WEBVTT 00:04.160 --> 00:08.440 So now that we have the first piece of the puzzle, which is the playlist ID, we now need to get the 00:08.440 --> 00:10.040 list of unique video IDs. 00:10.240 --> 00:14.880 To do this, we will need to build another function which will be similar in structure, but there are 00:14.880 --> 00:21.280 some changes, the first change being the resource we will use before we were using the channels resource, 00:21.280 --> 00:24.680 which we see in the Russian doll analogy in the bottom left. 00:25.240 --> 00:29.560 But for the video IDs we will need to use the playlist item resource. 00:30.080 --> 00:35.560 Another point that we need to mention is the max results parameter you see in the API docs under Playlist 00:35.560 --> 00:36.600 Item Resource. 00:36.760 --> 00:42.320 This parameter represents the maximum number of videos the YouTube API shows per page. 00:43.040 --> 00:48.040 As you can see, the default value is five and the maximum is 50. 00:48.080 --> 00:53.240 You will see here in the API explorer that if we change this parameter from the default of 5 to 10, 00:53.240 --> 00:56.920 for example, then ten videos will be listed for each page. 00:57.280 --> 01:02.060 We will do a demonstration of this shortly, but before I will tell you why we are mentioning this parameter. 01:02.660 --> 01:04.700 We know that from the previous lectures. 01:04.740 --> 01:09.180 The total number of videos that Mr. Beast currently has is 840. 01:09.580 --> 01:13.940 So it won't be a matter of setting this parameter to eight photo. 01:14.180 --> 01:17.420 And we will do just one API call and get all the videos. 01:18.100 --> 01:24.740 What we can do is set this parameter to the maximum of 50 and loop to each page, effectively making 01:24.740 --> 01:30.580 an API call each time until we have looped through each video that Mr. Beast has on his channel. 01:30.900 --> 01:36.060 Now let's do a quick demonstration, starting off by building the URL using the API explorer. 01:36.260 --> 01:39.700 I have pre-filled the parameters, but we will go over them quickly. 01:40.020 --> 01:44.980 For parts we use again the content detail scheme as this will contain the video ID key. 01:45.500 --> 01:50.620 And for the playlist id we found this using the previous function, which is this parameter. 01:51.940 --> 01:53.180 Now for max results. 01:53.180 --> 01:56.060 If we leave it as a blank, we know that five will be returned. 01:56.460 --> 02:00.820 So let's first try setting into one which should return one video. 02:01.720 --> 02:05.440 Let's copy the URL and paste it into postman. 02:09.480 --> 02:13.600 This time we can use postman inside vs code, so let me pull that up. 02:13.920 --> 02:19.840 So the extension, you can actually get it from the extensions and you can write it in postman. 02:21.680 --> 02:24.840 Make sure you install the official postman extension. 02:26.120 --> 02:29.520 Going back here we do a new HTTP request. 02:32.200 --> 02:33.640 Make sure it's a Get request. 02:34.880 --> 02:40.800 We paste the URL here and we just need to change the API key here. 02:41.000 --> 02:44.840 So this will get it from the dot env. 02:47.720 --> 02:50.560 Let's paste this here and hit send. 02:50.880 --> 02:52.440 So this is the response. 02:52.560 --> 02:55.800 Let's just switch the view a bit so it's more clear. 02:56.280 --> 03:01.860 And here you can see that one video was returned with the video ID of the video. 03:02.540 --> 03:08.700 If we had to change this max results to ten, then ten videos will be listed up to a maximum of 50. 03:09.500 --> 03:13.380 Actually lets try to give this a number greater than 50. 03:13.420 --> 03:16.380 So lets say 60 and see what happens. 03:16.860 --> 03:17.980 So I can press send. 03:20.340 --> 03:23.180 So our response was still given. 03:23.940 --> 03:28.180 But maybe if we scroll down here we see that the results per page is 50. 03:28.500 --> 03:34.780 So even though we set max result to 60, the maximum of 50 is still being shown in this page. 03:34.980 --> 03:39.340 So now that we have seen this in action, lets apply these changes to the Python script. 