Merge pull request #16 from nickgood/error_checking

Error checking

slides/CourseNotes_Week12.Rmd

@@ -13,6 +13,42 @@ library(knitr)
 library(ggmap)
 ```
 
+# Functions-- Error checking
+
+## Functions-- Error checking
+
+There are ways to check for errors in the arguments a user inputs to the function. One useful check is to see if user inputs are in the required class. 
+
+The `assertive` package has some functions that you can use for common checks of the inputs to a function. If someone inputs something of the wrong class, it will give a useful error message. For example, the `assert_is_numeric` function will check if an object is in a numeric class. If so, it will do nothing. If not, it will return an error message:
+
+```{r error = TRUE}
+library(assertive)
+assert_is_numeric(1:3)
+assert_is_numeric(c("a", "b"))
+```
+
+## Functions-- Error checking
+
+You could add this in the code for the `add_one` function, so a useful error message will be returned if a user tries to input something besides a numeric vector for `number`:
+
+```{r}
+add_one <- function(number){
+  assert_is_numeric(number)
+  number + 1 
+}
+```
+
+```{r error = TRUE}
+add_one(number = 1:3)
+add_one(number = c("a", "b"))
+```
+
+## Functions-- Error checking
+
+I would recommend that you not worry about this too much when you're learning to write functions for your own use. 
+
+However, once you have mastered the basics of writing functions and start writing them for others to use, you'll want to start incorporating this. 
+
 # Shiny applications
 
 ## Resources for learning Shiny

slides/CourseNotes_Week8a.Rmd

@@ -196,40 +196,6 @@ add_one <- function(number = 0){
 
 If using `return` helps you think about what's happening with the code in your function, you can use it. However, outside of a few exceptions, you usually won't need to do it.
 
-## Functions-- Error checking
-
-There are ways to check for errors in the arguments a user inputs to the function. One useful check is to see if user inputs are in the required class. 
-
-The `assertive` package has some functions that you can use for common checks of the inputs to a function. If someone inputs something of the wrong class, it will give a useful error message. For example, the `assert_is_numeric` function will check if an object is in a numeric class. If so, it will do nothing. If not, it will return an error message:
-
-```{r error = TRUE}
-library(assertive)
-assert_is_numeric(1:3)
-assert_is_numeric(c("a", "b"))
-```
-
-## Functions-- Error checking
-
-You could add this in the code for the `add_one` function, so a useful error message will be returned if a user tries to input something besides a numeric vector for `number`:
-
-```{r}
-add_one <- function(number){
-  assert_is_numeric(number)
-  number + 1 
-}
-```
-
-```{r error = TRUE}
-add_one(number = 1:3)
-add_one(number = c("a", "b"))
-```
-
-## Functions-- Error checking
-
-I would recommend that you not worry about this too much when you're learning to write functions for your own use. 
-
-However, once you have mastered the basics of writing functions and start writing them for others to use, you'll want to start incorporating this. 
-
 ## `if` / `else`
 
 In R, the `if` statement evaluates everything in the parentheses and, if that evaluates to `TRUE`, runs everything in the braces. This means that you can trigger code in an `if` statement with a single-value logical vector: