WEBVTT 0 00:00.720 --> 00:03.630 To create a useful rain alert application. 1 00:03.630 --> 00:04.920 We want our script 2 00:04.920 --> 00:08.640 to detect if it will rain within the next 12 hours 3 00:08.640 --> 00:11.130 of the Python script being run. 4 00:11.130 --> 00:13.680 Let's say every day you leave your home at 8:00 AM 5 00:13.680 --> 00:15.150 in order to go to work, 6 00:15.150 --> 00:18.990 then by 5:00 or 6:00 PM you might be done with work 7 00:18.990 --> 00:21.240 and on your way back home. 8 00:21.240 --> 00:24.000 Given that the free OpenWeatherMap API 9 00:24.000 --> 00:27.990 gives us the weather forecast in three hour intervals, 10 00:27.990 --> 00:30.120 we can run the script at 6:00 AM 11 00:30.120 --> 00:33.240 and check the weather for the next 12 hours. 12 00:33.240 --> 00:34.560 If there's rain forecast 13 00:34.560 --> 00:36.690 during the time you're away from home, 14 00:36.690 --> 00:39.630 you want your Python script to send you a text message, 15 00:39.630 --> 00:42.060 reminding you to bring an umbrella. 16 00:42.060 --> 00:45.270 That way you'll get notified before you leave home 17 00:45.270 --> 00:47.133 and know how to prepare for the day. 18 00:48.450 --> 00:50.700 In the previous lesson, we saw that the data 19 00:50.700 --> 00:53.010 that we got back from OpenWeatherMap 20 00:53.010 --> 00:55.830 contains the weather forecast every three hours 21 00:55.830 --> 00:57.720 for the next five days. 22 00:57.720 --> 01:00.960 There are forecasts for 40 different timestamps, 23 01:00.960 --> 01:04.950 starting from zero and going up all the way to 39. 24 01:04.950 --> 01:07.650 Now, I don't know about you, but my decision 25 01:07.650 --> 01:09.420 whether or not to bring an umbrella 26 01:09.420 --> 01:11.640 really only depends on the forecast 27 01:11.640 --> 01:13.530 for the next 12 hours. 28 01:13.530 --> 01:16.410 In the JSON that we got back from OpenWeatherMap, 29 01:16.410 --> 01:20.250 that would be only the first four forecasts in the list. 30 01:20.250 --> 01:22.410 If the script is run at 6:00 AM 31 01:22.410 --> 01:26.520 then a 12-hour window takes us up to 6:00 PM 32 01:26.520 --> 01:29.430 by which time, we're hopefully ready to head home 33 01:29.430 --> 01:30.480 and we don't need to worry 34 01:30.480 --> 01:32.793 if it starts raining much later than that. 35 01:33.690 --> 01:36.180 If we look at this JSON that we got back, 36 01:36.180 --> 01:38.550 we don't actually care about the forecasts 37 01:38.550 --> 01:42.300 that are for tomorrow or those even further in the future. 38 01:42.300 --> 01:43.650 We just care about the weather 39 01:43.650 --> 01:46.290 within the next four forecasts. 40 01:46.290 --> 01:49.260 Looking at the API documentation more closely, 41 01:49.260 --> 01:51.330 we can actually spot parameters 42 01:51.330 --> 01:55.140 that will help us to tailor OpenWeatherMaps response, 43 01:55.140 --> 01:56.850 so that we just get back the data 44 01:56.850 --> 01:58.590 that we actually care about. 45 01:58.590 --> 02:03.590 Of course, I'm talking about the cnt or count parameter. 46 02:03.660 --> 02:06.330 By setting the count to four, 47 02:06.330 --> 02:09.090 we'll request only four timestamps 48 02:09.090 --> 02:12.540 covering the 12-hour window that matters to us. 49 02:12.540 --> 02:14.820 That way we make our own lives easier 50 02:14.820 --> 02:18.240 and we avoid requesting more data than we actually need. 51 02:18.240 --> 02:20.550 That's why it's always a good idea to have a read 52 02:20.550 --> 02:24.000 through the API documentation first. 53 02:24.000 --> 02:25.950 There are usually some useful nuggets in there 54 02:25.950 --> 02:29.700 that will make your life as a developer much, much easier. 