wrap weight aesthetic into the dist aesthetic for #41

NAMESPACE

@@ -263,6 +263,7 @@ importFrom(distributional,generate)
 importFrom(distributional,hdr)
 importFrom(distributional,is_distribution)
 importFrom(distributional,support)
+importFrom(distributional,variance)
 importFrom(dplyr,"%>%")
 importFrom(dplyr,across)
 importFrom(dplyr,arrange_at)

R/density.R

@@ -358,7 +358,7 @@ density_histogram = auto_partial(name = "density_histogram", function(
   x_label = as_label(enexpr(x))
   x = check_na(x, na.rm)
 
-  h = weighted_hist(x, breaks = breaks, align = align)
+  h = weighted_hist(x, weights = weights, breaks = breaks, align = align)
   input_1 = h$breaks[-length(h$breaks)]  # first edge of bin
   input_2 = h$breaks[-1]                 # second edge of bin
   input_ = (input_1 + input_2)/2   # center of bin

R/distributions.R

@@ -186,7 +186,7 @@ distr_is_factor_like = function(dist) {
   inherits(dist, "rvar_factor") || if (inherits(dist, "distribution")) {
     is_factor_like = map_lgl_(vctrs::vec_data(dist), function(d) {
       inherits(d, c("dist_categorical", "ggdist__wrapped_categorical")) ||
-        (inherits(d, "dist_sample") && inherits(distr_get_sample(d), c("character", "factor"))) ||
+        (inherits(d, c("dist_sample", "ggdist__weighted_sample")) && inherits(distr_get_sample(d), c("character", "factor"))) ||
         is.character(vctrs::field(support(vec_restore(list(d), dist_missing())), "x")[[1]])
     })
     length(dist) > 0 && all(is_factor_like)
@@ -445,6 +445,7 @@ mean.ggdist__weighted_sample = function(x, ...) {
   }
 }
 
+#' @importFrom distributional variance
 #' @export
 variance.ggdist__weighted_sample = function(x, ...) {
   if (is.null(x[["weights"]])){

R/stat_slabinterval.R

@@ -618,9 +618,9 @@ StatSlabinterval = ggproto("StatSlabinterval", AbstractStatSlabinterval,
       }
 
       # dist aesthetic is not provided but x aesthetic is, and x is not a dist
-      # this means we need to wrap it as a dist_sample
+      # this means we need to wrap it as a weighted dist_sample
       data = summarise_by(data, c("PANEL", y, "group"), function(d) {
-        data.frame(dist = dist_sample(list(trans$inverse(d[[x]]))))
+        data.frame(dist = .dist_weighted_sample(list(trans$inverse(d[[x]])), list(d[["weight"]])))
       })
       data[[x]] = median(data$dist)
     }