feat: allow dynamic signals (#56)

R/workpatterns_classify.R

@@ -29,8 +29,11 @@
 #' and "abs".
 #'
 #' @param signals Character vector to specify which collaboration metrics to use:
-#' You may use "email" (default) for emails only, "IM" for Teams messages only,
-#' or a combination of the two `c("email", "IM")`.
+#'   - "email" (default) for emails only
+#'   - "IM" for Teams messages only,
+#'   - "unscheduled_calls" for Unscheduled Calls only
+#'   - "meetings" for Meetings only
+#'   - or a combination of signals, such as `c("email", "IM")`
 #'
 #' @param start_hour A character vector specifying starting hours,
 #' e.g. "0900". Note that this currently only supports **hourly** increments.

R/workpatterns_classify_bw.R

@@ -66,54 +66,16 @@ workpatterns_classify_bw <- function(data,
   end_hour <- as.numeric(gsub(pattern = "00$", replacement = "", x = end_hour))
 
   # Calculate hours within working hours
-  d <-(end_hour - start_hour)-1
+  d <- (end_hour - start_hour) - 1
 
-  ## Select input variable names
-  if("email" %in% signals & "IM" %in% signals){
+  # Text replacement only for allowed values
 
-    ## Create 24 summed `Signals_sent` columns
-    signal_cols <-
-      purrr::map(0:23, ~combine_signals(data2, hr = .)) %>%
-      dplyr::bind_cols()
+  if(any(signals %in% c("email", "IM", "unscheduled_calls", "meetings"))){
 
-    ## Use names for matching
-    input_var <- names(signal_cols)
-
-    ## Signals sent by Person and date
-    signals_df <-
-      data2 %>%
-      .[, c("PersonId", "Date")] %>%
-      cbind(signal_cols)
-
-    ## Signal label
-    sig_label <- "Signals_sent"
-
-  } else if(signals == "IM"){
-
-    match_index <- grepl(pattern = "^IMs_sent", x = names(data2))
-    input_var <- names(data2)[match_index]
-    input_var2 <- c("PersonId", "Date", input_var)
-
-    ## signals sent by Person and date
-    signals_df <-
-      data2 %>%
-      .[, ..input_var2]
-
-    sig_label <- "IMs_sent"
-
-
-  } else if(signals == "email"){
-
-    match_index <- grepl(pattern = "^Emails_sent", x = names(data2))
-    input_var <- names(data2)[match_index]
-    input_var2 <- c("PersonId", "Date", input_var)
-
-    ## signals sent by Person and date
-    signals_df <-
-      data2 %>%
-      .[, ..input_var2]
-
-    sig_label <- "Emails_sent"
+    signal_set <- gsub(pattern = "email", replacement = "Emails_sent", x = signals) # case-sensitive
+    signal_set <- gsub(pattern = "IM", replacement = "IMs_sent", x = signal_set)
+    signal_set <- gsub(pattern = "unscheduled_calls", replacement = "Unscheduled_calls", x = signal_set)
+    signal_set <- gsub(pattern = "meetings", replacement = "Meetings", x = signal_set)
 
   } else {
 
@@ -121,6 +83,75 @@ workpatterns_classify_bw <- function(data,
 
   }
 
+  ## Create 24 summed `Signals_sent` columns
+  signal_cols <- purrr::map(0:23, ~combine_signals(data, hr = ., signals = signal_set))
+  signal_cols <- bind_cols(signal_cols)
+
+  ## Use names for matching
+  input_var <- names(signal_cols)
+
+  ## Signals sent by Person and Date
+  signals_df <-
+    data2 %>%
+    .[, c("PersonId", "Date")] %>%
+    cbind(signal_cols)
+
+  ## Signal label
+  sig_label <- ifelse(length(signal_set) > 1, "Signals_sent", signal_set)
+
+  # ## Select input variable names
+  # if("email" %in% signals & "IM" %in% signals){
+  #
+  #   ## Create 24 summed `Signals_sent` columns
+  #   signal_cols <-
+  #     purrr::map(0:23, ~combine_signals(data2, hr = .)) %>%
+  #     dplyr::bind_cols()
+  #
+  #   ## Use names for matching
+  #   input_var <- names(signal_cols)
+  #
+  #   ## Signals sent by Person and date
+  #   signals_df <-
+  #     data2 %>%
+  #     .[, c("PersonId", "Date")] %>%
+  #     cbind(signal_cols)
+  #
+  #   ## Signal label
+  #   sig_label <- "Signals_sent"
+  #
+  # } else if(signals == "IM"){
+  #
+  #   match_index <- grepl(pattern = "^IMs_sent", x = names(data2))
+  #   input_var <- names(data2)[match_index]
+  #   input_var2 <- c("PersonId", "Date", input_var)
+  #
+  #   ## signals sent by Person and date
+  #   signals_df <-
+  #     data2 %>%
+  #     .[, ..input_var2]
+  #
+  #   sig_label <- "IMs_sent"
+  #
+  #
+  # } else if(signals == "email"){
+  #
+  #   match_index <- grepl(pattern = "^Emails_sent", x = names(data2))
+  #   input_var <- names(data2)[match_index]
+  #   input_var2 <- c("PersonId", "Date", input_var)
+  #
+  #   ## signals sent by Person and date
+  #   signals_df <-
+  #     data2 %>%
+  #     .[, ..input_var2]
+  #
+  #   sig_label <- "Emails_sent"
+  #
+  # } else {
+  #
+  #   stop("Invalid input for `signals`.")
+  #
+  # }
+
 
