how get max date from arraylist in typescript

You can use this function to compute the maximum date of an array of dates:

arr = ["2017-07-20T04:04:43.000Z", "2017-07-21T04:04:43.000Z","2017-07-20T04:04:43.000Z"];
new Date(Math.max.apply(null,
             arr
             .map(x => new Date(x))));

Specific solution (works only for this specific JSON containing 1-element arrays):

Get list of data points

var data = {
  "jobs" : [ {
    "job_id" : 10,
    "users" : [ {
      "user_id" : 11,
      "data_points" : [ {
        "duration" : 0,
        "start_time" : "2017-07-20T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      }, {
        "duration" : 0,
        "start_time" : "2017-07-21T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      }, {
        "duration" : 0,
        "start_time" : "2017-07-20T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      } ]
    } ]
  } ]
};

var d = data["jobs"][0]["users"][0]["data_points"];

var max_start_time = 
     new Date(Math.max.apply(null,
     d
       .map(z =>   
        z["start_time"]    
       ).map(w => new Date(w))));
       
console.log(max_start_time);

General solution (you get an array indexed by "job_id" and "user_id"). This would be the equivalent of something like the sql select job_id, user_id, max(start_time) from ... group by job_id, user_id

var data = {
  "jobs" : [ 
  {
    "job_id" : 110,
    "users" : [ {
      "user_id" : 11,
      "data_points" : [ {
        "duration" : 0,
        "start_time" : "2017-07-20T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      }, {
        "duration" : 0,
        "start_time" : "2017-07-21T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      }, {
        "duration" : 0,
        "start_time" : "2017-07-20T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      } ]
    } ]
  },
  {
    "job_id" : 111,
    "users" : [ {
      "user_id" : 12,
      "data_points" : [ {
        "duration" : 0,
        "start_time" : "2017-08-20T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      }, {
        "duration" : 0,
        "start_time" : "2017-08-21T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      }, {
        "duration" : 0,
        "start_time" : "2017-08-20T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      } ]
    },
    {
      "user_id" : 13,
      "data_points" : [ {
        "duration" : 0,
        "start_time" : "2017-08-20T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      }, {
        "duration" : 0,
        "start_time" : "2017-08-21T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      }, {
        "duration" : 0,
        "start_time" : "2017-08-20T04:04:43.000Z",
        "datastream" : "GENERIC",
        "username" : "[email protected]"
      } ]
    }]
  }]
};

var max_start_time = {};

var r=data["jobs"].forEach(job =>
  {
     if (job.hasOwnProperty("job_id")) {
       if (job.hasOwnProperty("users")) {
         job["users"].forEach(userData =>
           { 
             if (userData.hasOwnProperty("data_points")) {
               max_start_time[[job["job_id"],userData["user_id"]]]=
               new Date(Math.max.apply(null,
                 userData["data_points"]
                 .map(z =>   
                 z["start_time"]    
                 ).map(w => new Date(w))));        
             }
           }
         )
       } 
     }
  }
);

console.log(max_start_time);

Try this

var userSortedData=jobs[0].users[0]. data_points.sort(function(a, b){return (b.start_time >a.start_time)?1:-1 ;});
console.log(userSortedData[0]);  // max Date

This should do it. What it does is go trough each list (jobs, users, data_points) and reduce every one of them to the max date representing it.

let data = {
    "jobs": [{
        "job_id": 10,
        "users": [{
            "user_id": 11,
            "data_points": [{
                "duration": 0,
                "start_time": "2017-07-20T04:04:43.000Z",
                "datastream": "GENERIC",
                "username": "[email protected]"
            }, {
                "duration": 0,
                "start_time": "2017-07-21T04:04:43.000Z",
                "datastream": "GENERIC",
                "username": "[email protected]"
            }, {
                "duration": 0,
                "start_time": "2017-07-20T04:04:43.000Z",
                "datastream": "GENERIC",
                "username": "[email protected]"
            }]
        }]
    }]
}

let max = data.jobs.reduce((prev, job) => {
    let maxJobStartTime = job.users.reduce((prev, user) => {
        let maxUserStartTime = user.data_points.reduce((prev, point) => {
            return prev - new Date(point.start_time).getTime() > 0 ? prev : new Date(point.start_time).getTime();
        }, 0)
        console.log(maxUserStartTime);
        return prev - maxUserStartTime > 0 ? prev : maxUserStartTime;
    }, 0)
    return prev - maxJobStartTime > 0 ? prev : maxJobStartTime;
}, 0)

console.log(new Date(max));

See a working playground example here.

A breakdown:

interface User {
    user_id: number,
    data_points: { start_time: string }[]
}

interface Job {
    job_id: number,
    users: User[]
}

interface Data {
    jobs: Job[]
}

// max time from a user's data_points
function maxUserStartTime(user: User) {
    return user.data_points.reduce((prev, point) => {
        return prev - new Date(point.start_time).getTime() > 0 ? prev : new Date(point.start_time).getTime();
    }, 0);
}

//using the max time of an user calculate the max time of all users in a job
function maxJobStartTime(job: Job) {
    return job.users.reduce((prev, user) => {
        let max = maxUserStartTime(user);
        return prev - max > 0 ? prev : max;
    }, 0);
}


// using the max time for a job calculate the max time out of all jobs
function maxTime(data: Data) {
    return data.jobs.reduce((prev, job) => {
        let max = maxJobStartTime(job)
        return prev - max > 0 ? prev : max;
    }, 0);
}

console.log(new Date(maxTime(data)));

A working example of the clean version can be found here.