1 00:00:05,200 --> 00:00:09,520 Alright, so how did you get on with the challenge? It gave you a chance to practise iterating over 2 00:00:09,520 --> 00:00:15,913 a dictionary, and adding items to one. Now I've called my Python file simple_deepcopy, 3 00:00:15,913 --> 00:00:19,600 and I've already copied the sample code I showed at the end of the last video. 4 00:00:19,600 --> 00:00:25,120 So to copy the dictionary, we need to start by creating a new dictionary. Alright, so I'm 5 00:00:25,120 --> 00:00:31,280 going to remove this line 14, line 13 rather, and replace that with new_dict 6 00:00:31,280 --> 00:00:36,400 is equal to, and we want an empty dictionary, so left and right curly braces. Next we need to 7 00:00:36,400 --> 00:00:43,680 iterate over the keys and values of the original dictionary. So I'm going to type for key comma value 8 00:00:43,680 --> 00:00:52,560 in d dot items, parentheses colon. The challenge description stated that our code only has to cope 9 00:00:52,560 --> 00:00:58,025 with values that are lists or dictionaries. This means we can use their copy method to create a copy 10 00:00:58,025 --> 00:01:07,328 of the value. So I'm going to type new_value is equal to value.copy, parentheses there, 11 00:01:07,328 --> 00:01:12,419 and we finish by adding the new value to the new dictionary. We use the key that it had in the original. 12 00:01:12,419 --> 00:01:19,040 So I'm going to type new_dict square brackets, key in square brackets, equals 13 00:01:19,040 --> 00:01:23,840 new_value. Alright, and when the loop finishes, all the values will have been copied 14 00:01:23,840 --> 00:01:30,812 and added to the new dictionary. All that remains to do, is to return the new dictionary, so let's do that. 15 00:01:30,812 --> 00:01:39,280 On line 18, I'm going to type return new_dict. Alright. So on line 22, we change the amount of ginger 16 00:01:39,280 --> 00:01:44,560 that we have in the copied dictionary. Because we copied the inner dictionaries, the original value 17 00:01:44,560 --> 00:01:53,200 should remain unchanged. Now if we have a quick look at the recipe's dictionary in contents.py, 18 00:01:53,200 --> 00:02:00,277 down here, we can see that butter chicken needs three gingers. So back in simple_deepcopy again, 19 00:02:00,277 --> 00:02:07,200 line 24 should hopefully print three. Let's run the program now, and a space there to 20 00:02:07,200 --> 00:02:14,064 keep IntelliJ happy. We'll run it again. We confirm that the copy now has the value of 300 for ginger, 21 00:02:14,064 --> 00:02:19,680 but the original value is unchanged. Alright, so that's the challenge completed. We deliberately 22 00:02:19,680 --> 00:02:25,280 restricted what your code had to do to complete the challenge, because this isn't a full deep copy. 23 00:02:25,280 --> 00:02:30,480 It only creates a copy of dictionaries and lists that are values. If they also contain dictionaries 24 00:02:30,480 --> 00:02:36,635 or lists, a shallow copy will be performed on those inner objects. An example is better than words here, 25 00:02:36,635 --> 00:02:45,605 so let's actually create a new Python file. So I'm going to call mine deep copy is recursive - 26 00:02:45,605 --> 00:02:53,244 deepcopy_is_recursive. I'm going to import our my_deepcopy function 27 00:02:53,244 --> 00:02:59,440 and the copy module. So I'm going to do from simple_deepcopy 28 00:02:59,440 --> 00:03:06,160 import my_deepcopy. On the next line, import copy. Alright, so we need a dictionary to copy 29 00:03:06,160 --> 00:03:11,520 and I'm going to keep it very simple. So I'm going to paste this in, line four. 30 00:03:11,520 --> 00:03:14,900 So I kept the dictionary list small to make it easy to see what's going on. 31 00:03:14,900 --> 00:03:21,253 Let's continue on. I'm gonna type copy underscore 1 equals copy.deepcopy 32 00:03:21,253 --> 00:03:30,000 original. And on the next line, copy underscore two equals my_deepcopy original. 33 00:03:30,000 --> 00:03:40,720 Then line 12, print original, then we'll also print copy 1. We'll do the same - print copy 2. 34 00:03:40,720 --> 00:03:45,040 Alright, and again, so I've kept the dictionary and the list small, just to make it easy to 35 00:03:45,040 --> 00:03:50,960 see what's going on. So on line 9, we use the deep copy function to copy the original dictionary. 36 00:03:50,960 --> 00:03:55,585 Line 10 uses the my_deepcopy function from the challenge. 37 00:03:55,600 --> 00:03:59,840 So let's run the program and see what the results are. 38 00:04:04,000 --> 00:04:09,120 And you can see there, all three dictionaries look the same. So ignore the first two lines of output. 39 00:04:09,120 --> 00:04:13,520 They're produced because we imported simple underscore deepcopy. You'll learn all about 40 00:04:13,520 --> 00:04:19,360 that when we talk about importing, later in the course. The value for Tim, on line five, is a list 41 00:04:19,360 --> 00:04:24,800 containing my surname and a list of jobs. To check that our two copies contain copies of these lists, 42 00:04:24,800 --> 00:04:31,680 I'll append the country for Tim, and also for Jean-Paul. So we'll come down here, 43 00:04:31,680 --> 00:04:39,360 line 12. We're going to type original square brackets, Tim in double quotes, dot append. 