Data keys

One option for converting the input files to the internal data frames is a unique script. However much of the conversion is route so a genetic keyed template function soilDataR::processDataTemplet is provided. This function requires a key that links the input files to the internal data representation. The key must be a data frame with the original headers (header), what variable those headers refer to (var), what data frame that variable is in (dataframe), what class the variable is (class), what type of data is in that header file (type), and what unit is in the column if it's not a reference (unit) Our example case would look something like this:

header	var	dataframe	class	type	unit
study name	dataset_name	study	factor	NA	NA
author name	curator_name	study	factor	NA	NA
institute	curator_organization	study	factor	NA	NA
site_name	site_name	field	factor	NA	NA
profile_name	profile_name	field	factor	NA	NA
latitude	lat	field	numeric	NA	NA
longitude	lon	field	numeric	NA	NA
date	observation_date	field	factor	NA	NA
biome	biome	field	factor	NA	NA
soil org C	soc	sample	numeric	value	NA
soil org C (sigma)	soc	sample	numeric	sigma	NA
soil org C (unit)	soc	sample	factor	unit	NA
soil org C (method)	soc	sample	character	method	NA
bulk density (g cm-3)	bd_sample	sample	numeric	value	g cm-3
pH	pH	sample	numeric	value	NA
time on shelf (yrs)	treatment	treatment	character	method	NA