   ## Create binary variable 0 or 1
   num_cols <- names(which(sapply(signals_df, is.numeric))) # Get numeric columns

R/workpatterns_classify_pav.R

@@ -49,57 +49,14 @@ workpatterns_classify_pav <- function(data,
   start_hour <- as.numeric(sub(pattern = "00$", replacement = "", x = start_hour))
   end_hour <- as.numeric(sub(pattern = "00$", replacement = "", x = end_hour))
 
-  ## Select input variable names
-  if("email" %in% signals & "IM" %in% signals){
+  # Text replacement only for allowed values
 
-    ## Create 24 summed `Signals_sent` columns
-    signal_cols <-
-      purrr::map(0:23,
-                 ~combine_signals(data,
-                                  hr = .)) %>%
-      bind_cols()
+  if(any(signals %in% c("email", "IM", "unscheduled_calls", "meetings"))){
 
-    ## Use names for matching
-    input_var <- names(signal_cols)
-
-    ## Average signals sent by Person
-    signals_df <-
-      data %>%
-      select(PersonId) %>%
-      cbind(signal_cols) %>%
-      group_by(PersonId) %>%
-      summarise_all(~mean(.))
-
-    ## Signal label
-    sig_label <- "Signals_sent"
-
-  } else if(signals == "IM"){
-
-    match_index <- grepl(pattern = "^IMs_sent", x = names(data))
-    input_var <-names(data)[match_index]
-
-    ## Average signals sent by Person
-    signals_df <-
-      data %>%
-      select(PersonId, all_of(input_var)) %>%
-      group_by(PersonId) %>%
-      summarise_all(~mean(.))
-
-    sig_label <- "IMs_sent"
-
-  } else if(signals == "email"){
-
-    match_index <- grepl(pattern = "^Emails_sent", x = names(data))
-    input_var <-names(data)[match_index]
-
-    ## Average signals sent by Person
-    signals_df <-
-      data %>%
-      select(PersonId, all_of(input_var)) %>%
-      group_by(PersonId) %>%
-      summarise_all(~mean(.))
-
-    sig_label <- "Emails_sent"
+    signal_set <- gsub(pattern = "email", replacement = "Emails_sent", x = signals) # case-sensitive
+    signal_set <- gsub(pattern = "IM", replacement = "IMs_sent", x = signal_set)
+    signal_set <- gsub(pattern = "unscheduled_calls", replacement = "Unscheduled_calls", x = signal_set)
+    signal_set <- gsub(pattern = "meetings", replacement = "Meetings", x = signal_set)
 
