class: title-slide, middle, inverse .leftcol30[ <center> <img src="https://github.com/emse-p4a-gwu/emse-p4a-gwu.github.io/raw/master/images/p4a_hex_sticker.png" width=250> </center> ] .rightcol70[ # Week 4: .fancy[Conditionals & Testing] ### EMSE 4574: Intro to Programming for Analytics ### John Paul Helveston ### September 22, 2020 ] --- class: inverse, middle # Week 4: .fancy[Conditionals & Testing] ## 1. Conditionals ## 2. Testing ## 3. Tips --- class: inverse, middle # Week 4: .fancy[Conditionals & Testing] ## 1. .orange[Conditionals] ## 2. Testing ## 3. Tips --- # "Flow Control" ### Code that alters the otherwise linear flow of operations in a program. -- .leftcol[ ### This week: - `if` statements - `else` statements ] -- .rightcol[ ### Next week: - `for` loops - `while` loops - `break` statements - `next` statements ] --- # The `if` statement -- .leftcol40[.code90[ ### Basic format ```r if ( CONDITION ) { # Do stuff here } ``` ]] -- .rightcol[ ### Flow chart: <img src="images/condition_if.png" width="400"> ] --- # Quick code tracing .leftcol[.code80[ Cosider this function: ```r f <- function(x) { cat("A") if (x == 0) { cat("B") cat("C") } cat("D") } ``` ]] .rightcol[.code80[ What will this print? ```r f(1) f(0) ``` ]] --- class: inverse
03
:
00
# Quick practice ### Write the function `absValue(n)` that returns the absolute value of a number (and no cheating - you can't use the built-in `abs()` function!) Tests: - `absValue(7) == 7` - `absValue(-7) == 7` - `absValue(0) == 0` --- # Adding an `else` to an `if` .leftcol40[ ### Basic format: .code80[ ```r if ( CONDITION ) { # Do stuff here } else { # Do other stuff here } ``` ]] -- .rightcol[ ### Flow chart: <img src="images/condition_if_else.png" width="500"> ] --- # Quick code tracing .leftcol[.code70[ Consider this code: ```r f <- function(x) { cat("A") if (x == 0) { cat("B") cat("C") } else { cat("D") if (x == 1) { cat("E") } else { cat("F") } } cat("G") } ``` ]] .rightcol[.code80[ What will this print? ```r f(0) f(1) f(2) ``` ]] --- .leftcol[ # `else if` chains Example - "bracketing" problems: ```r getLetterGrade <- function(score) { if (score >= 90) { grade = "A" } else if (score >= 80) { grade = "B" } else if (score >= 70) { grade = "C" } else if (score >= 60) { grade = "D" } else { grade = "F" } return(grade) } ``` ] -- .rightcol[ Check function output: ```r getLetterGrade(99) ``` ``` ## [1] "A" ``` ```r getLetterGrade(88) ``` ``` ## [1] "B" ``` ```r getLetterGrade(70) ``` ``` ## [1] "C" ``` ```r getLetterGrade(61) ``` ``` ## [1] "D" ``` ```r getLetterGrade(22) ``` ``` ## [1] "F" ``` ] --- class: inverse
08
:
00
# Think-Pair-Share Write the function `getType(x)` that returns the type of the data (either `integer`, `double`, `character`, or `logical`). Basically, it does the same thing as the `typeof()` function (but you can't use `typeof()` in your solution). Tests: - `getType(3) == "double"` - `getType(3L) == "integer"` - `getType("foo") == "character"` - `getType(TRUE) == "logical"` --- class: inverse, middle # Week 4: .fancy[Conditionals & Testing] ## 1. Conditionals ## 2. .orange[Testing] ## 3. Tips --- # Why write test functions? <br> -- ## 1. They help you understand the problem -- ## 2. They verify that a function is working as expected --- # Test function "syntax" -- .leftcol[.code80[ ### Function: ```r functionName <- function(arguments) { # Do stuff here return(something) } ``` ]] -- .rightcol[.code80[ ### Test function: ```r test_functionName <- function() { cat("Testing functionName()...") # Put test cases here cat("Passed!\n") } ``` ]] --- # Writing test cases with `stopifnot()` `stopifnot()` stops the function if whatever is inside the `()` is not `TRUE`. -- .leftcol[.code80[ ### Function: ```r isEven <- function(n) { return((n %% 2) == 0) } ``` - `isEven(1)` should be `FALSE` - `isEven(2)` should be `TRUE` - `isEven(-7)` should be `FALSE` ]] -- .rightcol[.code80[ ### Test function: ```r test_isEven <- function() { cat("Testing isEven()...") stopifnot(isEven(1) == FALSE) stopifnot(isEven(2) == TRUE) stopifnot(isEven(-7) == FALSE) cat("Passed!