55 02:29.700 --> 02:31.530 As I always say, this is like going 56 02:31.530 --> 02:33.120 into somebody else's house, 57 02:33.120 --> 02:34.260 everything is different. 58 02:34.260 --> 02:36.780 The way that the washing machine works is different, 59 02:36.780 --> 02:39.060 the way that their dryer works is different, 60 02:39.060 --> 02:40.440 so you really have 61 02:40.440 --> 02:43.590 to look at the API documentation when you are working 62 02:43.590 --> 02:46.380 with a new API, just so that you're not caught out 63 02:46.380 --> 02:49.170 and you understand exactly what you have to do 64 02:49.170 --> 02:52.053 in order to be a good API user. 65 02:53.370 --> 02:55.890 Now, let's go back into our Python code 66 02:55.890 --> 02:58.350 and provide that extra parameter. 67 02:58.350 --> 03:00.240 The parameter was called count, 68 03:00.240 --> 03:04.110 and I'll just add it to our weather_params dictionary. 69 03:04.110 --> 03:05.730 Let's set cnt to 4, 70 03:05.730 --> 03:09.630 so that we request only the timestamps in the near future. 71 03:09.630 --> 03:14.490 Let's run it again and take a look at the JSON response 72 03:14.490 --> 03:15.810 we get back. 73 03:15.810 --> 03:17.190 I'm going to copy it again, 74 03:17.190 --> 03:19.590 go back to the online JSON viewer, 75 03:19.590 --> 03:22.530 clear that huge amount of data from before 76 03:22.530 --> 03:24.750 and paste in what I just got back. 77 03:24.750 --> 03:26.880 You can already see it's much shorter, 78 03:26.880 --> 03:29.580 and if I click on Viewer, we can see that our list 79 03:29.580 --> 03:31.920 of weather forecasts now only contain 80 03:31.920 --> 03:33.993 the next four timestamps. 81 03:35.070 --> 03:36.480 So that's the goal. 82 03:36.480 --> 03:39.360 And if we head back into our code, 83 03:39.360 --> 03:42.180 we know that at the moment the response code 84 03:42.180 --> 03:45.900 that we're getting back from calling this API is 200. 85 03:45.900 --> 03:47.520 So it's successful. 86 03:47.520 --> 03:48.750 But we want to make sure 87 03:48.750 --> 03:51.900 that we actually catch other response codes. 88 03:51.900 --> 03:54.750 So we're going to call raise_for_status(), 89 03:54.750 --> 03:57.150 so that if there is a problem 90 03:57.150 --> 03:59.220 and we don't get a 200 code, 91 03:59.220 --> 04:01.293 then we actually raise an exception. 92 04:02.400 --> 04:06.450 Now, next, we want to save our response.json() 93 04:06.450 --> 04:07.840 as the weather_data 94 04:09.480 --> 04:11.040 and then we're going to work with this data 95 04:11.040 --> 04:13.380 in order to get the particular pieces 96 04:13.380 --> 04:14.630 that we're interested in. 97 04:15.600 --> 04:17.520 So that's step one done. 98 04:17.520 --> 04:21.900 The next step is to dig through this hourly forecast 99 04:21.900 --> 04:24.750 and get hold of the thing that we're interested in, 100 04:24.750 --> 04:28.170 which is the actual weather condition. 101 04:28.170 --> 04:30.390 The way that weather services tend 102 04:30.390 --> 04:34.290 to provide the weather condition is through an id. 103 04:34.290 --> 04:38.730 And I know this because I read the API documentation. 104 04:38.730 --> 04:42.630 It's not because I'm some sort of weather geek. (laughs) 105 04:42.630 --> 04:45.930 Although, I mean that's not a bad thing to be I guess. 106 04:45.930 --> 04:50.930 So if we scroll down in this documentation past the examples, 107 04:51.000 --> 04:53.010 you can see it provides all of the fields 108 04:53.010 --> 04:54.600 in the API response. 109 04:54.600 --> 04:56.280 So these are all of the things 110 04:56.280 --> 04:59.760 that we could possibly get back and what they mean. 111 04:59.760 --> 05:01.920 And you can see there's some really interesting things, 112 05:01.920 --> 05:05.250 like what does the temperature feel like based 113 05:05.250 --> 05:08.850 on the wind chill and actual ground temperature. 