1
00:00:02,520 --> 00:00:04,960
We had a look at number int and number long.

2
00:00:05,220 --> 00:00:08,500
Now one obvious question you might have at this point is

3
00:00:08,880 --> 00:00:12,970
especially with that string thing which I mentioned at the end of the last lecture,

4
00:00:13,050 --> 00:00:16,710
why don't we just store all numbers as strings?

5
00:00:16,740 --> 00:00:24,990
Let's say we have accounts collection with bank accounts or something like this and in there, we could

6
00:00:25,050 --> 00:00:27,650
insert one bank account for one person,

7
00:00:27,780 --> 00:00:33,510
so for person Max, of course we don't need that name and the interesting part here is like what's in

8
00:00:33,510 --> 00:00:39,440
there, so amount and now we could store it as a value like this but let's say we only have integer values

9
00:00:39,530 --> 00:00:44,370
there, unrealistic of course but let's say we only have full numbers, so we could of course store it

10
00:00:44,460 --> 00:00:46,070
as a number

11
00:00:46,320 --> 00:00:48,950
or number long depending on how big the number will be,

12
00:00:48,960 --> 00:00:51,620
I can tell you number int would suffice for me here

13
00:00:51,780 --> 00:00:55,020
but why don't we just store it as text

14
00:00:55,140 --> 00:01:01,050
if we face no problem here and we can store as big of a number as we want like this, which is not my

15
00:01:01,050 --> 00:01:04,740
actual bank account but let's say this could be stored,

16
00:01:04,820 --> 00:01:07,650
it's a valid text, the text can be as long as you want,

17
00:01:07,650 --> 00:01:17,150
so why don't we store it like this? If I now have a look at my bank accounts here, we see that value was

18
00:01:17,150 --> 00:01:21,130
of course stored exactly because it's just some text.

19
00:01:21,350 --> 00:01:23,690
Now let's delete this quickly

20
00:01:31,010 --> 00:01:32,800
and insert this again a bit smaller

21
00:01:35,230 --> 00:01:39,490
so that we that we can see what's going on and let's just store 10.

22
00:01:39,640 --> 00:01:42,760
Now one reason why storing it as a text is a bad idea is

23
00:01:42,790 --> 00:01:48,440
because if I now need to do any calculations with it, it will just fail.

24
00:01:48,700 --> 00:01:55,800
So let's say I want to update my account here and I only have one entry in here, so I don't need to

25
00:01:55,800 --> 00:02:04,320
filter and update one and I do this by incrementing, you learned about the inc operator, by incrementing

26
00:02:04,790 --> 00:02:09,940
amount by 10. If I hit enter here,

27
00:02:09,950 --> 00:02:14,430
we see that we cannot apply inc to a value of non-numeric type,

28
00:02:14,570 --> 00:02:20,360
so it essentially fails because amount is a string and not a number.

29
00:02:20,360 --> 00:02:26,920
And that is one problem we'll face if we ever want to do calculations, we'll not work with a string,

30
00:02:27,140 --> 00:02:31,250
it will work with number int and number long though.

31
00:02:31,340 --> 00:02:33,800
So if I clear my account again

32
00:02:34,100 --> 00:02:38,530
and I now insert my account, my amount but not as a string

33
00:02:38,600 --> 00:02:45,770
but let's say as number int, 10 wrapped in quotation marks which is the preferred way of passing a value

34
00:02:45,770 --> 00:02:46,500
here, 

35
00:02:46,550 --> 00:02:52,570
if I do that and I repeat my update one operation, now this succeeded.

36
00:02:52,760 --> 00:02:54,290
And now there's one important thing,

37
00:02:54,390 --> 00:02:58,500
if I now output this, we see it's 20 but now it's not a number int anymore,

38
00:02:58,550 --> 00:03:01,660
it's a normal 64 bit float

39
00:03:01,670 --> 00:03:07,940
again. The reason for that is that I increment it by such a normal number and not by an int and therefore

40
00:03:08,030 --> 00:03:11,300
mongodb automatically converts it to this type of number,

41
00:03:11,300 --> 00:03:14,610
so to the float 64.

42
00:03:14,650 --> 00:03:23,140
So if I wanted to keep it a number int, I'll quickly delete everything, re-insert my initial 10 bucks

43
00:03:23,170 --> 00:03:29,570
into my account and then update, I would update by also wrapping this with number int,

44
00:03:29,620 --> 00:03:31,930
so I would update by passing a number int

45
00:03:32,050 --> 00:03:33,900
and this as you see succeeds.

