My question might be really simple but i will appreciate any help, am using java and firebase and i have a variable in the database called count == 0, i need to increment this value every day by 1, so after 30 days the variable count should be count == 30, the only trouble is making that increment programmaticaly.
Any suggestions will be highly appreciated.
A scheduled Cloud Function deployed with the Firebase CLI will be the proper way to update your data in firebase. Your can run a scheduled function which can be triggered at a specific time (everyday or week however you want) and you can update your variable in the database inside the function. Here is a sample from firebase official documentation :
exports.scheduledFunctionCrontab = functions.pubsub.schedule('5 11 * * *')
.timeZone('America/New_York') // Users can choose timezone - default is America/Los_Angeles
.onRun((context) => {
console.log('This will be run every day at 11:05 AM Eastern!');
return null;
});
Read documentation here : https://firebase.google.com/docs/functions/schedule-functions#write_a_scheduled_function
Related
I am writing an android app in java which uses Android phone's Step Counter sensor event values to record walking steps taken by the user.
Basically my java code to display current Steps is as follows:
if (running) {
totalSteps = event.values[0];
currentSteps = totalSteps - previousTotalSteps;
tv_stepsTaken.setText(String.valueOf(currentSteps));
}
I am using SharedPreferences to load, save and reset previousTotalSteps on a daily basis.
My problem is that when phone reboots, the totalSteps event value from Step Counter Sensor is reset automatically to zero, hence it turns my currentSteps value to negative as previousTotalSteps value is then being subtracted from zero. Only solution that is coming to my mind is to set previousTotalSteps to zero too. But it will defeat the purpose of recording daily steps. If the user reboots the phone in the middle of the day, then half day's data will be lost or otherwise turn negative in display. I am also recording this daily steps value to sqlite database as daily history to be shown graphically to user. So, this negative value goes to sqlite databse too and ruins the historical graph. I need suggestions to solve this problem. My reset data code is as follows:
previousTotalSteps = totalSteps;
tv_stepsTaken.setText("0");
saveStepsData();
return true;
So, I have solved this problem by myself. I am posting this solution for anyone else who find themselves in this situation, so that it might help them solve it.
For this, currentSteps as well as previousSteps need to be saved to SharedPreferences after each step. Earlier only previousSteps was being saved and that too once daily.
Thus, totalSteps-previousSteps will give an increment of one step to be added to currentSteps.
In case, the phone reboots, totalSteps value is reset to zero. If that happens(or by any chance, if totalSteps is less than previousSteps), then the previousSteps value will also be set equal to totalSteps for that condition before applying both variables for current steps calculation.
if (running) {
totalSteps = event.values[0];
loadCurrentStepsData();
loadPreviousStepsData();
if (totalSteps == 0 || totalSteps<previousSteps) {
previousSteps = totalSteps;
savePreviousStepsData();
}
currentSteps = currentSteps + (totalSteps - previousSteps);
previousSteps = totalSteps;
saveCurrentStepsData();
savePreviousStepsData();
}
tv_stepsTaken.setText(String.valueOf(currentSteps));
I have a date supplied by the user and of course today's date.
I'm attempting to verify that the difference between the 2 days is at least 2 weeks. I've done this using standard libraries - but I'm attempting to do this using jodaTime and I'm having some difficulty.
// BAD CODE - doesn't work
// NOTE: getUserSuppliedDate() returns an instance of java.sql.Date
// Also assume that validation prior to this call has been run that
// validates that the userSuppliedDate comes AFTER today's date - not sure if
// that is relevant in the context I'm attempting to use these particular jodaTime APIs.
DateTime jodaStartDate = new DateTime(getUserSuppliedDate());
if (Days.daysBetween(jodaStartDate, DateTime.now()).isLessThan(Days.days(14))) {
System.out.println("Bad user. You have chosen...poorly.");
}
else {
System.out.println("Well done user. You have supplied wisely.");
}
// GOOD CODE ---- ? Help =)
Your code gives you the wrong result because the dates supplied to Days.daysBetween() are in the wrong order. Since you specified that the user supplied date comes after the current date, your approach will result in a negative number of days.
It will work correctly if you switch the order, putting the earliest date first.
