WEBVTT 00:00:00.640 --> 00:00:05.990 Assigning a standard array to another one is very convenient and in this case, 00:00:05.990 --> 00:00:09.250 it makes more sense than utilizing the copy function. 00:00:09.640 --> 00:00:14.410 However, if we wanted to, we could have used this function on standard arrays, 00:00:14.410 --> 00:00:14.860 too. 00:00:15.240 --> 00:00:17.890 We just need to find a way to provide the correct pointers 00:00:17.890 --> 00:00:19.890 to the first elements of both arrays, 00:00:19.890 --> 00:00:23.150 and a pointer to the first memory address after the 00:00:23.150 --> 00:00:25.260 last element of the first array. 00:00:25.640 --> 00:00:26.920 And that's actually easy. 00:00:26.920 --> 00:00:28.790 To get the location of the first element, 00:00:28.790 --> 00:00:30.480 we can use the begin function, 00:00:30.480 --> 00:00:33.760 and to get the location of the memory after the last element, 00:00:33.760 --> 00:00:35.750 we can use the end function. 00:00:36.140 --> 00:00:40.720 These two functions provided by the standard array are also known as iterators. 00:00:41.540 --> 00:00:46.280 The other container classes from the standard library implement these iterators, 00:00:46.280 --> 00:00:46.670 too. 00:00:46.670 --> 00:00:51.650 Iterators provide an interface for algorithm functions, like this one. 00:00:52.040 --> 00:00:55.520 They bridge the gap between algorithms and containers, 00:00:55.520 --> 00:00:59.610 because algorithms need to know how to iterate over all of the 00:00:59.610 --> 00:01:02.260 elements inside of a specific container. 00:01:02.840 --> 00:01:04.860 This may seem silly in this case, 00:01:04.860 --> 00:01:07.550 because why doesn't this copy algorithm just know that 00:01:07.550 --> 00:01:10.850 the array starts at the index of 0 and finishes at the 00:01:10.850 --> 00:01:13.550 index of 0 plus the array size? 00:01:13.940 --> 00:01:17.760 But don't forget that these algorithms have to work on all kinds of containers, 00:01:17.760 --> 00:01:21.050 and most of them are not as simple as this basic array. 00:01:21.440 --> 00:01:25.470 By providing a common interface, algorithms can do what they do, 00:01:25.470 --> 00:01:29.560 without the knowledge of the underlying data structure implementation. 00:01:30.740 --> 00:01:35.260 I will define the begin and end functions for the myArray class. 00:01:35.640 --> 00:01:37.740 Iterators are a broad topic. 00:01:37.740 --> 00:01:40.420 There are all kinds of iterator categories: contiguous, 00:01:40.780 --> 00:01:44.960 random access, bi‑directional, forward, and input iterators. 00:01:45.340 --> 00:01:47.620 We could have a whole course on this topic. 00:01:47.630 --> 00:01:51.470 That is why I will treat this lesson as a soft introduction to how we can 00:01:51.470 --> 00:01:55.600 utilize these two basic functions to extend the abilities of the array to 00:01:55.600 --> 00:01:58.560 work with other parts of the standard library. 00:01:59.040 --> 00:02:02.180 The begin function returns a pointer to the first element, 00:02:02.180 --> 00:02:06.960 and the end function returns a pointer to the place after the last element. 00:02:08.039 --> 00:02:11.050 Okay, now let's go down to the main function. 00:02:11.170 --> 00:02:14.370 I will create a new array with three integers. 00:02:14.370 --> 00:02:17.800 Since myArray class now implements these two new functions, 00:02:17.800 --> 00:02:21.130 we can loop over this array with the range‑based for loop, 00:02:21.140 --> 00:02:27.080 which looks much cleaner and shorter than a regular for loop. If I try 00:02:27.080 --> 00:02:29.950 to compile this now, you will see that it works. 00:02:30.640 --> 00:02:35.230 The range‑based for loop will use the iterators in the background to figure 00:02:35.230 --> 00:02:38.160 out how to go through all of the elements of the array. 00:02:39.140 --> 00:02:43.400 This is what the implementation of the slope looks like under the hood. A 00:02:43.400 --> 00:02:46.750 local pointer is initialized to the beginning of the array, and we 00:02:46.750 --> 00:02:50.650 increment this pointer until the address of the pointer matches the end 00:02:50.650 --> 00:02:53.260 of the array. And in each iteration, 00:02:53.260 --> 00:02:57.840 we de‑reference the pointer to get the value of the pointed‑to element, and 00:02:57.840 --> 00:03:01.150 we store that value inside of a placeholder variable. 00:03:01.840 --> 00:03:05.850 The name of this alias is defined here, and that is why we 00:03:05.850 --> 00:03:08.560 can use it as a normal integer variable. 00:03:09.540 --> 00:03:12.140 This is one important use case for iterators. 00:03:12.660 --> 00:03:16.760 Another one is utilizing functions from the algorithm library. 00:03:17.140 --> 00:03:21.960 We already saw how the copy function works. Now, let's pick another one. 00:03:22.340 --> 00:03:24.830 For example, let's use the sort function, 00:03:24.830 --> 00:03:27.160 which will sort the elements from the array. 00:03:28.040 --> 00:03:31.060 We need to pass both iterator functions to determine 00:03:31.060 --> 00:03:33.030 the range inside of the array. 00:03:33.040 --> 00:03:36.130 In this case, we want to work with the whole array, 00:03:36.130 --> 00:03:39.040 and if I compile the program again, 00:03:39.050 --> 00:03:44.040 you can see that elements are now sorted by value. Iterators 00:03:44.040 --> 00:03:47.660 provided an interface for this sort algorithm.