44 00:04:39,360 --> 00:04:44,240 In parentheses, double quotes Australia, and for Jean-Paul I'm going to type 45 00:04:44,240 --> 00:04:54,880 original square brackets double quotes J-P dot append parentheses double quotes UK. 46 00:04:54,880 --> 00:04:58,000 Now if we run this, 47 00:04:58,000 --> 00:05:03,030 we can see in the case of the first copy, we can see that at the top, the first one, 48 00:05:03,030 --> 00:05:08,800 has got both the countries, but the countries don't get added to the lists in our copies - the second 49 00:05:08,800 --> 00:05:14,160 and third lines. To see the difference between our function, and the copy module's deepcopy function, 50 00:05:14,160 --> 00:05:18,960 what I'm going to do is mutate the inner lists in the original. Now I'm referring to the inner 51 00:05:18,960 --> 00:05:25,200 lists that contain Programmer and Teacher, back up here on lines 5 and 6. Jean-Paul 52 00:05:25,200 --> 00:05:30,160 worked as a motorcycle courier for a while, and I was a cashier in an amusement arcade, 53 00:05:30,160 --> 00:05:34,925 so I'm going to append those jobs to the inner lists. Let's go ahead and do that. Come down here, 54 00:05:34,925 --> 00:05:41,040 after the country additions there, and I'm going to type original, square brackets double quotes, Tim 55 00:05:41,040 --> 00:05:50,480 in those double quotes. Then another double quotes and 1 dot append parentheses double quotes cashier. 56 00:05:50,480 --> 00:05:55,840 Then on the next line I'm going to type jp_list equals original 57 00:05:55,840 --> 00:06:02,640 square brackets double quotes J-P. On the next line, jp_list then 1 58 00:06:02,640 --> 00:06:08,103 in square brackets, dot append parenthesis double quotes Courier, 59 00:06:09,440 --> 00:06:16,878 like so. So what I've done there, is I've made the change in two different ways. So on line 15, 60 00:06:16,878 --> 00:06:20,880 the code refers to the second item for the value of Tim. So I'll just scroll up so we can 61 00:06:20,880 --> 00:06:25,440 see the original there as well. So I've made the change in two different ways. On line 15, 62 00:06:25,440 --> 00:06:31,200 the code refers to the second item in the value for Tim. The value is a list, and the second item 63 00:06:31,200 --> 00:06:36,850 in that list is another list, which contains Programmer and Teacher. That's on line five. 64 00:06:36,850 --> 00:06:43,619 We append cashier to that inner list on line 15. Lines 16 and 17 do the same thing - jp_list 65 00:06:43,619 --> 00:06:49,760 is bound to the value for jp. jp_list 1 in square brackets is the second item in that list. 66 00:06:49,760 --> 00:06:55,520 I'll split the code up into two steps, in case you're unsure about the indexing on line 15. 67 00:06:55,520 --> 00:07:01,280 In both cases, we append a new job to the inner list. If you find those indices on line 15 confusing, 68 00:07:01,280 --> 00:07:06,560 review the lectures in the list and tuple section, starting with nested tuples and lists. 69 00:07:06,560 --> 00:07:12,160 Alright, so let's run the program to check the output. So we get the new jobs added to the list in the original 70 00:07:12,160 --> 00:07:21,280 dictionary. We can scroll over and see that; Cashier for me and Courier for Jean-Paul. Now the second 71 00:07:21,280 --> 00:07:27,680 dictionary in the output doesn't contain the new jobs. You can see that there, in both cases. 72 00:07:27,680 --> 00:07:33,709 We use the copy module's deepcopy function to create copy_1. We can see that on line9, 73 00:07:33,709 --> 00:07:39,840 and that performs a deep copy of all contained objects. Our my_deepcopy function 74 00:07:39,840 --> 00:07:45,680 only goes one level down. We use the copy method in the function, and that performs a shallow copy. 75 00:07:45,680 --> 00:07:54,387 As a result, the inner list in copy_2 is a reference to the same inner list in original. 76 00:07:54,387 --> 00:07:59,600 The deepcopy function uses a technique called recursion. When it finds an object 77 00:07:59,600 --> 00:08:05,280 that can contain other objects, it calls itself again to copy those containers. 78 00:08:05,280 --> 00:08:09,840 If we wanted to write a complete deep copy function, you'd need to use recursion. 79 00:08:09,840 --> 00:08:15,200 We'll cover recursion a bit later in the course, when we revisit functions. Alright, so that's the end of 80 00:08:15,200 --> 00:08:20,960 the practical use of dictionaries. In the next few lectures, we're going to be looking at the theory. 81 00:08:20,960 --> 00:08:26,720 A dictionary can contain millions of keys, but retrieving a value for a key is very fast. 82 00:08:26,720 --> 00:08:31,200 We'll look at the theory of how that works, using something called hashes. We saw them mentioned in 83 00:08:31,200 --> 00:08:38,640 the documentation, earlier - now it's time to find out what they are. See you in the next video.