1 00:00:04,660 --> 00:00:09,320 Alright so in the last video, we talked about the two challenges but I didn't really explain in detail what 2 00:00:09,320 --> 00:00:10,040 they were. 3 00:00:10,280 --> 00:00:12,450 So the first file challenge1.py, 4 00:00:12,500 --> 00:00:15,560 remember you can get that from the resources section of the last video. 5 00:00:15,590 --> 00:00:21,200 The challenge for that first one is to rewrite the following code to use a list comprehension, instead 6 00:00:21,200 --> 00:00:25,050 of a for loop, and you can see the for loop on lines 10 and 11. 7 00:00:25,060 --> 00:00:30,830 Now you need to add your solution below the loop, so that the resulting list is printed out below the 8 00:00:30,850 --> 00:00:36,890 output, so it makes it easier to check that it's producing exactly the same list and avoids entering the input 9 00:00:36,890 --> 00:00:37,900 text twice. 10 00:00:38,150 --> 00:00:39,510 So basically, we're going to do that. 11 00:00:39,630 --> 00:00:41,830 So let's go ahead and look at that first. 12 00:00:41,870 --> 00:00:44,540 And you can see that we've helpfully got a comment here 13 00:00:44,570 --> 00:00:48,330 to say type your solution here, and that's exactly where we're going to start typing it. 14 00:00:48,710 --> 00:00:53,030 And by the way, I mentioned down here that it could be useful to store the original words in a list 15 00:00:53,030 --> 00:00:53,980 as well. 16 00:00:53,990 --> 00:00:55,900 The for loop would look like this; 17 00:00:56,090 --> 00:00:57,720 down here you can see the example of that, 18 00:00:57,890 --> 00:01:03,320 and notice the extra parentheses so that we get tuples in the list. You can see the extra parentheses there 19 00:01:03,320 --> 00:01:05,930 on line 23. But we'll get to that shortly. 20 00:01:05,930 --> 00:01:07,730 We're going to do this first bit first. 21 00:01:07,800 --> 00:01:13,850 So for the first part, our comprehension will iterate over the list that's produced by splitting up the 22 00:01:13,850 --> 00:01:14,990 input text. 23 00:01:14,990 --> 00:01:16,730 Now we want the length of each word, 24 00:01:17,030 --> 00:01:22,940 so the comprehension's going to use the len function for that. So lets type in the code and we can see how 25 00:01:22,940 --> 00:01:24,430 we go about that. 26 00:01:24,580 --> 00:01:33,540 So I'm going to start on line 16, answer is equal to, left and right square brackets because it's a comprehension, 27 00:01:33,640 --> 00:01:43,580 a list comprehension, len parentheses x for x in text dot split parentheses. 28 00:01:43,820 --> 00:01:50,290 Then on the next line we can print parentheses answers, or answer I should say. 29 00:01:51,080 --> 00:01:56,510 So basically, we want the length of each word and the comprehension's using the len x function, as you 30 00:01:56,510 --> 00:02:00,530 can see on line 16, the same as the for loop did. Nothing unique there. 31 00:02:00,830 --> 00:02:03,750 So lets run the program to make sure it works. 32 00:02:03,860 --> 00:02:06,220 And obviously, we'll see the for loop version running as well. 33 00:02:07,590 --> 00:02:12,210 Now for the text, we're going to go with What have the Romans 34 00:02:14,860 --> 00:02:16,390 ever done for us. 35 00:02:18,610 --> 00:02:24,430 So you can see that we've got the same output for both the for loop and also for the comprehension. 36 00:02:24,520 --> 00:02:25,730 So that's a good start. 37 00:02:25,850 --> 00:02:31,250 Now for this second part of the challenge, I mentioned that it could be useful to store the original words 38 00:02:31,250 --> 00:02:32,840 in the list as well. 39 00:02:32,840 --> 00:02:38,900 Now the for loop would look like the example you can see there with the extra parentheses, so that 40 00:02:38,900 --> 00:02:40,490 we get tuples in the list. 41 00:02:40,770 --> 00:02:46,250 And obviously, you're going to type your corresponding comprehension down here on, starting on line 29. 42 00:02:46,250 --> 00:02:51,570 So basically, we want a tuple containing the original word and it's length. 43 00:02:51,580 --> 00:02:54,200 Let's go ahead and type in an answer to that. 44 00:02:54,200 --> 00:03:04,550 And we'll start this one on line 30: answer is equal to left and right square brackets, parentheses x comma space len 45 00:03:05,240 --> 00:03:14,990 left and right parentheses, x then we've got two closing parentheses as you can say there, then 4 x in text dot split 46 00:03:15,230 --> 00:03:18,420 parentheses. Then we'll print the answer on the next line, 47 00:03:18,440 --> 00:03:25,670 print parentheses answer. So basically, is we're producing a tuple containing the word and its length. 