Merge pull request #133 from ropensci/marianne-let-bargarp

Let bargap adapt and fix x-labels in category histograms

Author: mkcor

.gitignore

@@ -2,4 +2,6 @@ R/.Rhistory
 Rapp.history
 /Karthik_local.R
 *~
-.Rhistory
+.Rhistory
+.RData
+Makefile

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: plotly
 Type: Package
 Title: Interactive, publication-quality graphs online.
-Version: 0.5.4
+Version: 0.5.5
 Authors@R: c(person("Chris", "Parmer", role = c("aut", "cre"),
     email = "[email protected]"),
     person("Scott", "Chamberlain", role = "aut",

NEWS

@@ -1,3 +1,7 @@
+0.5.5 -- 24 October 2014.
+
+Support category histograms (with factors).
+
 0.5.4 -- 22 October 2014.
 
 Support conversion of geom_vline().

R/trace_generation.R

@@ -21,7 +21,6 @@ layer2traces <- function(l, d, misc) {
   barmode <- "group"
   if (g$geom == "bar" || g$geom == "histogram") {
     if (l$stat$objname == "bin") {
-      bargap <- 0
       if (g$geom != "histogram") {
         warning("You may want to use geom_histogram.")
       }
@@ -67,13 +66,15 @@ layer2traces <- function(l, d, misc) {
                                 "%Y-%m-%d %H:%M:%S")
         } else if (inherits(data.vec, "factor")) {
           ## Re-order data so that Plotly gets it right from ggplot2.
-          g$data <- g$data[order(g$data[[a]]),]
+          g$data <- g$data[order(g$data[[a]]), ]
           data.vec <- data.vec[match(g$data[[a]], as.numeric(data.vec))]
-          g$prestats.data <- g$prestats.data[order(g$prestats.data[[a]]),]
+          g$prestats.data <- g$prestats.data[order(g$prestats.data[[a]]), ]
           pdata.vec <- pdata.vec[match(g$prestats.data[[a]],
                                        as.numeric(pdata.vec))]
           if (length(pdata.vec) == length(data.vec))
             pdata.vec <- data.vec
+          if (!is.factor(pdata.vec))
+            pdata.vec <- g$prestats.data[[paste0(a, ".name")]]
         }
         g$data[[a]] <- data.vec
         g$prestats.data[[a]] <- pdata.vec

tests/testthat/test-ggplot-histogram.R

@@ -8,6 +8,9 @@ test_that("default position is translated to barmode=stack", {
   L <- gg2list(hist)
   expect_equal(length(L), 2)
   expect_identical(L$kwargs$layout$barmode, "stack")
+  expect_identical(L$kwargs$layout$xaxis$type, "category")
+  expect_identical(L[[1]]$type, "histogram")
+  expect_true(L[[1]]$x[1] %in% c("CDN", "MEX", "USA"))
 
   save_outputs(hist, "histogram-barmodestack")
 })