Make dirmult a required dependency for clusterSize

DESCRIPTION

@@ -30,6 +30,7 @@ Depends:
     R (>= 4.1.0)
 Imports:
     data.table,
+    dirmult,
     glue,
     methods,
     mvnfast,
@@ -61,7 +62,6 @@ Suggests:
     broom.helpers,
     survminer,
     katex,
-    dirmult,
     rms,
     lmerTest
 LinkingTo: 

NEWS.md

@@ -1,5 +1,8 @@
 # simstudy (development version)
 
+* `clusterSize` now works after a standard installation: `dirmult` is a
+  required dependency rather than an optional suggestion (#257).
+
 ## New feature
 
 * Added optional argument `withinClusterID` to `genCluster()` to generate

R/generate_dist.R

@@ -655,14 +655,6 @@
 
 .genclustsize <- function(n, formula, variance = 0,  envir = parent.frame()) { 
 
-  if (!requireNamespace("dirmult", quietly = TRUE)) { # nocov start
-    stop(
-      "Package \"dirmult\" must be installed to use this function with 
-       the Poisson distribution.",
-        call. = FALSE
-    )
-  } # nocov end
-
   formula <- .evalWith(formula, .parseDotVars(formula, envir), envir = envir)[1]
   variance <- .evalWith(variance, .parseDotVars(variance, envir), envir = envir)[1]
 

tests/testthat/test-generate_dist.R

@@ -177,6 +177,45 @@ test_that("clusterSize data is generated as expected.", {
 })
 
 
+test_that("genData preserves clusterSize totals for both dispersion branches", {
+  for (dispersion in c(0, 0.05)) {
+    def <- defData(
+      varname = "size", formula = 103, variance = dispersion,
+      dist = "clusterSize"
+    )
+    dt <- genData(10, def)
+
+    expect_length(dt$size, 10)
+    expect_true(all(dt$size >= 0))
+    expect_equal(dt$size, floor(dt$size))
+    expect_equal(sum(dt$size), 103)
+    if (dispersion == 0) {
+      expect_lte(diff(range(dt$size)), 1)
+    }
+  }
+})
+
+test_that("addColumns preserves clusterSize totals and existing identifiers", {
+  clusters <- genData(10)
+
+  for (dispersion in c(0, 0.05)) {
+    def <- defDataAdd(
+      varname = "size", formula = 103, variance = dispersion,
+      dist = "clusterSize"
+    )
+    dt <- addColumns(def, clusters)
+
+    expect_identical(dt$id, clusters$id)
+    expect_length(dt$size, 10)
+    expect_true(all(dt$size >= 0))
+    expect_equal(dt$size, floor(dt$size))
+    expect_equal(sum(dt$size), 103)
+    if (dispersion == 0) {
+      expect_lte(diff(range(dt$size)), 1)
+    }
+  }
+})
+
 # .gencustom ----
 test_that("custom data is generated as expected.", {
   skip_on_cran()