| 1 | #' getData |
| 2 | #' |
| 3 | #' Acquire data as a Data object; see ?Data. |
| 4 | #' |
| 5 | #' Since series are given in columns (database format), this function builds series one |
| 6 | #' by one and incrementally grows a Data object which is finally returned. |
| 7 | #' |
| 8 | #' @param ts_data Time-series, as a data frame (DB style: 2 columns, first is date/time, |
| 9 | #' second is value) or a CSV file. |
| 10 | #' @param exo_data Exogenous variables, as a data frame or a CSV file; first column is |
| 11 | #' dates, next block are measurements for the day, and final block are exogenous |
| 12 | #' forecasts (for the same day). |
| 13 | #' @param date_format How date/time are stored (e.g. year/month/day hour:minutes; |
| 14 | #' see ?strptime) |
| 15 | #' @param limit Number of days to extract (default: Inf, for "all") |
| 16 | #' |
| 17 | #' @return An object of class Data |
| 18 | #' |
| 19 | #' @examples |
| 20 | #' ts_data = read.csv(system.file("extdata","pm10_mesures_H_loc.csv",package="talweg")) |
| 21 | #' exo_data = read.csv(system.file("extdata","meteo_extra_noNAs.csv",package="talweg")) |
| 22 | #' data = getData(ts_data, exo_data, limit=120) |
| 23 | #' @export |
| 24 | getData = function(ts_data, exo_data, date_format="%d/%m/%Y %H:%M", limit=Inf) |
| 25 | { |
| 26 | # Sanity checks (not full, but sufficient at this stage) |
| 27 | if ( (!is.data.frame(ts_data) && !is.character(ts_data)) || |
| 28 | (!is.data.frame(exo_data) && !is.character(exo_data)) ) |
| 29 | stop("Bad time-series / exogenous input (data frame or CSV file)") |
| 30 | if (is.character(ts_data)) |
| 31 | ts_data = ts_data[1] |
| 32 | if (is.character(exo_data)) |
| 33 | exo_data = exo_data[1] |
| 34 | if (!is.character(date_format)) |
| 35 | stop("Bad date_format (character)") |
| 36 | date_format = date_format[1] |
| 37 | if (!is.numeric(limit) || limit < 0) |
| 38 | stop("limit: positive integer") |
| 39 | |
| 40 | ts_df = |
| 41 | if (is.character(ts_data)) |
| 42 | read.csv(ts_data) |
| 43 | else |
| 44 | ts_data |
| 45 | # Convert to GMT (pretend it's GMT; no impact) |
| 46 | dates_POSIXlt = strptime(as.character(ts_df[,1]), date_format, tz="GMT") |
| 47 | ts_df[,1] = format(as.POSIXct(dates_POSIXlt, tz="GMT"), tz="GMT", usetz=TRUE) |
| 48 | |
| 49 | exo_df = |
| 50 | if (is.character(exo_data)) |
| 51 | read.csv(exo_data) |
| 52 | else |
| 53 | exo_data |
| 54 | # Times in exogenous variables file are ignored: no conversions required |
| 55 | |
| 56 | line = 1 #index in PM10 file (24 lines for 1 cell) |
| 57 | nb_lines = nrow(ts_df) |
| 58 | nb_exos = ( ncol(exo_df) - 1 ) / 2 |
| 59 | data = Data$new() |
| 60 | i = 1 #index of a cell in data |
| 61 | while (line <= nb_lines) |
| 62 | { |
| 63 | time = c() |
| 64 | serie = c() |
| 65 | level_hat = c() |
| 66 | repeat |
| 67 | { |
| 68 | { |
| 69 | time = c(time, ts_df[line,1]) |
| 70 | serie = c(serie, ts_df[line,2]) |
| 71 | level_hat = c(level_hat, #if data file is incomplete... |
| 72 | ifelse(ncol(ts_df) > 2, ts_df[line,3], mean(serie,na.rm=TRUE))) |
| 73 | line = line + 1 |
| 74 | }; |
| 75 | if (line >= nb_lines + 1 |
| 76 | || as.POSIXlt(ts_df[line-1,1],tz="GMT")$hour == 0) |
| 77 | { |
| 78 | break |
| 79 | } |
| 80 | } |
| 81 | |
| 82 | data$append(time=time, value=serie, level_hat=level_hat, |
| 83 | exo=exo_df[i,2:(1+nb_exos)], exo_hat=exo_df[i,(1+nb_exos+1):(1+2*nb_exos)]) |
| 84 | if (i >= limit) |
| 85 | break |
| 86 | i = i + 1 |
| 87 | } |
| 88 | data |
| 89 | } |