\n") } ``` ]] --- # Writing test cases with `stopifnot()` `stopifnot()` stops the function if whatever is inside the `()` is not `TRUE`. .leftcol[.code80[ ### Function: ```r isEven <- function(n) { return((n %% 2) == 0) } ``` - `isEven(1)` should be `FALSE` - `isEven(2)` should be `TRUE` - `isEven(-7)` should be `FALSE` ]] .rightcol[.code80[ ### Test function: ```r test_isEven <- function() { cat("Testing isEven()...") stopifnot(isEven(1) == FALSE) stopifnot(isEven(2) == TRUE) stopifnot(isEven(-7) == FALSE) cat("Passed!\n") } ``` ```r test_isEven() ``` ``` ## Testing isEven()...Passed! ``` ]] --- # Write the test function _first_! .leftcol[.code80[ ### Step 1: Write the test function ```r test_isEven <- function() { cat("Testing isEven()...") stopifnot(isEven(1) == FALSE) stopifnot(isEven(2) == TRUE) stopifnot(isEven(-7) == FALSE) cat("Passed!\n") } ``` ]] -- .rightcol[.code80[ ### Step 2: Write the function ```r isEven <- function(n) { return((n %% 2) == 0) } ``` ### Step 3: Test the function ```r test_isEven() ``` ``` ## Testing isEven()...Passed! ``` ]] --- # Test cases to consider: NOLES .leftcol40[ - ### **N**ormal cases - ### **O**pposites - ### **L**arge & small cases - ### **E**dge cases - ### **S**pecial cases ] --- # Test cases to consider: NOLES .leftcol40[ - ### .red[**N**ormal cases] - ### **O**pposites - ### **L**arge & small cases - ### **E**dge cases - ### **S**pecial cases ] .rightcol60[.code80[ ### Example: ```r isEven <- function(n) { return((n %% 2) == 0) } ``` ] - `isEven(1) == FALSE` - `isEven(2) == TRUE` - `isEven(-7) == FALSE` ] --- # Test cases to consider: NOLES .leftcol40[ - ### **N**ormal cases - ### .red[**O**pposites] - ### **L**arge & small cases - ### **E**dge cases - ### **S**pecial cases ] .rightcol60[.code80[ ### Example: ```r isEven <- function(n) { return((n %% 2) == 0) } ``` ] Need cases that return both `TRUE` and `FALSE` - `isEven(52) == TRUE` - `isEven(53) == FALSE` ] --- # Test cases to consider: NOLES .leftcol40[ - ### **N**ormal cases - ### **O**pposites - ### .red[**L**arge & small cases] - ### **E**dge cases - ### **S**pecial cases ] .rightcol60[.code80[ ### Example: ```r isEven <- function(n) { return((n %% 2) == 0) } ``` ] - `isEven(8675309) == FALSE` - `isEven(-8675309) == FALSE` ] --- # Test cases to consider: NOLES .leftcol40[ - ### **N**ormal cases - ### **O**pposites - ### **L**arge & small cases - ### .red[**E**dge cases] - ### **S**pecial cases ] .rightcol60[.code80[ ### Example: ```r isPositive <- function(n) { return(n > 0) } ``` ] - `isPositive(0.000001) == TRUE` - `isPositive(0) == FALSE` - `isPositive(-0.000001) == FALSE` ] --- # Test cases to consider: NOLES .leftcol40[ - ### **N**ormal cases - ### **O**pposites - ### **L**arge & small cases - ### **E**dge cases - ### .red[**S**pecial cases] ] .rightcol60[ - Negative numbers - `0` and `1` for integers - The empty string, `""` - Strange input _types_, e.g. `"2"` instead of `2`. ] --- # Testing function inputs -- .leftcol45[.code80[ What if we gave `isEven()` the wrong input type? ```r isEven <- function(n) { return((n %% 2) == 0) } ``` ```r isEven('42') ``` ``` ## Error in n%%2: non-numeric argument to binary operator ``` ]] -- .rightcol55[.code80[ An improved function that checks inputs: ```r isEven <- function(n) { if (! is.numeric(n)) { return(NaN) } return((n %% 2) == 0) } ``` ]] --- # Testing function inputs .leftcol45[.code80[ What if we gave `isEven()` the wrong input type? ```r isEven <- function(n) { return((n %% 2) == 0) } ``` ```r isEven('42') ``` ``` ## Error in n%%2: non-numeric argument to binary operator ``` ]] .rightcol55[.code80[ An improved function that checks inputs: ```r isEven <- function(n) { if (! is.numeric(n)) { return(NaN) } return((n %% 2) == 0) } ``` ```r isEven('42') ``` ``` ## [1] NaN ``` ```r isEven(TRUE) ``` ``` ## [1] NaN ``` ]] --- class: inverse
15
:
00
