that this abused to squeeze commands in the queue
|
||||
// Look for an alternative
|
||||
// We abuse getPumpStatus to squeeze commands in the queue
|
||||
// The only actual status requests we send to the Pod are on startup (in initializeAfterRileyLinkConnection)
|
||||
// And when the user explicitly requested it by clicking the Refresh button on the Omnipod tab
|
||||
// We don't do periodical status requests because that can drain the Pod's battery
|
||||
// However that should be fine because we get a StatusResponse from all insulin commands sent to the Pod
|
||||
@Override
|
||||
public void getPumpStatus() {
|
||||
if (firstRun) {
|
||||
initializePump();
|
||||
initializeAfterRileyLinkConnection();
|
||||
} else if (!omnipodStatusRequestList.isEmpty()) {
|
||||
Iterator<OmnipodStatusRequest> iterator = omnipodStatusRequestList.iterator();
|
||||
|
||||
|
@ -741,19 +744,21 @@ public class OmnipodPumpPlugin extends PumpPluginBase implements PumpInterface,
|
|||
|
||||
@Override
|
||||
public boolean isUnreachableAlertTimeoutExceeded(long unreachableTimeoutMilliseconds) {
|
||||
long rileyLinkInitializationTimeout = 3 * 60 * 1000L; // 3 minutes
|
||||
// We have a separate notification for when no Pod is active, see doPodCheck()
|
||||
if (podStateManager.isPodActivationCompleted() && podStateManager.getLastSuccessfulCommunication() != null) { // Null check for backwards compatibility
|
||||
if (podStateManager.getLastSuccessfulCommunication().getMillis() + unreachableTimeoutMilliseconds < System.currentTimeMillis()) {
|
||||
// TODO update comment
|
||||
// We exceeded the alert threshold, and either our last command failed or we cannot reach the RL
|
||||
// We should show an alert
|
||||
// Don't trigger an alert when we exceeded the thresholds, but the last communication was successful & the RL is reachable
|
||||
// This happens when we simply didn't need to send any commands to the pump
|
||||
return !podStateManager.isPodRunning() ||
|
||||
(podStateManager.getLastFailedCommunication() != null && podStateManager.getLastSuccessfulCommunication().isBefore(podStateManager.getLastFailedCommunication())) ||
|
||||
long currentTimeMillis = System.currentTimeMillis();
|
||||
|
||||
if (podStateManager.getLastSuccessfulCommunication().getMillis() + unreachableTimeoutMilliseconds < currentTimeMillis) {
|
||||
// We exceeded the user defined alert threshold. However, as we don't send periodical status requests to the Pod to prevent draining it's battery,
|
||||
// Exceeding the threshold alone is not a reason to trigger an alert: it could very well be that we just didn't need to send any commands for a while
|
||||
// Below return statement covers these cases in which we will trigger an alert:
|
||||
// - Sending the last command to the Pod failed
|
||||
// - RileyLink is in an error state
|
||||
// - RileyLink has been connecting for over RILEY_LINK_CONNECT_TIMEOUT
|
||||
return (podStateManager.getLastFailedCommunication() != null && podStateManager.getLastSuccessfulCommunication().isBefore(podStateManager.getLastFailedCommunication())) ||
|
||||
rileyLinkServiceData.rileyLinkServiceState.isError() ||
|
||||
// The below clause is a hack for working around the RL service state forever staying in connecting state on startup if the RL is switched off / unreachable
|
||||
(rileyLinkServiceData.getRileyLinkServiceState().isConnecting() && rileyLinkServiceData.getLastServiceStateChange() + rileyLinkInitializationTimeout < System.currentTimeMillis());
|
||||
(rileyLinkServiceData.getRileyLinkServiceState().isConnecting() && rileyLinkServiceData.getLastServiceStateChange() + RILEY_LINK_CONNECT_TIMEOUT < currentTimeMillis);
|
||||
}
|
||||
}
|
||||
|
||||
|
@ -816,7 +821,7 @@ public class OmnipodPumpPlugin extends PumpPluginBase implements PumpInterface,
|
|||
return getOperationNotSupportedWithCustomText(info.nightscout.androidaps.core.R.string.pump_operation_not_supported_by_pump_driver);
|
||||
}
|
||||
|
||||
private void initializePump() {
|
||||
private void initializeAfterRileyLinkConnection() {
|
||||
if (podStateManager.isPodInitialized() && podStateManager.getPodProgressStatus().isAtLeast(PodProgressStatus.PAIRING_COMPLETED)) {
|
||||
PumpEnactResult result = executeCommand(OmnipodCommandType.GetPodStatus, aapsOmnipodManager::getPodStatus);
|
||||
if (result.success) {
|
||||
|
|
Loading…
Reference in a new issue