114 05:08.850 --> 05:13.020 But what we're mostly interested in is the weather data. 115 05:13.020 --> 05:18.020 So this weather.id is a weather condition id, 116 05:18.480 --> 05:21.660 and when you click on that link, it takes you to this table 117 05:21.660 --> 05:24.750 that shows you all of the weather condition codes 118 05:24.750 --> 05:29.103 that we could possibly get back in this particular field. 119 05:30.030 --> 05:31.860 Now, you can see that all the codes 120 05:31.860 --> 05:36.860 that start off with 2, means some sort of thunderstorm, 121 05:37.230 --> 05:38.730 and then starting with 3, 122 05:38.730 --> 05:40.380 that means some sort of drizzling, 123 05:40.380 --> 05:42.450 starting with 5 means rain, 124 05:42.450 --> 05:44.550 starting with 6 means snow, 125 05:44.550 --> 05:47.460 and then afterwards we have the seven hundreds. 126 05:47.460 --> 05:50.340 So these are atmospheric things like a bit of mist, 127 05:50.340 --> 05:53.040 a bit of smoke, a bit of dust or fog. 128 05:53.040 --> 05:55.920 And this is also incidentally the reason why 129 05:55.920 --> 06:00.000 this weather key actually has a value 130 06:00.000 --> 06:02.070 that's in the form of a list. 131 06:02.070 --> 06:05.250 You can see that's denoted by the square brackets here. 132 06:05.250 --> 06:08.730 So there could actually be multiple weather conditions 133 06:08.730 --> 06:12.180 for a particular place at a particular hour, 134 06:12.180 --> 06:16.020 and that's because you could maybe have snow, 135 06:16.020 --> 06:18.753 but you could also have fog at the same time. 136 06:19.740 --> 06:22.260 Now, when I looked through a lot of the examples 137 06:22.260 --> 06:26.010 and the documentation, it seems like the first item 138 06:26.010 --> 06:28.710 in that list is the main condition. 139 06:28.710 --> 06:30.030 So if it's going to rain, 140 06:30.030 --> 06:32.160 then it's going to be in that first item 141 06:32.160 --> 06:35.070 in the list of weather conditions. 142 06:35.070 --> 06:39.630 Inside that list, we have a dictionary or many dictionaries. 143 06:39.630 --> 06:42.870 Each of those contain a weather condition id, 144 06:42.870 --> 06:46.260 the main condition name, and the description. 145 06:46.260 --> 06:49.410 So if we look at this id code 802, 146 06:49.410 --> 06:51.780 we can decode it in this table 147 06:51.780 --> 06:54.540 and you can see it means scattered clouds. 148 06:54.540 --> 06:57.903 20-50% of the sky is covered in clouds, basically. 149 06:59.370 --> 07:01.500 Based on this list, we can say that, 150 07:01.500 --> 07:05.550 well, anything that has a code less than 700, 151 07:05.550 --> 07:08.730 then we probably will need an umbrella. 152 07:08.730 --> 07:11.400 I'm not sure how you stand on the umbrella 153 07:11.400 --> 07:15.180 in snow situation, but I personally do like 154 07:15.180 --> 07:17.130 to hold an umbrella when it's snowing, 155 07:17.130 --> 07:18.780 especially because I live in a country 156 07:18.780 --> 07:22.130 where the snow is not crazy, it's just sort of... 157 07:23.220 --> 07:26.670 It's never sort of the beautiful snow 158 07:26.670 --> 07:29.310 where it's thick and it gets caught on your eyelashes. 159 07:29.310 --> 07:33.810 It's the sort of annoying slush that's just sort of snow, 160 07:33.810 --> 07:37.380 but it's sort of like somebody spitting at you. 161 07:37.380 --> 07:38.940 In my case, I would prefer 162 07:38.940 --> 07:41.820 to have an umbrella if it was going to snow. 163 07:41.820 --> 07:44.040 I'm going to check for the codes 164 07:44.040 --> 07:47.010 that we get back from the OpenWeatherMap API, 165 07:47.010 --> 07:49.890 and if the code is less than 700, 166 07:49.890 --> 07:53.553 then I'm going to advise my user to bring an umbrella. 