03:40.540 --> 03:42.340 Lets go back to video stats. 03:42.820 --> 03:46.700 First we can set the global variable max results to 50. 03:47.580 --> 03:51.380 We can set it here max results equals 50. 03:51.820 --> 03:56.860 We can also copy and paste the URL from postman and give it the variable name base URL. 03:57.780 --> 03:58.820 Lets do that right now. 04:02.440 --> 04:06.080 This would be the first function and we will focus on the second function. 04:06.400 --> 04:07.880 But for now paste it here. 04:12.640 --> 04:17.400 Now let's substitute some of the variables so we don't have the values hard coded in the URL. 04:17.840 --> 04:19.760 Let's use f strings. 04:20.520 --> 04:24.040 We can keep the part as the quantum details. 04:24.520 --> 04:25.640 Mixed results. 04:25.840 --> 04:29.960 We can replace with the max results variable of 50. 04:31.720 --> 04:33.960 We can do that like that. 04:34.480 --> 04:37.040 The playlist ID we get from the previous function. 04:37.040 --> 04:42.840 So what we can do for now is to create a variable playlist id and set it equal to the function. 04:48.840 --> 04:54.200 The last change that we need to do is to replace the API key, as we did in the previous URL. 04:55.600 --> 04:57.760 So let's change this API key. 04:58.800 --> 05:01.280 And this is the base URL which we will be using. 05:01.620 --> 05:07.380 Going back to the API explorer, there is this page token parameter, which for the first page of results 05:07.380 --> 05:08.580 we don't need to fill in. 05:09.460 --> 05:12.700 If we have to press execute, we have to switch this off. 05:13.340 --> 05:17.220 Press execute and we get a response. 05:18.020 --> 05:23.420 You will see in this next page token parameter, which refers to the token request parameter that you 05:23.420 --> 05:24.100 see here. 05:25.100 --> 05:31.660 If we had to copy the next page token into page token, we would get the next 50 videos of Mr. Beast's 05:31.660 --> 05:32.180 channel. 05:32.860 --> 05:34.060 Let's do that right now. 05:34.060 --> 05:39.540 So we copy the next page token and paste it into the parameter. 05:39.780 --> 05:44.580 We press execute and now you see the response of the next page. 05:44.780 --> 05:50.020 Here you also see the previous page token which relates to the first videos that came up when we use 05:50.020 --> 05:50.660 postman. 05:51.500 --> 05:55.740 So here we are starting to see that to get the next batch of videos, we need to have the associated 05:55.740 --> 05:57.820 page token for that video batch. 05:58.060 --> 06:03.480 Before we move on to the next step, let me remind you that we are after the video IDs, which we know 06:03.480 --> 06:08.560 is under the content details, and we have seen before and we know it's this one. 06:08.720 --> 06:10.400 So we are interested in this video ID. 06:10.640 --> 06:12.960 So now the question is how will we code this up. 06:13.280 --> 06:17.640 I would say the only tricky part of coding this logic up is to keep in mind that for the first batch 06:17.640 --> 06:21.760 of 50 videos, we don't need to define a page token argument in the URL. 06:22.160 --> 06:27.560 However, in the next iterations we will need to and we will get the page token using the next page 06:27.560 --> 06:29.680 token key which we just went over. 06:30.280 --> 06:32.560 Knowing this, let's continue building the syntax. 06:32.560 --> 06:37.560 And we can do this by defining the second function which we can call get video IDs. 06:37.680 --> 06:41.600 Let's go back to the script and create a function. 06:43.520 --> 06:46.680 Get video IDs. 06:47.600 --> 06:52.600 Now as argument to this function, we will specify the playlist id variable which we get from the first 06:52.600 --> 06:53.200 function. 06:54.240 --> 06:59.000 So let's do some cleaning by first defining the playlist ID variable and moving the definition of this 06:59.000 --> 07:04.460 variable into the double underscore name Rename it with double underscore main part of the script. 07:07.740 --> 07:10.580 And this we can move it to here. 07:14.540 --> 07:15.660 Since we don't need it anymore. 