48 00:03:25,720 --> 00:03:32,980 And what I'll do is, I'm just going to copy the exact same output that we typed in up here. Run the code, 49 00:03:43,470 --> 00:03:50,130 and once again, you can see our comprehension produces the same list as the original for loop did. Converting 50 00:03:50,130 --> 00:03:55,650 a for loop to a comprehension is quite straightforward, and they're a very useful addition to your programming 51 00:03:55,650 --> 00:03:56,490 tool kit. 52 00:03:56,590 --> 00:04:00,930 Now before I move onto the solution to the second challenge, there's a couple of things to say about 53 00:04:00,930 --> 00:04:01,960 this one. 54 00:04:02,010 --> 00:04:08,040 Now the first is, we probably don't want duplicates. In other words, if the same word appears more than once 55 00:04:08,040 --> 00:04:09,040 in the input, 56 00:04:09,090 --> 00:04:11,260 there's no point in counting it twice. 57 00:04:11,640 --> 00:04:13,080 So to get around that, 58 00:04:13,140 --> 00:04:16,740 what we could do is use a set comprehension instead. 59 00:04:17,000 --> 00:04:21,360 Now we've already seen how to do that, to convert a list comprehension to a set comprehension. 60 00:04:21,420 --> 00:04:24,580 We just replace the square brackets with curly braces. 61 00:04:24,590 --> 00:04:25,710 So let's go ahead and do that, 62 00:04:30,060 --> 00:04:34,500 and let's run that. Now I'm going to provide a different input string here, 63 00:04:34,530 --> 00:04:36,430 so we've got some duplicated words. 64 00:04:36,720 --> 00:04:42,670 So the input text contains repeated words in the input. 65 00:04:47,710 --> 00:04:54,070 OK, save that, and you can see in this example down here now, we get a set returned, 66 00:04:54,270 --> 00:04:58,760 and because sets don't contain duplicates, we get each word only once. 67 00:04:58,770 --> 00:05:01,980 Now note that the words aren't necessarily appearing in order. 68 00:05:01,980 --> 00:05:08,620 In fact, they're definitely not in order as you can see here. We've got the word 'contains' here, c is after the letter r 69 00:05:09,120 --> 00:05:13,500 alphabetically, and that's because by definition there's no ordering in a set. 70 00:05:13,500 --> 00:05:18,410 Now you'll probably be doing this to produce some sort of metrics, such as an average word length for example, 71 00:05:18,780 --> 00:05:22,080 and therefore the ordering wouldn't be important. 72 00:05:22,080 --> 00:05:24,690 Now when you're considering using a comprehension, 73 00:05:24,750 --> 00:05:28,650 remember that set comprehensions can also be quite useful. 74 00:05:28,710 --> 00:05:33,810 They work the same as list comprehensions and can be very useful for avoiding duplicates. 75 00:05:34,120 --> 00:05:37,140 OK so the next thing isn't really a problem with comprehensions. 76 00:05:37,380 --> 00:05:39,750 It's got to do with the way that split works. 77 00:05:40,110 --> 00:05:42,120 So we split our text up to get words, 78 00:05:42,270 --> 00:05:48,570 and it works fine as long as each word's separated from the next with white space - white spaces, space tabs 79 00:05:48,570 --> 00:05:53,970 and the new line characters, although we can't enter line breaks in the input. Now to see what I mean, 80 00:05:54,210 --> 00:05:58,000 let's run the program again and use some punctuation. 81 00:05:58,380 --> 00:06:02,010 So this time we're going to enter What have they 82 00:06:02,190 --> 00:06:10,360 comma the Romans comma ever done for us, ever done for us 83 00:06:10,560 --> 00:06:18,200 with a question mark on the end, enter. Now the program still works, but they now show, for example, the word 'they' over 84 00:06:18,210 --> 00:06:24,000 here, as being of length five, five characters, and that's because of the comma after the word 'they'. 85 00:06:24,160 --> 00:06:30,210 We've also got 'Romans' showing a seven as well, and 'us' even having three over here, and that's because of 86 00:06:30,210 --> 00:06:31,090 the question mark. 87 00:06:31,260 --> 00:06:35,450 Now I'm mentioning this because we can use a nested comprehension to fix it. 88 00:06:35,460 --> 00:06:40,520 Remember the problem, and we'll come back to it when we look at nesting comprehensions later. 