|
| 1 | +--- |
| 2 | +description: Review your knowledge of error handling with Python via a number of examples. |
| 3 | +sidebar_label: "S05 L07: Handling User Errors Quiz" |
| 4 | +--- |
| 5 | + |
| 6 | +# Handling User Errors Quiz |
| 7 | + |
| 8 | +This lecture follows a quiz in the course that asks several questions about error handling. Let's take a moment to go over the quiz questions and their answers! |
| 9 | + |
| 10 | +## Question 1 |
| 11 | + |
| 12 | +A quick question to warm you up! |
| 13 | + |
| 14 | +Ben and Dan are two novice software developers. They are trying to build a function that asks user for a number and calculate its power of 2. Here is their first version of code: |
| 15 | + |
| 16 | +```python |
| 17 | +def power_of_two(): |
| 18 | + n = input('Please enter a number: ') |
| 19 | + n_square = n ** 2 |
| 20 | + return n_square |
| 21 | +``` |
| 22 | + |
| 23 | +What do you think of the code, will it work? Before submitting your answer, think about why. |
| 24 | + |
| 25 | +### Answer |
| 26 | + |
| 27 | +The answer here is that this code will never work, because the `input()` function always returns a string, and you can't square a string. |
| 28 | + |
| 29 | +## Question 2 |
| 30 | + |
| 31 | +In this second question, Ben and Dan improve the code slightly: |
| 32 | + |
| 33 | +```python |
| 34 | +def power_of_two(): |
| 35 | + user_input = input('Please enter a number: ') |
| 36 | + n = float(user_input) |
| 37 | + n_square = n ** 2 |
| 38 | + return n_square |
| 39 | +``` |
| 40 | + |
| 41 | +### Answer |
| 42 | + |
| 43 | +Now the code will work on some inputs, but not others. That's because if the user decides to not enter a number (e.g. a letter), the program will crash. |
| 44 | + |
| 45 | +## Question 3 |
| 46 | + |
| 47 | +Now, Ben and Dan add some error handling: |
| 48 | + |
| 49 | +```python |
| 50 | +def power_of_two(): |
| 51 | + user_input = input('Please enter a number: ') |
| 52 | + try: |
| 53 | + n = float(user_input) |
| 54 | + except ValueError: |
| 55 | + print('Your input was invalid.') |
| 56 | + finally: |
| 57 | + n_square = n ** 2 |
| 58 | + return n_square |
| 59 | +``` |
| 60 | + |
| 61 | +Dan wants to test this function with some different inputs when the program asks for a number. He decides to test with two cases. First case: he enters `4`, which is a valid numeric input. Second case: he enters `'dan'` which is an invalid input. |
| 62 | + |
| 63 | +What do you expect the function to return with Dan's input (Denoted as `[4, 'dan']`)? If the program breaks due to an error, we mark the function's return value as `Error`, and if it does not return anything, we mark the return value as `None`. |
| 64 | + |
| 65 | +This one's a bit of a tricky one! Type the code out and try it in your editor for optimal chances of getting it right! |
| 66 | + |
| 67 | +Potential answers: |
| 68 | + |
| 69 | +- `[16.0, None]` |
| 70 | +- `[Error, Error]` |
| 71 | +- `[16.0, 0.0]` |
| 72 | +- `[16.0, Error]` |
| 73 | +- None of the above |
| 74 | + |
| 75 | +### Answer |
| 76 | + |
| 77 | +The `finally` block gets executed regardless of the occurrence of `ValueError`, so the correct answer is `[16.0, Error]`. We only defined `n` in the try block. If the input was invalid, `n` never gets its value assigned, and thus will raise an error when we try to access it in the `finally` block. |
| 78 | + |
| 79 | +## Question 4 |
| 80 | + |
| 81 | +As Dan found the error in the code, he decided to make the following change: |
| 82 | + |
| 83 | +```python |
| 84 | +def power_of_two(): |