Compare the following two:
DateTime jodaStartDate = new DateTime().withYear(2018)
.withMonthOfYear(7)
.withDayOfMonth(5); // 15 days from now
System.out.println(Days.daysBetween(jodaStartDate, DateTime.now())); // "P-15D"
System.out.println(Days.daysBetween(DateTime.now(), jodaStartDate)); // "P15D"
In the first case, -15 days will evaluate to less than 14 days.
Using weeks instead of days, you'd run into the same problem:
System.out.println(Weeks.weeksBetween(jodaStartDate, DateTime.now())); // "P-2W"
System.out.println(Weeks.weeksBetween(DateTime.now(), jodaStartDate)); // "P2W"
I have a step counter app in which I am running a service which counts steps taken and then sends a broadcast to the fragment which is then updated on the fragment. The step counting is working fine but I want to calculate speed based on the steps. Here is what I am trying right now.
The receiver to get step count:
receiver = new BroadcastReceiver() {
#Override
public void onReceive(Context context, Intent intent) {
int steps = intent.getIntExtra(StepCounterService.STEP_INCREMENT_KEY, 0);
if (firstStepTime.equals("0")) {
firstStepTime = intent.getStringExtra(StepCounterService.TIME_STAMP_KEY);
} else if (secondStepTime.equals("0")) {
secondStepTime = intent.getStringExtra(StepCounterService.TIME_STAMP_KEY);
} else {
firstStepTime = secondStepTime;
secondStepTime = intent.getStringExtra(StepCounterService.TIME_STAMP_KEY);
}
updateAllUI(steps);
}
};
So what I am doing is as soon as I start getting steps, I see if the variable firstStepTime is empty. If it is, I save the time in firstStepTime variable. The in the next step I see if secondStepTime is empty, and if it is, I save that time in secondStepTime variable.
Now for the next steps both these are updated.
public void updateAllUI(int numberOfSteps) {
if (!(firstStepTime.equals("0")) && !(secondStepTime.equals("0"))) {
try {
Calendar c = Calendar.getInstance();
SimpleDateFormat timeFormat = new SimpleDateFormat("HH:mm:ss.SSS");
timeDifference = timeFormat.parse(secondStepTime).getTime() - timeFormat.parse(firstStepTime).getTime();
speed = (float) ((0.0254 / (timeDifference * 0.001)) * 3.6);
} catch (Exception e) {
timeDifference = 0;
speed = 0;
}
textview.settext(speed +"Km/h);
}
}
So in this I just check if both are not empty, I take the values and calculate the difference in times. The problem here is sometimes it doesn't count speed properly. And a bigger problem is if the user stops, the speed remains constant and doesn't drop to zero.
Is there any better way to do the speed calculation?
As you're only counting steps and would like to calculate speed with steps taken only i am assuming that you would not like to use GPS for counting speed or you would like to provide backup speed counter if GPS is not there.
You can start a new thread by doing
int interval=1*1000; //specifying interval to check steps
Handler.postDelayed(mySpeedChecker(),1000);
You'll need to do this for every step taken
public void mySpeedChecker(){
// assuming steps taken stored in a global variable
// and there is one more variable which is updated by this function
// to store number of steps before as global variable to calculate number of steps in time interval.
// this should check if another step has been taken in the interval to check
// if stepsTaken>0 then speed is not zero and you'll know how much steps are taken
// if zero steps are taken then speed is zero
// also it will tell you number of steps taken in last second
// so you can use this to calculate and update speed
// increase the interval for high accuracy but lower frequency of updates
}
Also, it's a good idea to calculate user stride length when GPS is available so, whenever user is using GPS the step length can be calculated by finding out the distance covered/number of steps taken with the help of GPS, this will help your system calculate accurately the distance covered by user with the help of number of steps taken.
The usual way to calculate speed ideally would be using Location.getSpeed(), or some analysis on Accelerometer values. but I'm assuming you want to get it based on steps counted.
To solve one of the problems: " if the user stops, the speed remains constant and doesn't drop to zero."
You can use the Android Activity Recognition Api to see the users current activity. Use
ActivityRecognition.getMostProbableActivity()
If the DetectedActivity is type STILL, you can set your speed to be zero.
