Added tz parameter to functions that use check_data_columns interna…

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: iglu
 Type: Package
 Title: Interpreting Glucose Data from Continuous Glucose Monitors
-Version: 4.1.6
+Version: 4.1.7
 Authors@R: c(person("Elizabeth", "Chun",
             role = c("aut")),
             person("Steve", "Broll",
@@ -41,7 +41,7 @@ Description: Implements a wide range of metrics for measuring glucose control an
 License: GPL-2
 Encoding: UTF-8
 LazyData: true
-RoxygenNote: 7.3.1
+RoxygenNote: 7.3.2
 Depends: R (>= 3.5.0)
 Imports:
     caTools,

NEWS.md

@@ -1,3 +1,6 @@
+# iglu 4.1.7
+* added tz parameter to functions that require it
+
 # iglu 4.1.6
 * Updates to the MAGE vignette
 * Minor documentation updates

R/active_percent.R

@@ -4,10 +4,12 @@
 #' The function `active_percent` produces the % of time CGM is active together with the length of the measurement period
 #'
 #' @usage
-#' active_percent(data, dt0 = NULL)
+#' active_percent(data, dt0 = NULL, tz = "")
 #'
 #' @inheritParams plot_lasagna
 #'
+#' @param tz \strong{tz = "".} A character string specifying the time zone to be used. System-specific (see \code{\link{as.POSIXct}}), but " " is the current time zone, and "GMT" is UTC (Universal Time, Coordinated). Invalid values are most commonly treated as UTC, on some platforms with a warning.
+#'
 #' @details
 #'The function `active_percent` produces a tibble object with values equal to the
 #' percentage of time the CGM was active, the total number of observed days, the start date, and the end date. For example, if a CGM's (5 min frequency) times were 0, 5, 10, 15 and
@@ -38,14 +40,14 @@
 #' data(example_data_5_subject)
 #'
 #' active_percent(example_data_5_subject)
-#' active_percent(example_data_5_subject, dt0 = 5)
+#' active_percent(example_data_5_subject, dt0 = 5, tz = 'GMT')
 #'
 
 
-active_percent <- function(data, dt0 = NULL) {
+active_percent <- function(data, dt0 = NULL, tz = "") {
   active_percent = gl = id = NULL
   rm(list = c("gl", "id", "active_percent"))
-  data = check_data_columns(data, time_check = TRUE)
+  data = check_data_columns(data, time_check = TRUE, tz = tz)
   is_vector = attr(data, "is_vector")
 
   subject = unique(data$id)

R/agp.R

@@ -35,7 +35,7 @@ agp <- function(data, maxd = 14, inter_gap = 45, dt0 = NULL, tz = "", daily = TR
   }
 
   # Calculate range of measurements
-  out_range = active_percent(data)
+  out_range = active_percent(data, tz = tz)
 
   # Calculate metrics
   tableStat = data.frame("Glucose statistics", "Value")

R/agp_metrics.R

@@ -6,13 +6,15 @@
 #' `cv_glu`, `below_percent`, `in_range_percent`, `above_percent`.
 #'
 #' @usage
-#' agp_metrics(data, shinyformat = FALSE)
+#' agp_metrics(data, shinyformat = FALSE, tz = '')
 #'
 #' @param data DataFrame object with column names "id", "time", and "gl".
 #'
 #' @param shinyformat \strong{Default: FALSE.} Boolean indicating whether the output should be formatted for the single subject
 #' AGP page in shiny.
 #'
+#' @param tz \strong{Default: "".} A character string specifying the time zone to be used. System-specific (see \code{\link{as.POSIXct}}), but " " is the current time zone, and "GMT" is UTC (Universal Time, Coordinated). Invalid values are most commonly treated as UTC, on some platforms with a warning.
+#'
 #' @return
 #' By default, a tibble object with 1 row for each subject, and 13 columns is returned:
 #' a column for subject id,
@@ -66,12 +68,12 @@
 #'
 
 
-agp_metrics <- function (data, shinyformat = FALSE) {
+agp_metrics <- function (data, shinyformat = FALSE, tz = "") {
 
   id = . = start_date = end_date = ndays = CV = GMI = NULL
   rm(list = c("id", ".", "start_date", "end_date", "ndays", "CV", "GMI"))
 