| 85 | + user_input = input('Please enter a number: ') |
| 86 | + try: |
| 87 | + n = float(user_input) |
| 88 | + except ValueError: |
| 89 | + print('Your input was invalid. Using default value 0') |
| 90 | + n = 0 |
| 91 | + else: |
| 92 | + n_square = n ** 2 |
| 93 | + return n_square |
| 94 | +``` |
| 95 | + |
| 96 | +Dan wants to test with the same test cases from last time: `[4, 'dan']`. |
| 97 | + |
| 98 | +What do you expect the function to return with the two inputs (denoted as `[4, 'dan']`) this time? If the program breaks due to an error, we mark the function's return value as `Error`, and if it does not return anything, we mark the return value as `None`. |
| 99 | + |
| 100 | +### Answer |
| 101 | + |
| 102 | +This one's tricky! The `else` block only gets executed if no error occurs in the `try` block. So the code will finish after assigning `n` with the default value `0` if it ever reaches the `except` block. Since there is no `return` statement other than in the `else` clause, the function returns `None`. |
| 103 | + |
| 104 | +## Question 5 |
| 105 | + |
| 106 | +Now that Dan's code still cannot work, Ben insisted on using a finally block, here is his code: |
| 107 | + |
| 108 | +```python |
| 109 | +def power_of_two(): |
| 110 | + user_input = input('Please enter a number: ') |
| 111 | + try: |
| 112 | + n = float(user_input) |
| 113 | + except ValueError: |
| 114 | + print('Your input was invalid. Using default value 0') |
| 115 | + n = 0 |
| 116 | + else: |
| 117 | + n_square = n ** 2 |
| 118 | + finally: |
| 119 | + return n_square |
| 120 | +``` |
| 121 | + |
| 122 | +They decided to use the same test cases: `[4, 'dan']`. |
| 123 | + |
| 124 | +What do you expect the function to return with the two inputs (denoted as `[4, 'dan']`) this time? If the program breaks due to an error, we mark the function's return value as `Error`, and if it does not return anything, we mark the return value as `None`. |
| 125 | + |
| 126 | +Potential answers: |
| 127 | + |
| 128 | +- `[16.0, None]` |
| 129 | +- `[Error, Error]` |
| 130 | +- `[16.0, 0.0]` |
| 131 | +- `[16.0, Error]` |
| 132 | +- None of the above. |
| 133 | + |
| 134 | +### Answer |
| 135 | + |
| 136 | +In this case, the answer is `[16.0, Error]`. The `finally` block gets executed regardless of the occurrence of `ValueError`, while the `else` block only gets executed if no error occurs in the `try` block. However, we only defined `n_square` in the `else` block. If the input was invalid, `n_square` never gets its value assigned, thus will raise an error when we try to access it in the `finally` block. |
| 137 | + |
| 138 | +## Question 6 |
| 139 | + |
| 140 | +Ben and Dan just won't give up, here's their last try: |
| 141 | + |
| 142 | +```python |
| 143 | +def power_of_two(): |
| 144 | + user_input = input('Please enter a number: ') |
| 145 | + try: |
| 146 | + n = float(user_input) |
| 147 | + except ValueError: |
| 148 | + print('Your input was invalid. Using default value 0') |
| 149 | + return 0 |
| 150 | + finally: |
| 151 | + n_square = n ** 2 |
| 152 | + return n_square |
| 153 | +``` |
| 154 | + |
| 155 | +They decided to use the same test cases: `[4, 'dan']`. |
| 156 | + |
| 157 | +What do you expect the function to return with the two inputs (denoted as `[4, 'dan']`) this time? If the program breaks due to an error, we mark the function's return value as `Error`, and if it does not return anything, we mark the return value as `None`. |
| 158 | + |
| 159 | +Potential answers: |
| 160 | + |
| 161 | +- `[16.0, None]` |
| 162 | +- `[16.0, 0.0]` |
