1
00:00:00,002 --> 00:00:05,002
(upbeat music)

2
00:00:05,002 --> 00:00:06,006
- [Instructor] All right, let's examine my code

3
00:00:06,006 --> 00:00:07,006
for this challenge.

4
00:00:07,006 --> 00:00:10,005
We needed to add a new column to the inventory data

5
00:00:10,005 --> 00:00:12,003
to calculate the margin

6
00:00:12,003 --> 00:00:16,000
between the consumer prices and wholesale prices.

7
00:00:16,000 --> 00:00:18,007
So I already have two functions

8
00:00:18,007 --> 00:00:23,004
for reading and writing the CSV data.

9
00:00:23,004 --> 00:00:26,000
So I just needed to write the code

10
00:00:26,000 --> 00:00:28,003
to perform the calculations.

11
00:00:28,003 --> 00:00:33,004
My code starts by reading the Inventory.CSV file,

12
00:00:33,004 --> 00:00:37,000
and then I get the first row of the result,

13
00:00:37,000 --> 00:00:38,007
which is the row of headers.

14
00:00:38,007 --> 00:00:43,005
So I add the new string margin on the end of that list.

15
00:00:43,005 --> 00:00:45,005
That's the easy part, all right?

16
00:00:45,005 --> 00:00:48,003
So then I use an array slice

17
00:00:48,003 --> 00:00:51,008
to get the rows from one until the end of the file.

18
00:00:51,008 --> 00:00:53,001
So this is all the data, right?

19
00:00:53,001 --> 00:00:55,005
This is all the data after the headers.

20
00:00:55,005 --> 00:00:58,002
So for each one of these rows, I loop over

21
00:00:58,002 --> 00:00:59,006
and I calculate the difference

22
00:00:59,006 --> 00:01:03,002
between the values at index four and index three.

23
00:01:03,002 --> 00:01:07,001
And I cast each one of those to a decimal object.

24
00:01:07,001 --> 00:01:09,006
And you can see that if I scroll back up,

25
00:01:09,006 --> 00:01:13,001
I imported the decimal class from the decimal module,

26
00:01:13,001 --> 00:01:15,009
and I'll get to that in a second.

27
00:01:15,009 --> 00:01:18,001
So I calculate the difference

28
00:01:18,001 --> 00:01:21,000
and then I add that value onto the end

29
00:01:21,000 --> 00:01:23,007
of the existing row of the data

30
00:01:23,007 --> 00:01:26,002
and then I just simply write out my array

31
00:01:26,002 --> 00:01:28,007
to the output CSV file.

32
00:01:28,007 --> 00:01:31,008
Now, the reason I need to cast each of these strings

33
00:01:31,008 --> 00:01:33,007
to a decimal value

34
00:01:33,007 --> 00:01:37,001
in order to preserve the floating point precision

35
00:01:37,001 --> 00:01:38,002
of the currency,

36
00:01:38,002 --> 00:01:42,003
and this was the hint that I gave in the previous value,

37
00:01:42,003 --> 00:01:45,001
you could have just cast each of these values

38
00:01:45,001 --> 00:01:47,001
to floating point numbers,

39
00:01:47,001 --> 00:01:50,009
but that actually results in a loss of precision.

40
00:01:50,009 --> 00:01:53,009
So let's see what happens when we do that.

41
00:01:53,009 --> 00:01:59,007
I'm going to change this to float.

42
00:01:59,007 --> 00:02:00,005
All right?

43
00:02:00,005 --> 00:02:05,009
And I'm going to run the code again, all right?

44
00:02:05,009 --> 00:02:08,006
And when I do that, if we look at the output file,

45
00:02:08,006 --> 00:02:11,006
look at the difference, right?

46
00:02:11,006 --> 00:02:13,002
There's all these zeros

47
00:02:13,002 --> 00:02:16,006
and then some teeny tiny little, you know, decimal point

48
00:02:16,006 --> 00:02:17,004
on the end.

49
00:02:17,004 --> 00:02:19,009
You know, in this case, it's, you know, really large.

50
00:02:19,009 --> 00:02:21,004
Some aren't affected that much,

51
00:02:21,004 --> 00:02:25,000
but some, they have all these extra decimals in them, right?

52
00:02:25,000 --> 00:02:27,006
Now, this is not a bug, just to be clear.

53
00:02:27,006 --> 00:02:29,005
This is not a Python bug or anything.

54
00:02:29,005 --> 00:02:34,000
It's a result of how computers store floating point numbers.

55
00:02:34,000 --> 00:02:36,002
So to prevent this from happening,

56
00:02:36,002 --> 00:02:38,007
when we work with currency values,

57
00:02:38,007 --> 00:02:41,006
we use the decimal module,

58
00:02:41,006 --> 00:02:44,006
which I've imported right here at the top, right?

59
00:02:44,006 --> 00:02:48,005
We use the decimal module to preserve currency precision.

60
00:02:48,005 --> 00:02:50,007
And you can learn more about this module,

61
00:02:50,007 --> 00:02:52,003
just go into the Python documentation

62
00:02:52,003 --> 00:02:55,007
and it'll explain how the decimal module works.

63
00:02:55,007 --> 00:02:57,004
All right, so that's my solution.

64
00:02:57,004 --> 00:02:59,000
How does my code compare to yours?