89 00:06:40,550 --> 00:06:40,780 Right, 90 00:06:40,780 --> 00:06:44,410 so that's the first challenge. I'm going to close down that challenge1.py now. 91 00:06:45,230 --> 00:06:51,130 So the second challenge, we had to create a list comprehension to convert a tuple from inches to centimetres. 92 00:06:51,740 --> 00:06:57,470 And just as a note here, in case it's not obvious, a list comprehension produces a list, but it doesn't 93 00:06:57,470 --> 00:06:59,540 have to be given a list to iterate over. 94 00:06:59,840 --> 00:07:02,640 Now you can use a list comprehension with any iterable type. 95 00:07:02,720 --> 00:07:06,500 So we're going to write a comprehension here - that was your challenge - to convert dimensions from 96 00:07:06,500 --> 00:07:08,600 inches to centimetres. 97 00:07:08,680 --> 00:07:14,700 Now our dimensions will be replaced by tuple for the length, width and height, and as a reminder it's 2.5 98 00:07:14,770 --> 00:07:16,560 4 centimeters to 1 inch. 99 00:07:16,940 --> 00:07:21,800 And then down here, once you've got the correct values, change the code to produce a tuple rather 100 00:07:21,800 --> 00:07:22,910 than a list. 101 00:07:23,140 --> 00:07:27,950 Alright, so we've got our code there printing out the measurement, and again in case it's not obvious, a list 102 00:07:27,950 --> 00:07:32,960 comprehension produces a list, but it doesn't have to be given the list to iterate over - that's the important 103 00:07:32,960 --> 00:07:33,590 part. 104 00:07:33,680 --> 00:07:37,970 And that's really why I actually created this challenge in the first place, really just 105 00:07:37,970 --> 00:07:41,950 to demonstrate that a comprehension can use any iterable type or sequence. 106 00:07:42,050 --> 00:07:47,840 You don't have to provide a list to a list comprehension. So the list, or set, in list comprehension refers 107 00:07:47,840 --> 00:07:54,790 to the type of object produced by the comprehension. Alright, so we've got the value here, so on line 13, let's now 108 00:07:54,890 --> 00:08:01,010 enter our comprehension. I'm going to delete that code there in between the square brackets, 109 00:08:01,010 --> 00:08:02,320 and we're going to go with inch 110 00:08:04,970 --> 00:08:13,670 multiplied by two point five four for inch in inch underscore measurement. 111 00:08:13,670 --> 00:08:16,830 I'll leave the code to print the cm underscore measurement. 112 00:08:17,240 --> 00:08:23,960 So the comprehension iterates over each item in the tuple, and multiplies it by the number of centimeters 113 00:08:24,010 --> 00:08:26,190 in an inch, two point five four. 114 00:08:26,480 --> 00:08:31,420 So let's actually run this to confirm that it works. 115 00:08:31,800 --> 00:08:34,059 And you can see we've correctly got the values there, 116 00:08:34,110 --> 00:08:39,179 seven point six two, twenty point three two and fifty point eight. Alright, now our 117 00:08:39,179 --> 00:08:43,659 next part here, we need to convert the code to produce a tuple rather than a list. 118 00:08:43,919 --> 00:08:48,890 All we have to do is parse the comprehension to the tuple function to get that to work. 119 00:08:48,920 --> 00:08:55,860 So let's add the code there to do that. Now line 18, we'll do a cm underscore measurement is equal to tuple, 120 00:08:57,450 --> 00:09:03,660 then we want parentheses, then we need to add the list comprehension to the tuple call. So in left to right 121 00:09:03,690 --> 00:09:04,860 square brackets, 122 00:09:04,910 --> 00:09:05,960 it's going to be - 123 00:09:05,980 --> 00:09:12,690 now I could just paste that code in, but I'll just quickly type it in - inch times two point five four, for 124 00:09:12,690 --> 00:09:20,320 inch in inch underscore measurement. And then on the last line we'll print out 125 00:09:21,000 --> 00:09:23,070 cm_measurement again. 126 00:09:23,070 --> 00:09:30,020 So basically, this call here on line 18 is converting a tuple of values in one unit, inches here, to an 127 00:09:30,020 --> 00:09:33,990 equivalent tuple using another set of units, centimetres. 128 00:09:34,040 --> 00:09:36,420 And I think that's pretty impressive for one line of code, 129 00:09:36,610 --> 00:09:39,440 but lets actually just run it to make sure we do get the right results. 130 00:09:41,600 --> 00:09:45,130 We've now got the identical results in a tuple. 131 00:09:45,520 --> 00:09:51,540 Alright, so I'm going to stop the video here. In the next one, we're going to look at adding conditions to our comprehensions. 132 00:09:51,590 --> 00:09:52,980 So I'll see you in the next video.