dispatch_on_field: Operator overloading function

In steveneschrich/pgreportr: NIH Program Grant Reporting Library

Operator overloading function

Description

For many cases, we would like to not expose the specifics of the indicator variables in the dataset explicitly, since function calls are the things that are well documented in the library.

Usage

dispatch_on_field(tb, name)

Arguments

tb	A data.frame or tibble to extract a value from. May be NULL to retrieve the name itself.
name	The name of the variable to extract

Details

Therefore, we essentially have getters for a variety of variables but with a small twist. The simplest form of usage is to say grants %>% is_phsu_cancer_related() or more simply, is_phsu_cancer_related(grants). This would return the underlying variable (is_phsu_cancer_related) which is already a logical. However, it would be clearly documented as a function call (rather than having to know the specific field).

The second, less clear usage of the function is when trying to use it in a dplyr statement. For instance, in filter: grants %>% filter(!!is_phsu_cancer_related()) allows us to filter on a field that I might not know the name of but is clearly documented within the function.

This function is the general-purpose operator overload dispatcher that figures out, based on whether or not an argument is provided, whether to return the variable name (for !!) or to return a list of values (if I provided a table).

In practical terms (within the library), you can then define a new function is_phsu_related() that just calls this dispatcher with the appropriate variable name. Essentially a java-like getter.

Value

Either a name (as a symbol, if the table value is NULL) or a list of values from the field.

steveneschrich/pgreportr documentation built on Jan. 13, 2025, 7:09 p.m.