WEBVTT 00:00.210 --> 00:03.510 Hello again! In this video, we are going to look at the C++ map. 00:04.350 --> 00:08.160 This is another associative container. In the set, 00:08.520 --> 00:13.410 the elements just consisted of the key and nothing else. In a map, 00:13.530 --> 00:15.150 the elements actually have two parts. 00:15.810 --> 00:18.750 There is the key, and a so-called "value". 00:19.530 --> 00:22.540 These are stored as a pair object. 00:23.040 --> 00:24.990 And the first member of the pair is the key, 00:25.440 --> 00:28.170 and the second member is the value of the element. 00:29.760 --> 00:31.500 Again, the keys have to be unique. 00:32.070 --> 00:36.840 We cannot have two elements with the same key, but we can have two elements which have the same value, 00:37.200 --> 00:37.650 provided 00:37.650 --> 00:38.700 the keys are different. 00:39.150 --> 00:43.980 And again, they are stored in a tree, in order, using the less-than operator of the key. 00:47.610 --> 00:54.600 The value of the pair is used to store the data for that element, and the key is used to locate it. 00:55.230 --> 00:59.400 Normally there is some kind of relationship between the key and the value. 01:00.870 --> 01:07.760 For example, if we have a map of the company's staff, we could have the employee number as the key 01:08.250 --> 01:10.200 and their personnel record as the data. 01:11.370 --> 01:17.370 If we have a map of results from a sporting event, we could have the name of the competitor as the 01:17.370 --> 01:21.450 key, and their result in the event as the value. 01:22.830 --> 01:26.080 When we use a map, we search for a particular key. 01:26.100 --> 01:31.980 So one of these. And then we get an element, and we look up the data, from the value member of the element. 01:32.580 --> 01:34.800 So the key acts as an index for the map. 01:35.190 --> 01:39.090 This is very similar to a "hash map" or a "directory" in other languages. 01:43.920 --> 01:45.660 The structure will look like this. 01:45.990 --> 01:49.590 We have two parts to the element, the key and the value. 01:50.160 --> 01:58.170 So here we have the key, which is presumably a job level, or a pay grade, and the value is a job description. 01:58.890 --> 02:01.680 So these are all organized depending on the key. 02:02.820 --> 02:04.470 The value is not taken into account. 02:04.680 --> 02:07.860 So "Driver" is greater than "Analyst", but that does not matter. 02:08.900 --> 02:11.540 All that matters is that 42 is less than 67. 02:14.130 --> 02:20.640 As with set, we can use insert() and erase() for adding and removing elements. The element type is 02:20.640 --> 02:24.660 a pair, so we need to provide a pair object as the argument to insert. 02:25.530 --> 02:27.840 And typically we use make_pair(). 02:28.080 --> 02:31.770 The first argument to make_pair() is going to be the key of our new element. 02:32.250 --> 02:36.120 And the second member is going to be its value. In C++11, 02:36.540 --> 02:41.640 we can also use list initializers, so we can put the key and value inside a pair of braces. 02:42.930 --> 02:44.660 And again, the keys have to be unique. 02:44.670 --> 02:50.760 So if we try to insert an element, and there is already an element in the map which has the same key, 02:51.120 --> 02:57.900 then this will fail. And we check for that exactly the same way as we did in the set: we get a paper 02:57.900 --> 03:02.100 object returned. The second member is a bool, which will tell us if it succeeded. 03:03.660 --> 03:05.670 So we include the header. 03:06.720 --> 03:08.610 We create an empty map object. 03:09.900 --> 03:13.560 This is going to have a key of type string and a value of type int. 03:14.790 --> 03:16.590 And then we add some elements to it. 03:17.070 --> 03:19.710 We first call make_pair(), to show how to do that. 03:20.280 --> 03:27.870 So this is going to create a new element, with key "Maybelline" and value 86. And then using the list 03:27.870 --> 03:27.990 initializer. 03:28.380 --> 03:31.020 This is going to have the key "Graham" and the value 78. 03:31.530 --> 03:33.330 And then we can use this function to print them out. 03:33.330 --> 03:38.520 It takes a map by reference to const and then we have a range for loop. 03:39.480 --> 03:41.610 So, all pretty straightforward, I hope. 03:43.670 --> 03:46.330 And then we have the same logic as we did in the set example. 03:46.700 --> 03:51.020 We try to add another element, which has the same key as an element that is already in the map. 03:52.160 --> 03:55.040 We get a pair object returned from the call to insert(). 03:55.550 --> 03:57.950 The second member will tell us if it succeeded. 03:58.640 --> 04:05.120 And the first member will be an iterator to this element, which made it fail, or the new element if it succeeded. 04:06.710 --> 04:12.320 This is iterator is actually going to be an iterator to an object of the element type, which is a pair. 04:13.220 --> 04:15.590 So this is actually going to be a pair itself. 04:16.280 --> 04:19.370 And the first member is going to be the key of this element. 04:19.970 --> 04:22.640 And the second member is going to be the value of the element. 04:25.160 --> 04:27.920 Then we remove the element which has this key. 04:28.250 --> 04:31.700 And then we try to insert it again. And this time it should succeed. 04:34.720 --> 04:36.700 So there is our initial map elements. 