167 07:55.290 --> 07:57.690 Here's a challenge for you. 168 07:57.690 --> 08:00.150 Can you figure out how to look 169 08:00.150 --> 08:01.800 through the data that we get back, 170 08:01.800 --> 08:05.073 which remember looks something like this, 171 08:05.910 --> 08:08.910 and then look at the weather 172 08:08.910 --> 08:13.410 and the first item in the weather list and also the id? 173 08:13.410 --> 08:16.800 So you're going to go all the way down to this, 174 08:16.800 --> 08:19.830 but remember you're going to do that for all of the items 175 08:19.830 --> 08:22.410 and you're going to check for the condition code. 176 08:22.410 --> 08:26.700 Now if any of those id codes are less than 700, 177 08:26.700 --> 08:30.330 then you want to be able to print out, "Bring an umbrella". 178 08:30.330 --> 08:32.340 So that's the goal 179 08:32.340 --> 08:35.460 and the actual implementation, I'll leave up to you 180 08:35.460 --> 08:37.800 'cause there's quite a few ways that you can do this, 181 08:37.800 --> 08:40.710 but I'm sure by now, you are well prepared 182 08:40.710 --> 08:44.400 with all of your tools in Python to figure this out. 183 08:44.400 --> 08:48.180 If you want to make sure that your code is actually working, 184 08:48.180 --> 08:50.730 you can switch to a latitude and longitude 185 08:50.730 --> 08:52.860 that is definitely raining. 186 08:52.860 --> 08:56.130 So if you go to ventusky.com, 187 08:56.130 --> 08:59.340 they actually show you the live weather forecast. 188 08:59.340 --> 09:03.270 So we can look at precipitation, which is basically rain, 189 09:03.270 --> 09:06.000 and we can find some sort of unfortunate place, 190 09:06.000 --> 09:09.060 which seems to be really heavily raining. 191 09:09.060 --> 09:11.313 For example, this place in Poland, Lodz, 192 09:12.270 --> 09:16.145 and if I spell that name correctly. 193 09:16.145 --> 09:19.200 (keyboard typing) 194 09:19.200 --> 09:22.560 Lodz and then I'll add the country code Poland, 195 09:22.560 --> 09:24.750 and just to make sure in the map, 196 09:24.750 --> 09:27.510 it's actually found the correct place. 197 09:27.510 --> 09:29.280 Yet that looks pretty much it, 198 09:29.280 --> 09:32.310 then I can switch out my latitude and longitude 199 09:32.310 --> 09:34.260 with this rainy place, 200 09:34.260 --> 09:37.470 and that way I know that at least one of the results 201 09:37.470 --> 09:39.480 I get back is definitely going 202 09:39.480 --> 09:41.463 to contain some sort of rain. 203 09:44.400 --> 09:46.620 Now if we look at their hourly data, 204 09:46.620 --> 09:48.450 you can see that the weather 205 09:48.450 --> 09:51.780 is basically just rain, rain, rain. 206 09:51.780 --> 09:55.020 So between looking at places which are sunny 207 09:55.020 --> 09:58.020 and places which are rainy, you should be able 208 09:58.020 --> 09:59.370 to get your code to work, 209 09:59.370 --> 10:02.610 so that it tells you in the next 12 hours, 210 10:02.610 --> 10:07.350 if any of those condition codes are less than 700, 211 10:07.350 --> 10:10.263 which means it's got some form of precipitation. 212 10:11.310 --> 10:13.173 Pause the video and give this a go. 213 10:18.300 --> 10:22.290 All right, so let's narrow down into this weather. 214 10:22.290 --> 10:24.090 If we want to get the weather, 215 10:24.090 --> 10:28.320 then we have to tap into the first item in our list 216 10:28.320 --> 10:30.030 and then we get hold of the weather 217 10:30.030 --> 10:32.010 and then we get hold of the first item 218 10:32.010 --> 10:37.010 in that list. Under the id key, we'll get the actual value. 