   } else {
 
@@ -107,6 +64,82 @@ workpatterns_classify_pav <- function(data,
 
   }
 
+  ## Create 24 summed `Signals_sent` columns
+  signal_cols <- purrr::map(0:23, ~combine_signals(data, hr = ., signals = signal_set))
+  signal_cols <- bind_cols(signal_cols)
+
+  ## Use names for matching
+  input_var <- names(signal_cols)
+
+  ## Average signals sent by Person
+  signals_df <-
+    data %>%
+    select(PersonId) %>%
+    cbind(signal_cols) %>%
+    group_by(PersonId) %>%
+    summarise_all(~mean(.))
+
+  ## Signal label
+  sig_label <- ifelse(length(signal_set) > 1, "Signals_sent", signal_set)
+
+  # ## Select input variable names
+  # if("email" %in% signals & "IM" %in% signals){
+  #
+  #   ## Create 24 summed `Signals_sent` columns
+  #   signal_cols <-
+  #     purrr::map(0:23,
+  #                ~combine_signals(data,
+  #                                 hr = .)) %>%
+  #     bind_cols()
+  #
+  #   ## Use names for matching
+  #   input_var <- names(signal_cols)
+  #
+  #   ## Average signals sent by Person
+  #   signals_df <-
+  #     data %>%
+  #     select(PersonId) %>%
+  #     cbind(signal_cols) %>%
+  #     group_by(PersonId) %>%
+  #     summarise_all(~mean(.))
+  #
+  #   ## Signal label
+  #   sig_label <- "Signals_sent"
+  #
+  # } else if(signals == "IM"){
+  #
+  #   match_index <- grepl(pattern = "^IMs_sent", x = names(data))
+  #   input_var <-names(data)[match_index]
+  #
+  #   ## Average signals sent by Person
+  #   signals_df <-
+  #     data %>%
+  #     select(PersonId, all_of(input_var)) %>%
+  #     group_by(PersonId) %>%
+  #     summarise_all(~mean(.))
+  #
+  #   sig_label <- "IMs_sent"
+  #
+  # } else if(signals == "email"){
+  #
+  #   match_index <- grepl(pattern = "^Emails_sent", x = names(data))
+  #   input_var <-names(data)[match_index]
+  #
+  #   ## Average signals sent by Person
+  #   signals_df <-
+  #     data %>%
+  #     select(PersonId, all_of(input_var)) %>%
+  #     group_by(PersonId) %>%
+  #     summarise_all(~mean(.))
+  #
+  #   sig_label <- "Emails_sent"
+  #
+  # } else {
+  #
+  #   stop("Invalid input for `signals`.")
+  #
+  # }
+
 
   ## Normalised pattern data
   ptn_data_norm <-

R/workpatterns_hclust.R

@@ -28,8 +28,11 @@
 #' or absolute values in "data" and "plot". Valid values are "percent" (default)
 #' and "abs".
 #' @param signals Character vector to specify which collaboration metrics to use:
-#' You may use "email" (default) for emails only, "IM" for Teams messages only,
-#' "unscheduled_calls" for Unscheduled Calls only, or a combination, such as `c("email", "IM")`.
+#'   - "email" (default) for emails only
+#'   - "IM" for Teams messages only,
+#'   - "unscheduled_calls" for Unscheduled Calls only
+#'   - "meetings" for Meetings only
+#'   - or a combination of signals, such as `c("email", "IM")`
 #' @param start_hour A character vector specifying starting hours,
 #' e.g. "0900"
 #' @param end_hour A character vector specifying starting hours,

man/workpatterns_classify.Rd

@@ -28,8 +28,13 @@ or absolute values in "data" and "plot". Valid values are "percent" (default)
 and "abs".}
 
 \item{signals}{Character vector to specify which collaboration metrics to use:
-You may use "email" (default) for emails only, "IM" for Teams messages only,
-or a combination of the two \code{c("email", "IM")}.}
+\itemize{
+\item "email" (default) for emails only
+\item "IM" for Teams messages only,
+\item "unscheduled_calls" for Unscheduled Calls only
+\item "meetings" for Meetings only
+\item or a combination of signals, such as \code{c("email", "IM")}
+}}
 
 \item{start_hour}{A character vector specifying starting hours,
 e.g. "0900". Note that this currently only supports \strong{hourly} increments.}

man/workpatterns_hclust.Rd

@@ -30,8 +30,13 @@ or absolute values in "data" and "plot". Valid values are "percent" (default)
 and "abs".}
 
 \item{signals}{Character vector to specify which collaboration metrics to use:
-You may use "email" (default) for emails only, "IM" for Teams messages only,
-"unscheduled_calls" for Unscheduled Calls only, or a combination, such as \code{c("email", "IM")}.}
+\itemize{
+\item "email" (default) for emails only
+\item "IM" for Teams messages only,
+\item "unscheduled_calls" for Unscheduled Calls only
+\item "meetings" for Meetings only
+\item or a combination of signals, such as \code{c("email", "IM")}
+}}
 
 \item{start_hour}{A character vector specifying starting hours,
 e.g. "0900"}