04:37.030 --> 04:42.970 They are in alphabetical order, the order of the keys. "Graham" comes before "Maybelline". Then we try 04:42.970 --> 04:46.180 to add another element with key "Graham". The map already has one. 04:46.750 --> 04:48.340 So the map is not modified. 04:49.240 --> 04:54.190 Then we remove the element with key "Graham", and then we try to add our element again. 04:54.490 --> 04:58.630 So this time we have a new element, with key "Graham" and score 66. 05:03.660 --> 05:06.300 Map has a feature which the set does not. 05:06.660 --> 05:09.270 And that is subscript access to elements. 05:09.750 --> 05:13.960 However, it does work differently from the way that we are used to, with strings, vectors and arrays. 05:16.200 --> 05:21.240 So we put the name of the map, and then square brackets. And this will finish the element, which 05:21.240 --> 05:26.430 has this argument as its key, and it will return the value of that element. 05:26.820 --> 05:33.560 So this will return the value of the element with key "Graham". If we assign to this, then this will update 05:33.570 --> 05:35.150 the value of that element. 05:35.610 --> 05:37.140 So Graham now has the score 05:37.140 --> 05:37.830 66. 05:40.390 --> 05:44.890 However, if we put something in here, and there are no elements in the map which already have 05:44.890 --> 05:51.430 that argument as key, then this will create a new element, and this will be the key of that element. 05:51.850 --> 05:57.010 So this will now create a new element with key "Grace", and set its value to 66. 05:58.760 --> 06:04.190 So if the element does not already exist, it will be created. If the elements does exist, 06:04.460 --> 06:07.310 it willl overwrite the existing value of the element. 06:10.750 --> 06:12.100 So let's try this out. 06:12.400 --> 06:18.190 So we are starting off with the same map as we had before, with the same initial elements. Then we use 06:18.520 --> 06:21.730 subscripts, to access Graham's value. 06:22.600 --> 06:29.290 Then we set Graham's value to 66, and then we see what happens when we search Grace's score to 66. 06:32.720 --> 06:34.460 So those are the original elements. 06:35.690 --> 06:37.850 We get Graham's score, which is 78. 06:38.420 --> 06:40.720 Then we set Graham's score to 66. 06:40.730 --> 06:41.630 So it has done that. 06:42.410 --> 06:44.960 And then we add Grace with a score of 66. 06:45.530 --> 06:47.990 So we now have this new element for Greece. 06:48.590 --> 06:52.730 And you will see that the elements are still in alphabetical order. 06:53.450 --> 06:54.320 There are still ordered. 06:54.650 --> 06:58.730 And also that we can have two elements with the same value and different keys. 07:02.630 --> 07:06.500 We can call the find() and count() member functions, as we did with the set. 07:07.130 --> 07:14.540 So here we call find() to get Graham's score, and we call count() to get the number of "Graham"'s in 07:14.540 --> 07:15.290 the map. 07:16.980 --> 07:22.290 We saw with the set that we are not allowed to modify the returned element from the find(). 07:22.740 --> 07:24.270 So let's see if we can do this with map. 07:27.030 --> 07:28.500 And no, we cannot! 07:29.220 --> 07:30.320 Lots of compiler errors! 07:31.170 --> 07:33.870 But maybe it is just the key which is the problem? 07:35.400 --> 07:36.900 So let's see if we can change the value. 07:40.520 --> 07:44.840 Yeah, so you can do that, so we can change Graham's scores to 67. 07:45.410 --> 07:49.190 So we found an element with key "Graham", and we set its score to 67. 07:50.090 --> 07:51.530 And there is one element. 07:54.180 --> 07:59.130 So you can modify the value of a map element, but not the key. 08:00.480 --> 08:02.910 And with algorithms, we have the same constraint again. 08:03.660 --> 08:09.870 We can use algorithms provided they only reach each iterators for a map, and they do not try to modify them 08:09.870 --> 08:10.860 or rearrange them. 08:11.970 --> 08:18.790 So we have our call to find_if(). This is going to take an element object as argument. 08:18.820 --> 08:24.220 So this needs to be a pair, of string and int, because that is what the elements of the map are. 08:25.140 --> 08:29.760 And then we have our lambda expression, which will compare the key to "Graham". 08:31.920 --> 08:35.070 And similarly for count_if(). 08:37.970 --> 08:40.490 So we get the same results as we did with the member functions. 08:42.540 --> 08:48.870 And maps have the same strengths and weaknesses as sets. They can search or locate an element very quickly. 08:49.560 --> 08:54.870 Inserting and removing elements is usually very fast, unless you are unlucky and the tree has to rebalance. 08:55.740 --> 09:00.240 The main application for maps is for data, which has an index of some kind. 09:00.900 --> 09:06.540 For example, with a contact list, you could have the contact's name as their key, and then the data 09:06.540 --> 09:10.410 would be their phone number or email address or some other data. 09:11.340 --> 09:18.240 The employee records with the employee ID as the key and the record as the value. Results from a database 09:18.240 --> 09:18.650 query. 09:18.930 --> 09:24.690 Quite often, you have a row with an index, so you could use the index as the key and the rest of the 09:24.690 --> 09:26.400 row as the data. And so on. 09:28.150 --> 09:33.130 They are also useful for storing data which naturally falls into key-value pairs. 09:33.400 --> 09:35.290 So things like JSON and XML. 09:36.160 --> 09:37.600 Okay, so that is it for this video. 09:37.960 --> 09:40.750 I will see you next time, but until then, keep coding!