Merge pull request #7 from tscnlab/more_imports

R/gg_day.r

@@ -17,6 +17,15 @@
 #' [symlog_trans()] for details on tweaking this scale. The scale can also be
 #' changed to a normal or logarithmic scale - see the y.scale argument for more.
 #'
+#' The default scaling of the color and fill scales is discrete, with the
+#' [ggsci::scale_color_jco()] and [ggsci::scale_fill_jco()] scales. To use a
+#' continuous scale, use the `jco_color = FALSE` setting. Both `fill` and
+#' `color` aesthetics are set to `NULL` by default. For most geoms, this is not
+#' important, but geoms that automatically use those aesthetics (like
+#' geom_bin2d, where fill = stat(count)) are affected by this. Manually adding
+#' the required aesthetic (like `aes_fill = ggplot2::stat(count)` will fix
+#' this).
+#'
 #' @param dataset A light logger dataset. Expects a `dataframe`. If not imported
 #'   by [LightLogR], take care to choose a sensible variable for the `x.axis.`.
 #' @param x.axis,y.axis column name that contains the datetime (x, defaults to
@@ -82,7 +91,6 @@
 #' sample.data.environment,
 #' scales = "fixed",
 #' end.date = "2023-08-16",
-#' x.axis = Datetime,
 #' y.axis.label = "mEDI (lx)",
 #' aes_col = Id)
 #' plot
@@ -210,10 +218,11 @@ gg_day <- function(dataset,
     ggplot2::ggplot(ggplot2::aes(x=Time.data, y = !!y)) +
     eval(geom_function_expr)(
       ggplot2::aes(
+        x=Time.data, y = !!y,
         group = {{ group }},
         col = {{ aes_col }},
         fill = {{ aes_fill }},
-      ), ...) +
+        ), ...) +
     ribbon +
     # Scales --------------------------------------------------------------
     jco_color_scheme+

R/import_LL.R

@@ -204,6 +204,24 @@ imports <- function(device,
 }
 
 import_arguments <- list(
+  #SpectraWear
+  SpectraWear = rlang::expr({
+    tmp <-suppressMessages( 
+      readr::read_csv(filename,
+                      n_max = n_max,
+                      id = "file.name",
+                      locale = locale,
+                      name_repair = "universal",
+                      ...
+      ))
+    tmp <- tmp %>%
+      dplyr::rename(MEDI = Mel
+                    ) %>% 
+      dplyr::mutate(Datetime =
+                      lubridate::dmy_hms(paste(Date, Time), tz = tz),
+                    Id = paste(.data$id, .data$ls, sep = ".")
+                      )
+  }),
   #Speccy
   Speccy = rlang::expr({
     tmp <-suppressMessages( 
@@ -228,6 +246,7 @@ import_arguments <- list(
                       id = "file.name",
                       locale = locale,
                       name_repair = "universal",
+                      col_types = c("fccdddddddddddddddddd"),
                       ...
       ))
     tmp <- tmp %>%

inst/CITATION

@@ -11,6 +11,6 @@ bibentry(
   pages    = "",
   url      = "https://github.com/tscnlab/LightLogR",
   textVersion = paste(
-  "Zauner, J.; Spitschan, M. (2023): LightLogR: Working With Wearable Light Logger Data. R Package, Available on https://github.com/tscnlab/LightLogR. Funded by EURAMET, co-funded by the European Union, Grant#22NRM05 MeLiDos"
+  "Zauner, J.; Spitschan, M. (2023): LightLogR: Working With Wearable Light Logger Data. R Package, Available on https://github.com/tscnlab/LightLogR"
   )
 )

vignettes/articles/Day.Rmd

@@ -281,16 +281,18 @@ dataset.LL.partial  %>%
 
 To create a boxplot representation, we not only need to specify the geom but also what time interval we want the boxplot to span. Here the `cut_Datetime()` function from **LightLogR** comes to the rescue. It will round the datetimes to the desired interval, which then can be specified as the `group` argument of `gg_day()`. While this can be a nice representation, I don´t think it fits our goal for the overall figure in our specific case.
 
-#### `geom_hexplot`
+#### `geom_bin2d`
 
 ```{r, fig.retina=2, fig.width=7, fig.height = 4, warning=FALSE}
-dataset.LL.partial  %>% 
-  gg_day(
-    size = 0.25, facetting = FALSE, geom = "bin2d", bins = 24) + #base plot
+dataset.LL.partial  %>% gg_day(
+  size = 0.25, facetting = FALSE, geom = "bin2d", 
+  jco_color = FALSE, bins = 24, aes_fill = stat(count)) + #base plot
   solar.reference + brown.reference + scale.correction
 ```
 
-The geom family of `hex`, `bin2d`, or `density_2d` is particularly well suited if you have many, possibly overlaying observations. It reduces complexity by cutting the x- and y-axis into bins and counts how many observations fall within this bin. By choosing 24 bins, we see dominant values for every hour of the day.
+The geom family of `hex`, `bin2d`, or `density_2d` is particularly well suited if you have many, possibly overlaying observations. It reduces complexity by cutting the x- and y-axis into bins and counts how many observations fall within this bin. By choosing 24 bins, we see dominant values for every hour of the day. The `jco_color = FALSE` argument is necessary to disable the default discrete color scheme of `gg_day()`, because a continuous scale is necessary for counts or densities. Finally, we habe to use the `aes_fill = stat(count)` argument to color the bins according to the number of observations in the bin[^4].
+
+[^4]: Normally, using `ggplot2::geom_bin2d()` would automatically use the fill aesthetic for this, but `gg_day()` has certain defaults that are favorable in most cases, but is a bit more work here.
 
 ###  {.unnumbered}