You can also get the confidence DetectedActivity.getConfidence() to be sure.
For the other problem when you said it doesn't count speed properly, could you elaborate more on that question?
Every information coming from the sensors has an amount of error and even loose of data.
You should store the information received in a queue or database including the time stamp.
Periodically, you will pick the data and analyze it, including filtering to get a more accurate information.
Hope it helps.
With this C code:
int a = time(NULL);
_daylight = 0;
_timezone = 0;
int b = time(NULL);
assert(a != b);
"a" and "b" will have different values (and not just because they are called a few milliseconds apart). The difference will be whatever the offset of your PC's timezone is from UTC time. Also, changing the _daylight and _timezone values effect pretty much every other function I might use in my C app -- I assume because they all respect that value.
Is there anything like that in Java, or specifically for Java on Android OS? I tried TimeZone.setDefault(), but that didn't change the value that System.currentTimeMillis() returned, so I assume it isn't going to have a "global" effect like the C variables.
I understand that System.currentTimeMillis() is different than time(), in that it "always" returns the number of millis since now and epoch, and the time() function allows you to get "false" (fudged) values that are adjusted according to these global variables you can set.
Just trying to emulate a legacy C app on Android OS. It clears those _timezone and _daylight values which pretty much means it ignores any timezones. So if a user running the app on the west coast enters a time of 3pm, and then they change their timezone settings, or a user on the est coast views that item, it will still show as 3pm.
I know I can use the Calendar object and other methods to make sure I do the proper conversions, but I'd rather just have an easy "I don't care about timezones" settings like I did in the C app and then truely not have to worry about them.
Edit: I would still like to hear what other options I have, but for now I came up with this Java code that I'll do my best to always use for any code that needs to mimic the C app:
// IMPORTANT: Use this function everywhere a Calendar object is needed, instead of calling
// Calendar.getInstance() directly. This returns the correct kludged time that matches
// what our PC application uses (_daylight=0, _timezone=0, time(NULL) in C)
public static Calendar GetCalendarInstance()
{
// Get the current UTC time
Calendar cal = Calendar.getInstance(TimeZone.getTimeZone("UTC"));
// Offset it by the system time zone offset.
// This mimics what the C time(NULL) function does when you set _timezone=0 and _daylight=0
cal.add(Calendar.MILLISECOND, TimeZone.getDefault().getOffset(cal.getTimeInMillis()));
return(cal);
}
Also, I did already find one place in my Android app that I need the real, not adjusted, system time (when using AlarmManager to schedule a PendingIntent). So I guess "global" could be dangerous either way. I still think 95% of my code will be using the version that mimics the C app though, so if possible I'd like to default to that and then only have to do special handling for the other few places.
I want to have my application execute code at a point in the future.
I want to do:
Date now = new Date();
for (Date beep : scheduledBeeps) {
if (beep.after(now))
{
Logger.i("adding beep");
m_beepTimer.postAtTime(beepNow, beep.getTime());
}
}
In the log I can see 4 beeps added, however they never fire. I'm assuming it has something to do with uptimeMillis, but I'm not sure what to do.
You will have to get the difference between now and beep.gettime() and pass it to postattime function. Since uptime is used as base, it may not be accurate if the phone goes to deep sleep.
beep.gettime - now + SystemCLock.uptimeMillis()
should be passed to postattime function
You are currently passing a very large number equivalent to current milliseconds from jan 1 1970.
You could use the Calendar class to set a certain point in time.
Calendar beepTime = Calendar.getInstance();
beepTime.set(Calendar.DAY_OF_MONTH, 2);
beepTIme.set(Calendar.HOUR_OF_DAY, 01);
beepTime.set(Calendar.MINUTE, 55);
beepTime.set(Calendar.SECOND, 00);
getInstance will set it to the current time, and you can change any variable you like, such as the ones above. For example this would create a time at 1:55 on the 2nd of the current month. You would then set this to be the time to go off with
beepTime.getTimeInMillis()
just pop that into your postAtTime method
Edit: Also I don't know enough about your problem to say for sure, but it may be better to use AlarmManager. I know that that still works even if the program is not running, whereas I don't think PostDelayed does. Feel free to correct me if I'm wrong!