07:17.420 --> 07:20.340 And thus we can also move inside the function. 07:25.220 --> 07:29.140 Now that this is done, let's also do some initialization of some variables. 07:29.780 --> 07:33.860 Let's define a list with the variable name video IDs. 07:36.220 --> 07:40.100 And this will contain all the video IDs of the channel. 07:40.220 --> 07:42.540 For now we will initialize it as an empty list. 07:42.740 --> 07:46.660 Also, we should define a page token variable and initialize it to null. 07:51.540 --> 07:54.060 This will change as we loop through each batch of videos. 07:54.220 --> 08:01.260 We will now encapsulate all of this code in a try except clause, same as we did in the previous function. 08:03.290 --> 08:08.810 Requests dot exceptions dot requests exception. 08:10.850 --> 08:13.650 Then we raise the error if there is one. 08:13.770 --> 08:17.690 Now we need to start a loop that will break only when a certain condition is met. 08:17.850 --> 08:22.610 In our case, this condition will be when there are no page tokens left to go through. 08:23.170 --> 08:26.810 At this point, we will have all the video IDs in our list. 08:27.730 --> 08:31.850 A while true condition will take care of this looping until the condition is satisfied. 08:32.530 --> 08:37.890 In our case, we will know that the loop is broken when in the API response, there won't be a next 08:37.890 --> 08:38.850 page token key. 08:39.650 --> 08:40.810 So inside the API. 08:41.530 --> 08:45.890 Let's write the while true clause. 08:49.210 --> 08:55.050 Now, while the condition is true, let's create a URL variable that is based on the base URL. 08:55.530 --> 08:59.650 So we can say URL is equal to the base URL. 08:59.810 --> 09:04.550 And now let's define a condition in which for the first iteration will be false with regards to the 09:04.550 --> 09:08.270 page token variable, since it is initially set to none. 09:08.470 --> 09:09.950 So now we can write. 09:10.230 --> 09:19.950 If each token, the URL will append the page token part to the main URL. 09:24.630 --> 09:31.590 What this part does is it says that if this is true meaning page token has a non-none value, then to 09:31.630 --> 09:36.750 the URL variable you need to append the page token string inside this f string. 09:36.790 --> 09:41.510 Again, I will repeat this statement is false in the first iteration as we have initialized page token 09:41.510 --> 09:42.030 to none. 09:42.350 --> 09:47.030 So the appending of the URL will not happen and the URL will be simply the base URL. 09:47.510 --> 09:53.510 As soon as the previous function, we get the response check for errors using the raise for status and 09:53.510 --> 09:56.990 if there are no errors, get the response data in JSON format. 09:57.670 --> 09:59.070 So let's write all of this. 09:59.750 --> 10:02.970 Actually we can get from the previous function. 10:03.410 --> 10:05.290 So let's copy all of this. 10:08.410 --> 10:10.490 And paste it here. 10:10.530 --> 10:12.690 As we have seen in the API explorer. 10:12.970 --> 10:17.770 We know that our variable of interest, which is the video ID is in the item scheme. 10:17.770 --> 10:22.890 And if we drill down the JSON file, we need to go through content details to find the get the video 10:22.930 --> 10:24.050 ID value. 10:24.090 --> 10:29.610 To manipulate this response in Python we use dictionaries, and here we can use the get method to get 10:29.610 --> 10:31.210 the value from the items key. 10:31.330 --> 10:36.450 Using a for loop, we can therefore loop through and extract the video IDs and their content details, 10:36.850 --> 10:40.690 and while doing so, appending to the initially empty video ID list. 10:40.850 --> 10:43.170 If we had to write this down, we would have this. 10:43.570 --> 10:53.530 So starting the for loop for item in data dot get, we are going to loop through the items key as a 10:53.530 --> 10:55.130 failsafe with the empty list. 10:55.130 --> 10:57.250 I will soon explain why we will use it. 10:58.170 --> 11:01.630 Now the video ID will be equal to item Tim. 11:02.470 --> 11:10.950 We're going to pass through contents, details, key and the video ID will be in the video ID final 11:10.950 --> 11:11.270 key. 11:13.590 --> 11:21.630 Finally, we will append to the initially empty video IDs list and we will append the video ID inside 11:21.630 --> 11:22.110 the loop. 11:22.630 --> 11:26.870 Now we have used this way of writing the get methods with the empty list. 11:27.070 --> 11:30.110 This one over here because there's a good reason for it. 11:30.590 --> 11:37.350 It prevents the program from crashing if the key items is not found in the data dictionary, and also 11:37.350 --> 11:41.310 it ensures that we always have a list to work with, even if it's empty. 