1
00:00:02,110 --> 00:00:10,570
Now to conclude the geospatial queries we can run, let's find all places in a certain radius around

2
00:00:10,570 --> 00:00:11,220
the user.

3
00:00:11,260 --> 00:00:14,540
We kind of did this before with the near command,

4
00:00:14,740 --> 00:00:17,500
now I also want to give you a different way.

5
00:00:17,740 --> 00:00:20,430
The near command also sorted the results,

6
00:00:20,440 --> 00:00:21,730
that is the important difference

7
00:00:21,730 --> 00:00:23,010
and I'll come back to that,

8
00:00:23,240 --> 00:00:30,600
let's now place the user here and let's check for a certain radius and all the points in that radius.

9
00:00:30,850 --> 00:00:33,510
So for that, we'll need the coordinates of the user of course

10
00:00:33,850 --> 00:00:40,030
and I'll write a new query, again against my places collection not the areas collection because the

11
00:00:40,030 --> 00:00:46,060
places collection is the collection with these four places we added, right.

12
00:00:46,120 --> 00:00:53,200
So this is the collection I'm interested in now and there, I want to find all elements in an unsorted

13
00:00:53,260 --> 00:00:56,250
order that are within a certain radius.

14
00:00:56,590 --> 00:01:00,750
So here the location field is the field which stores the coordinates,

15
00:01:00,760 --> 00:01:05,360
so this is the field which I will now use with the geoWithin

16
00:01:05,380 --> 00:01:12,520
operator again. Why geoWithin and not geoIntersects? Because now I want to find all places that are within

17
00:01:12,820 --> 00:01:16,780
a place or an area which I'll define here in the query,

18
00:01:16,780 --> 00:01:18,020
previously we had the opposite,

19
00:01:18,040 --> 00:01:24,070
we had a point in the query and wanted to find an area that surrounds the point.

20
00:01:24,100 --> 00:01:32,180
So geoWithin normally takes a geometry operator which describes a geo json object but there is

21
00:01:32,180 --> 00:01:34,680
a special operator you can use in

22
00:01:34,680 --> 00:01:35,360
mongodb

23
00:01:35,570 --> 00:01:41,360
and that is the centerSphere operator, $centerSphere.

24
00:01:41,450 --> 00:01:46,900
This is a helpful operator that allows you to quickly get a circle around a point,

25
00:01:46,940 --> 00:01:53,780
so essentially it use a radius and a center and gives you the whole circle around that center therefore.

26
00:01:54,530 --> 00:02:01,280
centerSphere takes an array as a value and that array has two elements,

27
00:02:01,280 --> 00:02:08,120
the first element is another array which holds your coordinates, again longitude and then latitude

28
00:02:08,330 --> 00:02:14,180
and with coordinates, I mean the coordinates of the center of the circle you want to draw.

29
00:02:14,660 --> 00:02:17,970
The second element in this array is the radius then

30
00:02:18,290 --> 00:02:25,580
and now this radius needs to be translated manually from meters or miles to radians.

31
00:02:25,580 --> 00:02:31,190
Now this conversion is relatively easy to do and you'll find an article in the mongodb docs

32
00:02:31,190 --> 00:02:31,850
too,

33
00:02:32,240 --> 00:02:38,150
you'll find an article which mentions how to translate miles to radians or kilometers.

34
00:02:38,150 --> 00:02:39,920
Now I will work with kilometers here,

35
00:02:39,920 --> 00:02:42,450
so this is the value which is interesting to me

36
00:02:42,950 --> 00:02:48,270
and now I want to look in a radius of let's say one kilometer,

37
00:02:48,560 --> 00:02:54,360
so one divided by this value without the comma is my radius here.

38
00:02:55,490 --> 00:03:02,120
If I hit enter, I find two points, let me pretty print this, I find the Conservatory of Flowers and the

39
00:03:02,120 --> 00:03:03,530
Golden Gate Tennis Park,

40
00:03:03,620 --> 00:03:07,300
what I don't find is to California Academy of Sciences,

41
00:03:07,460 --> 00:03:12,640
let's quickly measure the distance of this point to this point.

42
00:03:12,710 --> 00:03:15,300
Now it looks like this should be in there,

43
00:03:15,320 --> 00:03:21,910
so my assumption would be that it's simply not in there because I have the wrong coordinates stored.

44
00:03:21,950 --> 00:03:28,430
Let's quickly look into our database for the California Academy of Sciences and indeed, the coordinates

45
00:03:28,430 --> 00:03:29,700
here don't match mine,

46
00:03:29,750 --> 00:03:34,820
that is because previously I didn't have my browser perfectly centered on them.

47
00:03:34,820 --> 00:03:39,130
So let's quickly update that to be really sure that our query here works correctly.

48
00:03:39,380 --> 00:03:45,530
So I'll update by simply searching for that objectId here, places and you know how to update,

49
00:03:45,650 --> 00:03:46,770
I hope you still know it,

50
00:03:47,710 --> 00:03:53,230
first of all you filter for _id being equal to my objectId here

51
00:03:53,240 --> 00:03:59,010
and the second value is then the value where you describe the update with $set.

52
00:03:59,050 --> 00:04:03,200
I will set my location to a new value,

53
00:04:03,370 --> 00:04:10,900
the type will still be a point, that hasn't changed but the coordinates, that will now be a new array

54
00:04:10,900 --> 00:04:12,450
with new values,

55
00:04:12,640 --> 00:04:17,910
first the longitude here and second, the latitude.

56
00:04:18,110 --> 00:04:24,020
So let's grab this and this is now the exact point, the exact location of that institute.

57
00:04:24,110 --> 00:04:28,570
So with this updated, let's run our query again and

58
00:04:28,670 --> 00:04:32,320
now the California Academy of Sciences is included too

59
00:04:32,510 --> 00:04:38,630
but please note that Nopa is not included and that makes sense because Nopa should definitely not

60
00:04:38,630 --> 00:04:40,900
be inside of our circle.

61
00:04:41,270 --> 00:04:46,060
If I measure the distance of this point to this point, it's above two kilometers,

62
00:04:46,130 --> 00:04:48,630
so it makes sense that this is not included.

63
00:04:49,190 --> 00:04:52,660
So this is how we can run such queries as well,

64
00:04:52,670 --> 00:04:59,330
the important difference to the near query which also allowed us to find places within a certain min

65
00:04:59,360 --> 00:05:02,120
and max distance is that with this query,

66
00:05:02,150 --> 00:05:07,100
we get back an unsorted list, this keeps the order of the elements in the database

67
00:05:07,190 --> 00:05:14,570
or we could of course manually sort them with sort method. The near method did give us the elements in

68
00:05:14,570 --> 00:05:18,130
a certain radius and did sort them by proximity

69
00:05:18,320 --> 00:05:23,780
and that really depends on your requirements, what you need, do you need sorted results with the nearest

70
00:05:23,780 --> 00:05:24,860
result first

71
00:05:24,860 --> 00:05:30,410
or do you need unsorted results and you just want to get a list of the elements in an area in general?

72
00:05:30,800 --> 00:05:32,840
This is what geoWithin can do,

73
00:05:32,840 --> 00:05:35,960
near is otherwise your solution for the sorted list.