-  activity <- active_percent(data)
+  activity <- active_percent(data, tz = tz)
 
   out = list("Start Date" = activity[, c(1,4)],
              "End Date" = activity[, c(1,5)],

R/all_metrics.R

@@ -76,10 +76,10 @@ all_metrics <- function(data, dt0 = NULL, inter_gap = 45, tz = "", timelag = 15,
       "SD_GLU" = sd_glu(data),
       "Mean_GLU" = mean_glu(data),
       "CV_GLU" = cv_glu(data),
-      "Active_Percent" = active_percent(data), # TODO: potentially only keep the 'active_percent' column
+      "Active_Percent" = active_percent(data, tz = tz), # TODO: potentially only keep the 'active_percent' column
       "Percent_In_Tight_Range" = in_range_percent(data, target_ranges = list(c(70,140))),
       "GMI" = gmi(data),
-      "GRI" = gri(data),
+      "GRI" = gri(data, tz = tz),
       "Extended_Hypo" = extended_hypo,
       "Extended_Hyper" = extended_hyper
     )
@@ -92,7 +92,7 @@ all_metrics <- function(data, dt0 = NULL, inter_gap = 45, tz = "", timelag = 15,
                "eA1C" = ea1c(data),
                "episodes" = ep_out,
                "GMI" = gmi(data),
-               "GRI" = gri(data),
+               "GRI" = gri(data, tz = tz),
                "GRADE" = grade(data),
                "GRADE_Euglycemia" = grade_eugly(data),
                "GRADE_Hyperglycemia" = grade_hyper(data),
@@ -107,7 +107,7 @@ all_metrics <- function(data, dt0 = NULL, inter_gap = 45, tz = "", timelag = 15,
                "M_Value" = m_value(data),
                "Mad_GLU" = mad_glu(data),
                "MAGE" = mage(data),
-               "Percent_Active" = active_percent(data),
+               "Percent_Active" = active_percent(data, tz = tz),
                "Percent_Above" = above_percent(data),
                "Percent_Below" = below_percent(data),
                "Percent_In_Range" = in_range_percent(data),

R/gri.R

@@ -6,10 +6,12 @@
 #' GRI value. ' The output rows correspond to subjects.
 #'
 #' @usage
-#' gri(data)
+#' gri(data, tz = "")
 #'
 #' @inheritParams mean_glu
 #'
+#' @param tz \strong{Default: "".} A character string specifying the time zone to be used. System-specific (see \code{\link{as.POSIXct}}), but " " is the current time zone, and "GMT" is UTC (Universal Time, Coordinated). Invalid values are most commonly treated as UTC, on some platforms with a warning.
+#'
 #' @details
 #' A tibble object with 1 row for each subject, a column for subject id and
 #' column for GRI is returned. The formula for GRI is as follows:
@@ -40,10 +42,10 @@
 #' gri(example_data_1_subject)
 #'
 #' data(example_data_5_subject)
-#' gri(example_data_5_subject)
+#' gri(example_data_5_subject, tz = 'GMT')
 #'
 
-gri <- function(data){
+gri <- function(data, tz = ""){
 
   gri_single <- function(data) {
     # get percent in each range from agp aggregated metrics
@@ -61,7 +63,7 @@ gri <- function(data){
 
   id = NULL
   rm(list = c("id"))
-  data = check_data_columns(data, time_check = TRUE)
+  data = check_data_columns(data, time_check = TRUE, tz = tz)
 
