rbind.fill 1d arrays to vectors

crowding · Feb 21, 2014 · 37e60ae · 37e60ae
1 parent 68617ea
commit 37e60ae
Show file tree

Hide file tree

Showing 2 changed files with 25 additions and 21 deletions.
diff --git a/R/rbind-fill.r b/R/rbind-fill.r
@@ -127,28 +127,27 @@ allocate_column <- function(example, nrows, dfs, var) {
   }
 
   if (is.array(example)) {
+    if (length(dim(example)) > 1) {
+      if ("dimnames" %in% names(a)) {
+        a$dimnames[1] <- list(NULL)
+        if (!is.null(names(a$dimnames)))
+            names(a$dimnames)[1] <- ""
+      }
+
+      # Check that all other args have consistent dims
+      df_has <- vapply(dfs, function(df) var %in% names(df), FALSE)
+      dims <- unique(lapply(dfs[df_has], function(df) dim(df[[var]])[-1]))
+      if (length(dims) > 1)
+          stop("Array variable ", var, " has inconsistent dims")
 
-    if ("dimnames" %in% names(a)) {
-      a$dimnames[1] <- list(NULL)
-      if (!is.null(names(a$dimnames)))
-          names(a$dimnames)[1] <- ""
-    }
-
-    # Check that all other args have consistent dims
-    df_has <- vapply(dfs, function(df) var %in% names(df), FALSE)
-    dims <- unique(lapply(dfs[df_has], function(df) dim(df[[var]])[-1]))
-    if (length(dims) > 1)
-        stop("Array variable ", var, " has inconsistent dims")
-
-    if (length(dims[[1]]) == 0) { #is dropping dims necessary for 1d arrays?
+      a$dim <- c(nrows, dim(example)[-1])
+      length <- prod(a$dim)
+    } else {
+      #1d arrays devolve into vectors
       a$dim <- NULL
       a$dimnames <- NULL
       length <- nrows
-    } else {
-      a$dim <- c(nrows, dim(example)[-1])
-      length <- prod(a$dim)
     }
-
   } else {
     length <- nrows
   }

diff --git a/inst/tests/test-rbind.r b/inst/tests/test-rbind.r
@@ -101,22 +101,27 @@ test_that("time zones are preserved", {
 
 })
 
-test_that("arrays are ok", {
+test_that("1d arrays treated as vectors", {
   df <- data.frame(x = 1)
   df$x <- array(1, 1)
 
+  #1d arrays converted into vectors
   df2 <- rbind.fill(df, df)
-  #this asserts that dim is stripped off 1d arrays. Necessary?
   expect_that(df2$x, is_equivalent_to(rbind(df, df)$x))
   expect_that(dim(df2$x), equals(dim(rbind(df, df)$x)))
-  #this would be more consistent
-  #expect_that(df2$x, is_equivalent_to(rbind(array(1,1), array(1,1))))
 
   #if dims are stripped, dimnames should be also
   df <- data.frame(x = 1)
   df$x <- array(2, 1, list(x="one"))
   df2 <- rbind.fill(df, df)
   expect_that(is.null(dimnames(df2$x)), is_true())
+
+  #can bind 1d array to vector
+  dfV <- data.frame(x=3)
+  dfO1 <- rbind.fill(df, dfV)
+  dfO2 <- rbind.fill(dfV, df)
+  expect_equal(dfO1, data.frame(x=c(2, 3)))
+  expect_equal(dfO2, data.frame(x=c(3, 2)))
 })
 
 test_that("multidim arrays ok", {