46
00:03:33,910 --> 00:03:35,700
So we're not working with strings here,

47
00:03:35,710 --> 00:03:40,360
we are working with these special number types provided by mongodb

48
00:03:40,470 --> 00:03:44,370
and therefore now, if I have to look at this, we see the amount 20.

49
00:03:44,500 --> 00:03:52,870
Now unfortunately this still looks like a normal 64 bit because it is received to automatically display

50
00:03:52,870 --> 00:03:54,450
that such by the shell here

51
00:03:54,520 --> 00:03:57,150
but it will actually be stored as a number int

52
00:03:57,140 --> 00:04:04,120
behind the scenes because we only work with number ints for both inserting the document and updating

53
00:04:04,120 --> 00:04:08,650
it. Now the same does work for longs,

54
00:04:08,780 --> 00:04:16,340
if we go back to the companies example and I first of all, delete everything in there so that we can start

55
00:04:16,340 --> 00:04:17,770
from scratch,

56
00:04:17,840 --> 00:04:22,280
let me then insert one document into companies.

57
00:04:25,270 --> 00:04:31,260
Let's say there we have the funding or not the funding, let's say the evaluation and let's add number

58
00:04:31,260 --> 00:04:32,380
long here.

59
00:04:32,410 --> 00:04:34,580
Now let's enter a big number,

60
00:04:36,680 --> 00:04:44,750
like this which is too big for int 32 but which is ok as a number long. Let's first of all look into our

61
00:04:44,750 --> 00:04:45,560
companies here,

62
00:04:45,560 --> 00:04:52,670
so companies find one, this is the company I entered, the value was stored in there, it was stored in exactly

63
00:04:53,180 --> 00:05:00,030
the value I added. By the way, just because I haven't mentioned it yet but of course you can also store

64
00:05:00,030 --> 00:05:00,870
negative numbers,

65
00:05:00,870 --> 00:05:05,190
so I'm using positive numbers here but the same is possible for the same range as you saw in the slide

66
00:05:05,280 --> 00:05:06,490
with negative numbers.

67
00:05:06,840 --> 00:05:08,530
So I store this

68
00:05:08,700 --> 00:05:15,430
and now let's update this to show that we again can do calculations, mathematical calculations on number

69
00:05:15,450 --> 00:05:21,390
longs too. If I update one document, the only one I have in there, I can do this with the inc

70
00:05:21,410 --> 00:05:30,670
operator and I want to update my valuation by let's say, let's first of all say 100 or let's say only

71
00:05:30,670 --> 00:05:32,670
one because that will be easier to check,

72
00:05:32,680 --> 00:05:34,770
so by one. It succeeds

73
00:05:34,780 --> 00:05:41,060
but if I find my value, we actually see we got a problem now. The value we get back is incorrect

74
00:05:41,080 --> 00:05:49,030
and the reason for that is simply by changing it by one, this was converted to a 64 bit float which can

75
00:05:49,030 --> 00:05:50,410
store a number this big

76
00:05:50,470 --> 00:05:53,440
and therefore simply well rounds it in the end.

77
00:05:53,440 --> 00:05:55,840
So this is of course not what we wanted,

78
00:05:57,060 --> 00:05:59,770
so let's reset this,

79
00:05:59,790 --> 00:06:01,100
delete everything,

80
00:06:01,140 --> 00:06:03,180
insert our value with the number long

81
00:06:03,420 --> 00:06:04,730
and now update again

82
00:06:04,740 --> 00:06:10,130
but now I update with number long and that will be my value of one

83
00:06:10,320 --> 00:06:13,970
but now again, created as a long number

84
00:06:14,070 --> 00:06:19,740
with that special constructor provided by the mongodb shell. This succeeds

85
00:06:19,740 --> 00:06:26,350
and if I now have a look at my element here, we see this is indeed the correct number

86
00:06:26,340 --> 00:06:28,770
now, previously we had 89 at the end,

87
00:06:28,770 --> 00:06:31,540
now it's 90 because we incremented by 1.

88
00:06:31,830 --> 00:06:36,160
So this is how we can do math with these number longs and number ints

89
00:06:36,270 --> 00:06:40,210
and that is of course a crucial part of using them, so you can use them.

90
00:06:40,230 --> 00:06:45,280
You can also filter by them by the way, you can use them with greater than filters and sort by them,

91
00:06:45,360 --> 00:06:53,850
so they really work like normal numbers, just with the benefits of well this enhanced range or this bigger

92
00:06:53,850 --> 00:06:55,280
range you have available.