Update to 0.10.2

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: LightLogR
 Title: Process Data from Wearable Light Loggers and Optical Radiation Dosimeters
-Version: 0.10.1
+Version: 0.10.2
 Authors@R: c(
     person("Johannes", "Zauner", 
       email = "johannes.zauner@tum.de", role = c("aut", "cre"),

NEWS.md

@@ -1,3 +1,7 @@
+# LightLogR 0.10.2
+
+* `add_states()` no longer creates duplicate observations at identical start/endpoints. The default behavior is now that the `start` is inclusive and `end` is exclusive. Adjust the new `bounds` argument for more options.
+
 # LightLogR 0.10.1
 
 * `extract_metric()` give more helpful error messages when using a base dataset without start/end columns.

R/add_states.R

@@ -36,6 +36,10 @@
 #'   will be forced to the same time zone as the `dataset` using
 #'   [lubridate::force_tz()]. If `FALSE` (default), the start and end times of
 #'   the `States.dataset` will be used as is.
+#' @param bounds How to treat the edge case, when a state start or endpoint
+#'   falls onto a data point. See [dplyr::join_by()] for details on the bounds
+#'   argument (Join types > Overlap joins > between). Basically `[` or `]`
+#'   signal an inclusive joint, whereas `(` or `)` signal exclusivity.
 #'
 #' @returns a modified `dataset` with the states added. The states are added as
 #'   new columns to the `dataset`. The columns are named after the columns in
@@ -65,6 +69,7 @@ add_states <- function(dataset,
                        start.colname = start,
                        end.colname = end,
                        force.tz = FALSE,
+                       bounds = c("[)", "[]", "(]", "()"),
                        leave.out = c("duration", "epoch")
 ){
   # Initial Checks ----------------------------------------------------------
@@ -75,6 +80,8 @@ add_states <- function(dataset,
     "States.dataset is not a dataframe" = is.data.frame(States.dataset)
   )
 
+  bounds <- rlang::arg_match(bounds)
+
   # Check if Datetime.colname is part of the dataset
   Datetime.colname.defused <- colname.defused({{ Datetime.colname }})
   if (!Datetime.colname.defused %in% names(dataset)) {
@@ -171,8 +178,12 @@ add_states <- function(dataset,
       by =
         dplyr::join_by(
           !!!groups,
-          {{ Datetime.colname }} >= {{ start.colname}},
-          {{ Datetime.colname }} <= {{ end.colname }}
+          dplyr::between({{ Datetime.colname }},
+          {{ start.colname}},
+          {{ end.colname }},
+          bounds = !!bounds)
+          # {{ Datetime.colname }} >= {{ start.colname}},
+          # {{ Datetime.colname }} <= {{ end.colname }}
         )
     ) |>
     dplyr::select(-{{ start.colname}}, -{{ end.colname }})

tests/testthat/test-extract_states.R

@@ -195,7 +195,7 @@ test_that("add_states basic functionality", {
     Intensity = c(5, 3)
   ) # No group needed if main has only one group that matches
 
-  result <- add_states(main_data, states_data, Datetime.colname = Timestamp)
+  result <- add_states(main_data, states_data, Datetime.colname = Timestamp, bounds = "[]")
 
   expect_s3_class(result, "tbl_df")
   expect_equal(nrow(result), 4) # Same as main_data
@@ -226,7 +226,7 @@ test_that("add_states with multiple groups and NAs for non-matching times", {
     StateInfo = c("StateA_Mid", "StateB_Start")
   ) %>% dplyr::group_by(Id) # Match grouping
 
-  result <- add_states(main_data, states_data)
+  result <- add_states(main_data, states_data, bounds = "[]")
   # For Id A: Datetime 0,1,2. State defined for 1h.
   # For Id B: Datetime 0,1,2. State defined for 0h.
   expect_equal(result$StateInfo[result$Id == "A"], c(NA, "StateA_Mid", NA))