intermediate: R6, too slow

[talweg.git] / pkg / R / getData.R

#' @title Acquire data in a clean format
#'
#' @description Take in input data frames and/or files containing raw data, and timezones, and
#'   output a Data object, roughly corresponding to a list where each cell contains all value
#'   for one day (see \code{?Data}).
#'
#' @param ts_data Time-series, as a data frame (DB style: 2 columns, first is date/time,
#'   second is value) or a CSV file
#' @param exo_data Exogenous variables, as a data frame or a CSV file; first comlumn is dates,
#'   next block are measurements for the day, and final block are exogenous forecasts
#' @param input_tz Timezone in the input files ("GMT" or e.g. "Europe/Paris")
#' @param date_format How date/time are stored (e.g. year/month/day hour:minutes;
#'   see \code{strptime})
#' @param working_tz Timezone to work with ("GMT" or e.g. "Europe/Paris")
#' @param predict_at When does the prediction take place ? Integer, in hours. Default: 0
#' @param limit Number of days to extract (default: Inf, for "all")
#'
#' @return A list where data[[i]] contains
#'   \itemize{
#'     \item time: vector of times
#'     \item centered_serie: centered serie
#'     \item level: corresponding level
#'     \item exo: exogenous variables
#'     \item exo_hat: predicted exogenous variables
#'   }
#'
#' @examples
#' ts_data = read.csv(system.file("extdata","pm10_mesures_H_loc.csv",package="talweg"))
#' exo_data = read.csv(system.file("extdata","meteo_extra_noNAs.csv",package="talweg"))
#' data = getData(ts_data, exo_data, limit=120)
#' @export
getData = function(ts_data, exo_data, input_tz="GMT", date_format="%d/%m/%Y %H:%M",
	working_tz="GMT", predict_at=0, limit=Inf)
{
	# Sanity checks (not full, but sufficient at this stage)
	if (!is.character(input_tz) || !is.character(working_tz))
		stop("Bad timezone (see ?timezone)")
	input_tz = input_tz[1]
	working_tz = working_tz[1]
	if ( (!is.data.frame(ts_data) && !is.character(ts_data)) ||
			(!is.data.frame(exo_data) && !is.character(exo_data)) )
		stop("Bad time-series / exogenous input (data frame or CSV file)")
	if (is.character(ts_data))
		ts_data = ts_data[1]
	if (is.character(exo_data))
		exo_data = exo_data[1]
	predict_at = as.integer(predict_at)[1]
	if (predict_at<0 || predict_at>23)
		stop("Bad predict_at (0-23)")
	if (!is.character(date_format))
		stop("Bad date_format (character)")
	date_format = date_format[1]

	ts_df =
		if (is.character(ts_data))
			read.csv(ts_data)
		else
			ts_data
	exo_df =
		if (is.character(exo_data))
			read.csv(exo_data)
		else
			exo_data
	# Convert to the desired timezone (usually "GMT" or "Europe/Paris")
	formatted_dates_POSIXlt = strptime(as.character(ts_df[,1]), date_format, tz=input_tz)
	ts_df[,1] = format(as.POSIXct(formatted_dates_POSIXlt), tz=working_tz, usetz=TRUE)

	line = 1 #index in PM10 file (24 lines for 1 cell)
	nb_lines = nrow(ts_df)
	nb_exos = ( ncol(exo_df) - 1 ) / 2
	data = list()
	i = 1 #index of a cell in data
	while (line <= nb_lines)
	{
		time = c()
		serie = c()
		repeat
		{
			{
				time = c(time, ts_df[line,1])
				serie = c(serie, ts_df[line,2])
				line = line + 1
			};
			if (line >= nb_lines + 1 || as.POSIXlt(ts_df[line-1,1])$hour == predict_at)
				break
		}

		exo = as.data.frame( exo_df[i,2:(1+nb_exos)] )
		exo_hat = as.data.frame( exo_df[i,(1+nb_exos+1):(1+2*nb_exos)] )
		level = mean(serie, na.rm=TRUE)
		centered_serie = serie - level
		data[[i]] = list("time"=time, "centered_serie"=centered_serie, "level"=level,
			"exo"=exo, "exo_hat"=exo_hat)
		if (i >= limit)
			break
		i = i + 1
	}
	start = 1
	end = length(data)
	if (length(data[[1]]$centered_serie) < length(data[[2]]$centered_serie))
		start = 2
	if (length(data[[length(data)]]$centered_serie) <
		length(data[[length(data)-1]]$centered_serie))
	{
		end = end-1
	}
	if (start>1 || end<length(data))
		data = data[start:end]
	data
}