Skip to contents

This function defines the column types for the article data frame used in the epireview package. readr is generally good at guessing the column types, but it is better to be explicit. Moreover, it reads a column of NAs as a logical vector, which is particularly undesirable for us. The function is intended to be used internally by load_epidata_raw where the files are being read.

Usage

article_column_type(pathogen)

Arguments

pathogen

name of pathogen. Must be one of the priority pathogens You can get a list of the priority pathogens currently included in the package by calling the function priority_pathogens.

Value

A list of column types for the article data frame

See also

parameter_column_type, outbreak_column_type, model_column_type