219 10:37.050 --> 10:40.050 We've talked about this in detail in previous lessons. 220 10:40.050 --> 10:42.030 If you've skipped a lot of lessons 221 10:42.030 --> 10:43.860 and you've come here directly, 222 10:43.860 --> 10:46.050 then it's going to be a little bit confusing 223 10:46.050 --> 10:48.630 and I recommend to review the previous lessons 224 10:48.630 --> 10:50.580 before you continue. 225 10:50.580 --> 10:53.820 Instead of printing the weather data, let's drill down. 226 10:53.820 --> 10:56.250 Let's get to the first item here. 227 10:56.250 --> 10:58.650 If we provide a set of square brackets 228 10:58.650 --> 11:00.630 and then we can access the value 229 11:00.630 --> 11:04.500 inside the list key, like this. 230 11:04.500 --> 11:08.910 And now, we've got a list with all of the hourly data 231 11:08.910 --> 11:11.613 and it looks pretty much like this. 232 11:12.780 --> 11:14.250 Now, you can confirm this 233 11:14.250 --> 11:17.790 by also pasting this into the JSON Viewer, 234 11:17.790 --> 11:20.460 replacing the previous text that was there. 235 11:20.460 --> 11:24.330 You can see we now have our items in this list. 236 11:24.330 --> 11:25.650 To drill further down, 237 11:25.650 --> 11:28.260 let's get hold of the first item in that list 238 11:28.260 --> 11:30.960 by providing a square bracket 239 11:30.960 --> 11:33.990 and then the index, which is zero. 240 11:33.990 --> 11:36.660 Now, we are into the first item 241 11:36.660 --> 11:39.120 and this is what the data looks like. 242 11:39.120 --> 11:40.980 It's a lot shorter. 243 11:40.980 --> 11:44.820 Now, we want to tap into the weather condition, 244 11:44.820 --> 11:47.340 so that means yet another set of square brackets 245 11:47.340 --> 11:50.823 and then the name of the key, which is weather. 246 11:52.560 --> 11:53.730 Now, it's pretty simple. 247 11:53.730 --> 11:58.470 It's simply giving us a list with only one item. 248 11:58.470 --> 12:02.820 Let's tap into that one item by using again, 249 12:02.820 --> 12:05.070 square brackets, zero. 250 12:05.070 --> 12:10.070 And now, we've got just a simple dictionary, essentially. 251 12:10.470 --> 12:12.330 So if we want to get hold of the id, 252 12:12.330 --> 12:15.570 then it's the final square bracket 253 12:15.570 --> 12:18.450 and the key, which is id. 254 12:18.450 --> 12:21.690 Now, what if I created a new list 255 12:21.690 --> 12:24.600 that contains the condition codes? 256 12:24.600 --> 12:26.610 We know that we've got a list 257 12:26.610 --> 12:31.470 of all of the weather conditions for the next 12 hours. 258 12:31.470 --> 12:36.470 How can we loop through that list to find out the actual id 259 12:36.780 --> 12:39.873 of the weather condition for each of those hours? 260 12:40.770 --> 12:42.570 Well, we can create a for loop 261 12:42.570 --> 12:46.590 that looks at each of the data from the hour. 262 12:46.590 --> 12:50.640 So, 'for hour_data in weather_data["list"]', 263 12:50.640 --> 12:53.070 so this is going to go through the 12 hours. 264 12:53.070 --> 12:56.220 And then for each, we're going to tap into the hour data 265 12:56.220 --> 12:58.830 and try to get hold of the item 266 12:58.830 --> 13:01.653 that is in the weather key, 267 13:03.510 --> 13:04.413 like this. 268 13:05.370 --> 13:09.180 Now, I'm going to print each of these 269 13:09.180 --> 13:12.360 and you can see, we've now got a list 270 13:12.360 --> 13:17.360 of all of the weather conditions for each of those hours. 271 13:17.760 --> 13:20.790 Now, let's narrow down that a little bit further. 272 13:20.790 --> 13:24.300 Let's get hold of the first item of each of these lists. 273 13:24.300 --> 13:26.580 And you can see, in fact, there's only one item 274 13:26.580 --> 13:27.930 in all of these lists. 