   out = data %>%
     dplyr::group_by(id) %>%

R/heatmap.R

@@ -6,6 +6,7 @@
 #' @param clustering_method the agglomeration method for hierarchical clustering, accepts same values as \code{\link{hclust}}, default value is 'complete'
 #' @param clustering_distance_metrics the distance measure for metrics clustering, accepts same values as \code{\link{dist}}, default value is 'correlation' distance
 #' @param clustering_distance_subjects the distance measure for subjects clustering, accepts same values as \code{\link{dist}}, default value is 'correlation' distance
+#' @param tz \strong{Default: "".} A character string specifying the time zone to be used. System-specific (see \code{\link{as.POSIXct}}), but " " is the current time zone, and "GMT" is UTC (Universal Time, Coordinated). Invalid values are most commonly treated as UTC, on some platforms with a warning.
 #'
 #' @return A heatmap of metrics by subjects generated via \code{\link{pheatmap}}
 #' @export
@@ -16,14 +17,14 @@
 #' metrics_heatmap(metrics = mecs)
 metrics_heatmap <- function(data = NULL, metrics = NULL, metric_cluster = 6, clustering_method = "complete",
                             clustering_distance_metrics = "correlation",
-                            clustering_distance_subjects = "correlation"){
+                            clustering_distance_subjects = "correlation", tz = ""){
   # Check whether pre-calculated metrics are provided
   mecs = metrics
   if(is.null(mecs)){
     if(is.null(data)){
       stop("Either CGM data or precalculated metrics must be provided")
     }else{
-      data = check_data_columns(data, time_check=TRUE)
+      data = check_data_columns(data, time_check=TRUE, tz = tz)
       # No metrics are provided, calculate all by default
       mecs = all_metrics(data)
     }

R/plot_meals.R

@@ -4,11 +4,13 @@
 #' The function plot_meals produces a visual for meals data
 #'
 #' @usage
-#' plot_meals(data, mealtimes, plot_type=c('ggplot','plotly'))
+#' plot_meals(data, mealtimes, plot_type=c('ggplot','plotly'), tz = "")
 #'
 #' @inheritParams meal_metrics
 #' @inheritParams mage_ma_single
-#' @param plot_type Default: "ggplot". One of 'ggplot', 'plotly'. Determines whether the function returns a static publication-ready image or an interactive GUI.
+#' @param plot_type \strong{Default: "ggplot".} One of 'ggplot', 'plotly'. Determines whether the function returns a static publication-ready image or an interactive GUI.
+#' @param tz \strong{Default: "".} A character string specifying the time zone to be used. System-specific (see \code{\link{as.POSIXct}}), but " " is the current time zone, and "GMT" is UTC (Universal Time, Coordinated). Invalid values are most commonly treated as UTC, on some platforms with a warning.
+#'
 #' @return Plot to visualize meals data.
 #'
 #' @export
@@ -34,10 +36,10 @@
 #'
 #' select_subject = example_data_hall[example_data_hall$id == "2133-018", ]
 #' select_meals = example_meals_hall[example_meals_hall$id == "2133-018", ]
-#' plot_meals(select_subject, select_meals)
+#' plot_meals(select_subject, select_meals, tz = 'GMT')
 
 
-plot_meals = function(data, mealtimes, plot_type=c('ggplot','plotly')) {
+plot_meals = function(data, mealtimes, plot_type=c('ggplot','plotly'), tz = '') {
 
   id = meal = mealtime = peaktime = recovertime = basegl = peakgl = recovergl =
     time_window = x = y = NULL
@@ -53,7 +55,7 @@ plot_meals = function(data, mealtimes, plot_type=c('ggplot','plotly')) {
     data = data[data$id == subject, ]
   }
 
-  metrics = meal_metrics(data, mealtimes, glucose_times = TRUE)
+  metrics = meal_metrics(data, mealtimes, glucose_times = TRUE, tz = tz)
 
   metrics_expanded = metrics %>%
     dplyr::rowwise() %>%

R/sd_measures.R

@@ -88,7 +88,7 @@ sd_measures <- function(data, dt0 = NULL, inter_gap = 45, tz = ""){
 
   id = NULL
   rm(list = c("id"))
-  data = check_data_columns(data, time_check=TRUE)
+  data = check_data_columns(data, time_check=TRUE, tz = tz)
   subject = unique(data$id)
   ns = length(subject)
 

R/utils.R

@@ -62,6 +62,10 @@ check_data_time <- function(data, tz = ""){
   if (!lubridate::is.POSIXct(data$time)){ # Check if already in date format
     tr = as.character(data$time)
     data$time = as.POSIXct(tr, format='%Y-%m-%d %H:%M:%S', tz = tz)
+
+    if (any(is.na(data$time))) {
+      stop('Automatic identification of timezone unsuccessful. Please manually set `tz` parameter or double check if the timezone is correct.')
+    }
   }
 
   out = data %>%