# Think-Pair-Share For each of the following functions, start by writing a test function that tests the function for a variety of values of inputs. Consider cases that you might not expect! .leftcol[ 1. Write the function `isFactor(f, n)` that takes two integer values and returns `TRUE` if `f` is a factor of `n`, and `FALSE` otherwise. Note that every integer is a factor of `0`. Assume `f` and `n` will only be numeric values, e.g. `2` is a factor of `6`. ] .rightcol[ 2. Write the function `isMultiple(m, n)` that takes two integer values and returns `TRUE` if `m` is a multiple of `n` and `FALSE` otherwise. Note that `0` is a multiple of every integer other than itself. Hint: You may want to use the `isFactor(f, n)` function you just wrote above. Assume `m` and `n` will only be numeric values. ] --- class: inverse, center # .fancy[Break]
05
:
00
--- class: inverse, middle # Week 4: .fancy[Conditionals & Testing] .leftcol[ ## 1. Conditionals ## 2. Testing ## 3. .orange[Tips] ] .rightcol[ <center> <img src="images/test_cases_fry.png"> </center> ] --- # Debugging your code Use `traceback()` to find the steps that led to an error (the "call stack") -- Example: ```r f <- function(x) { return(x + 1) } g <- function(x) { return(f(x) - 1) } ``` -- ```r g('a') ``` ``` ## Error in x + 1: non-numeric argument to binary operator ``` -- ```r traceback() ``` ``` 2: f(x) at #2 1: g("a") ``` --- # When testing _numbers_, use `almostEqual()` .leftcol[.code80[ Rounding errors can cause headaches: ```r x <- 0.1 + 0.2 x ``` ``` ## [1] 0.3 ``` ```r x == 0.3 ``` ]] --- # When testing _numbers_, use `almostEqual()` .leftcol[.code80[ Rounding errors can cause headaches: ```r x <- 0.1 + 0.2 x ``` ``` ## [1] 0.3 ``` ```r x == 0.3 ``` ``` ## [1] FALSE ``` ```r print(x, digits = 20) ``` ``` ## [1] 0.30000000000000004441 ``` ]] --- # When testing _numbers_, use `almostEqual()` .leftcol[.code80[ Rounding errors can cause headaches: ```r x <- 0.1 + 0.2 x ``` ``` ## [1] 0.3 ``` ```r x == 0.3 ``` ``` ## [1] FALSE ``` ```r print(x, digits = 20) ``` ``` ## [1] 0.30000000000000004441 ``` ]] .rightcol[.code80[ Define a function that checks if two values are _almost_ the same: ```r almostEqual <- function(n1, n2, threshold = 0.00001) { return(abs(n1 - n2) <= threshold) } ``` ```r x <- 0.1 + 0.2 almostEqual(x, 0.3) ``` ``` ## [1] TRUE ``` ]] --- # Checking for integer values .leftcol[ Since numbers are doubles by default, the `is.integer(x)` function can be confusing: .code80[ ```r is.integer(7) ``` ``` ## [1] FALSE ``` ]] -- .rightcol[ Define a new function that returns `TRUE` if the _value_ is an integer: .code80[ ```r is.integer.val <- function(x) { y <- round(x) return(almostEqual(x, y)) } is.integer.val(7) ``` ``` ## [1] TRUE ``` ]] --- # Checking for special data types -- .leftcol[ **Not available**: `NA`<br>_value is "missing"_ ```r x <- NA x == NA ``` ``` ## [1] NA ``` ] -- .rightcol[ **No value**: `NULL`<br>_no value whatsoever_ ```r x <- NULL x == NULL ``` ``` ## logical(0) ``` ] --- # Checking for special data types .leftcol[ **Not available**: `NA`<br>_value is "missing"_ ```r x <- NA x == NA ``` ``` ## [1] NA ``` Have to use special function: ```r is.na(x) ``` ``` ## [1] TRUE ``` ] .rightcol[ **No value**: `NULL`<br>_no value whatsoever_ ```r x <- NULL x == NULL ``` ``` ## logical(0) ``` Have to use special function: ```r is.null(x) ``` ``` ## [1] TRUE ``` ] --- class: inverse
15
:
00
# Think-Pair-Share Write the function `getInRange(x, bound1, bound2)` which takes 3 numeric values: `x`, `bound1`, and `bound2` (`bound1` is not necessarily less than `bound2`). If `x` is between the two bounds, just return `x`, but if `x` is less than the lower bound, return the lower bound, or if `x` is greater than the upper bound, return the upper bound. For example: - `getInRange(1, 3, 5)` returns `3` (the lower bound, since 1 is below [3,5]) - `getInRange(4, 3, 5)` returns `4` (the original value, since 4 is between [3,5]) - `getInRange(6, 3, 5)` returns `5` (the upper bound, since 6 is above [3,5]) - `getInRange(6, 5, 3)` returns `5` (the upper bound, since 6 is above [3,5]) **Bonus**: Re-write `getInRange(x, bound1, bound2)` without using conditionals --- # [HW 4](https://p4a.seas.gwu.edu/2020-Fall/hw4-conditionals-testing.html) ## You'll need to write a _test function_ for each function!