275 13:27.930 --> 13:31.200 It's very rare that you have multiple weather conditions 276 13:31.200 --> 13:32.463 for each hour. 277 13:33.420 --> 13:36.420 Now that gives us a Python dictionary 278 13:36.420 --> 13:38.100 for each of those hours 279 13:38.100 --> 13:41.310 and we can tap into that final value we're interested in 280 13:41.310 --> 13:43.800 under the key id. 281 13:43.800 --> 13:47.760 So now, we've looped through the next 12 hours 282 13:47.760 --> 13:51.810 and got hold of the weather condition id 283 13:51.810 --> 13:53.223 for each of those hours. 284 13:54.120 --> 13:57.870 Now, all we need to do is to save this, 285 13:57.870 --> 14:02.870 instead of printing it, and we'll call it the condition_code. 286 14:02.871 --> 14:06.270 (keyboard typing) 287 14:06.270 --> 14:07.710 And remember previously we said 288 14:07.710 --> 14:11.340 that if the condition code is less than 700, 289 14:11.340 --> 14:15.000 then we're going to print, "Bring an umbrella". 290 14:15.000 --> 14:17.940 So we can check if condition_code, 291 14:17.940 --> 14:21.540 which remember at this point is still a string. 292 14:21.540 --> 14:25.350 So we have to turn that into an integer 293 14:25.350 --> 14:29.010 in order to be able to compare it against another number. 294 14:29.010 --> 14:33.570 If that condition_code as an integer is less than 700, 295 14:33.570 --> 14:37.173 then we're going to print, "Bring an umbrella". 296 14:39.510 --> 14:42.420 And you can see that for the next 12 hours, 297 14:42.420 --> 14:46.470 there's rain at this particular place in Lodz in Poland 298 14:46.470 --> 14:49.860 that I've put into the latitude and longitude. 299 14:49.860 --> 14:51.270 Now, if I don't want 300 14:51.270 --> 14:54.720 to call this print statement many times, 301 14:54.720 --> 14:57.718 then we can define a variable outside the for loop 302 14:57.718 --> 14:59.135 called will_rain, 303 15:00.000 --> 15:03.180 and we can set that to false to begin with. 304 15:03.180 --> 15:06.570 Now if during the next 12 hours the condition_code 305 15:06.570 --> 15:08.310 is less than 700, 306 15:08.310 --> 15:12.540 then we'll switch that to equal true instead. 307 15:12.540 --> 15:16.140 Now, after the for loop has completed, 308 15:16.140 --> 15:17.850 then we can check to see 309 15:17.850 --> 15:20.130 if it will_rain in the next 12 hours, 310 15:20.130 --> 15:22.503 and if so, it will print, "Bring an umbrella". 311 15:23.760 --> 15:26.790 This way we'll only get one print statement being called, 312 15:26.790 --> 15:28.560 instead of every single time 313 15:28.560 --> 15:32.280 we land on a condition_code for rain. 314 15:32.280 --> 15:36.330 So this involved quite a bit of JSON passing 315 15:36.330 --> 15:41.330 and also understanding how to work with parameters and APIs. 316 15:41.970 --> 15:44.490 Now, I hope you managed to get this far by yourself, 317 15:44.490 --> 15:47.370 but if not, be sure to review what I've written 318 15:47.370 --> 15:49.710 and fix your code as required. 319 15:49.710 --> 15:51.330 If any of it was confusing, 320 15:51.330 --> 15:54.570 be sure to review previous lessons before you continue 321 15:54.570 --> 15:56.340 because we're now building heavily 322 15:56.340 --> 15:57.870 on your previous knowledge 323 15:57.870 --> 16:00.390 that you learned through the course. 324 16:00.390 --> 16:01.800 Now, in the next lesson, 325 16:01.800 --> 16:03.930 we're going to be looking at how we can, 326 16:03.930 --> 16:06.600 instead of printing, "Bring an umbrella", 327 16:06.600 --> 16:09.870 to send an SMS text message 328 16:09.870 --> 16:12.993 that notifies you to bring an umbrella instead. 329 16:13.860 --> 16:15.960 So for all of that and more, 330 16:15.960 --> 16:17.560 I'll see you on the next lesson.