Signaling Conditions (in R)

Many programming languages rely on being able to signal “conditions” when code doesn’t work as intended. These conditions range from effectively ‘for your information’ notes all the way to full-blown errors. See the tabs below for the three most common conditions in R

Message

Messages indicate that an action has been taken on the user’s behalf but not necessarily a problem. These can be useful to explicitly inform a user about an assumed default value or–for code that iterates for a long time–reassure users that the function is still working.

message("Pssst")

Pssst

Warning

Warnings indicate that something has gone wrong but the function could at least partially recover. These can be useful when some facet of a user’s input is incorrect but the code can still complete. I often uses warnings in my custom functions that have at least one argument that expects a logical (i.e., T or F). If the user supplies anything other than a logical, I return a warning and coerce that argument to whatever default logical I originally defined.

warning("Oops")

Warning: Oops

Error

Errors indicate that that the function cannot continue and execution must stop. Including custom input checks with informative messages in an important facet of package development! And in non-function code, error messages are your first indicating that something is not working as it should.

stop("Oh no")
#> Error: Oh no

1

I included what this looks like as a commented-out line because otherwise the website can’t render this page.