11:41.830 --> 11:44.270 This makes the code more robust at this point. 11:44.270 --> 11:49.430 If we were to run the code, we would extract the first batch of 50 videos from the first page, but 11:49.430 --> 11:50.430 it would stop there. 11:51.030 --> 11:56.070 We now need to specify the next page token where we will get the next batch of videos. 11:56.110 --> 12:01.390 We do this by using the get command, this time specifying the next page token parameter. 12:02.450 --> 12:06.290 So let's continue the code from here to define the page token. 12:07.450 --> 12:11.930 And we'll set it equal to the page token variable. 12:17.130 --> 12:19.850 Finally we just need to break out of the infinite loop. 12:19.850 --> 12:23.010 If there is no next page token this will be in the last page. 12:23.410 --> 12:27.610 And we can do this by writing if not page token. 12:28.130 --> 12:30.810 Which means if it doesn't find it, then we break the loop. 12:31.450 --> 12:38.050 Now, the very last thing that we need to do is to return the video IDs final list. 12:38.530 --> 12:41.850 So we write return video IDs. 12:42.050 --> 12:47.490 Looking at our function, something that I am now just realizing is that in the base URL, we didn't 12:47.490 --> 12:51.730 apply the change of replacing the hardcoded playlist ID. 12:52.290 --> 12:55.210 As you can see, it is still hardcoded. 12:55.730 --> 12:57.450 So let's apply the change right now. 12:57.650 --> 13:05.670 This can easily be done by replacing this with the playlist ID curly brackets, and this playlist ID 13:05.670 --> 13:08.270 is referring the input argument of the function. 13:08.670 --> 13:13.870 Now, in order to execute the function, we need to write it in the double underscore name equals double 13:13.870 --> 13:15.950 underscore main part of the script. 13:17.950 --> 13:23.190 Here we have already defined the playlist id variable, which takes the output from the get playlist 13:23.190 --> 13:24.430 id function. 13:24.430 --> 13:32.630 So the only change that we need to do is to simply write the function name to just defined, referencing 13:32.670 --> 13:34.590 the playlist ID variable. 13:35.950 --> 13:41.830 For the purpose of this demonstration, I will also add a print to showcase the output in the terminal. 13:43.830 --> 13:49.910 So now what is left is to press run and we should get the full list of video IDs in list format. 13:51.870 --> 13:56.430 There appears to be an error and here we are seeing it is something in the URL. 13:58.270 --> 14:08.370 So let's look at the URL URL we be copied from the API explorer, so it should be okay and the change 14:08.370 --> 14:15.250 with it for the playlist id appears to be okay, so maybe there is somewhere here I guess. 14:15.290 --> 14:19.650 So what I forgot to include was the end, which is this one here. 14:19.690 --> 14:26.610 Notice that when we are defining a new variable, we use the At to append this. 14:26.610 --> 14:29.050 You can see here here and even for the key. 14:29.770 --> 14:31.970 So let's include the end. 14:32.450 --> 14:35.730 And hopefully with this change the code will now run. 14:35.930 --> 14:39.170 Let's click save and press run again. 14:43.330 --> 14:44.450 Seems to be executing. 14:44.450 --> 14:45.570 And there we have it. 14:46.330 --> 14:49.410 These are all the video IDs from the Mr. Beast YouTube channel. 14:49.730 --> 14:56.010 So now what I will do is I will remove the print, save and commit this change to GitHub. 14:56.010 --> 14:59.210 And I will see you in part three of the video stats script.