| 163 | +- `[16.0, 0]` |
| 164 | +- `[16.0, Error]` |
| 165 | +- None of the above. |
| 166 | + |
| 167 | +### Answer |
| 168 | + |
| 169 | +The answer is `[16.0, Error]`. It's a tricky one. The `finally` block will get executed regardless of the `return` statement in both `try` block and `except` block. However, we only defined `n` in the `try` block. If the input was invalid, `n` never gets its value assigned, thus will raise an error when we try to access it in the `finally` block. |
| 170 | + |
| 171 | +## Improvements to the code |
| 172 | + |
| 173 | +This is the latest code that Ben and Dan wrote. As you know, it still has some issues! |
| 174 | + |
| 175 | +```python |
| 176 | +def power_of_two(): |
| 177 | + user_input = input('Please enter a number: ') |
| 178 | + try: |
| 179 | + n = float(user_input) |
| 180 | + except ValueError: |
| 181 | + print('Your input was invalid. Using default value 0') |
| 182 | + return 0 |
| 183 | + finally: |
| 184 | + n_square = n ** 2 |
| 185 | + return n_square |
| 186 | +``` |
| 187 | + |
| 188 | +To fix this, we need to use the `else` clause! |
| 189 | + |
| 190 | +First, let's think about what parts of the code can raise an exception. In our case, that's converting the user input to a float. |
| 191 | + |
| 192 | +That line is already inside the `try` block, so that's good! |
| 193 | + |
| 194 | +```python |
| 195 | +def power_of_two(): |
| 196 | + user_input = input('Please enter a number: ') |
| 197 | + try: |
| 198 | + n = float(user_input) |
| 199 | +``` |
| 200 | + |
| 201 | +Next up, let's write what should happen if there is no error: |
| 202 | + |
| 203 | +```python |
| 204 | +def power_of_two(): |
| 205 | + user_input = input('Please enter a number: ') |
| 206 | + try: |
| 207 | + n = float(user_input) |
| 208 | + else: |
| 209 | + n_square = n ** 2 |
| 210 | + return n_square |
| 211 | +``` |
| 212 | + |
| 213 | +Doing it this way is not absolutely necessary, of course, and you may like writing the `except` part first. It's up to you. |
| 214 | + |
| 215 | +But now we've got the "happy path" of our code, or what happens if there are no errors. |
| 216 | + |
| 217 | +So let's handle any errors that come up. In our case that's just `ValueError`, so let's add an `except` clause for that. |
| 218 | + |
| 219 | +```python |
| 220 | +def power_of_two(): |
| 221 | + user_input = input('Please enter a number: ') |
| 222 | + try: |
| 223 | + n = float(user_input) |
| 224 | + except ValueError: |
| 225 | + print('Your input was invalid. Using default value 0') |
| 226 | + return 0 |
| 227 | + else: |
| 228 | + n_square = n ** 2 |
| 229 | + return n_square |
| 230 | +``` |
| 231 | + |
| 232 | +So all that was needed in the end was to change the `finally` to an `else`, because there wasn't anything that we needed to run _no matter what_. |
| 233 | + |
| 234 | +A small improvement we might do is change the `return 0` for `return 0.0`, so that the function always returns a `float`: |
| 235 | + |
| 236 | +```python |
| 237 | +def power_of_two(): |
| 238 | + user_input = input('Please enter a number: ') |
| 239 | + try: |
| 240 | + n = float(user_input) |
| 241 | + except ValueError: |
| 242 | + print('Your input was invalid. Using default value 0') |
| 243 | + return 0.0 |
| 244 | + else: |
| 245 | +<
741A
/span> n_square = n ** 2 |
| 246 | + return n_square |
| 247 | +``` |
| 248 | + |
| 249 | +That way the caller of `power_of_two()` can always expect a `float` to come back, and not "sometimes a `float`, sometimes an `int`". Consistency helps! |
0 commit comments