Merge pull request #76 from r-spark/bugfix/rayrender-cran

Install rayrender from CRAN

.travis.yml

@@ -46,6 +46,7 @@ matrix:
         - ggraph
         - graphframes
         - RCurl
+        - rayrender
         - rsample
         - sparktf
         - sparkxgb

DESCRIPTION

@@ -20,5 +20,4 @@ Suggests:
   testthat,
   geospark
 Remotes:
-  r-spark/geospark,
   rstudio/mleap

distributed-r.Rmd

@@ -392,7 +392,7 @@ Next you will learn a use case for large-compute where R is used to perform dist
 
 R can be used in combination with Spark to perform large-scale computing. The use case we will explore in this section is rendering computationally-expensive images using the `rayrender` package, which uses ray tracing -- a photorealistic technique commonly used in movie production.
 
-Let's use this package to render a simple scene that includes a few spheres with using a lambertian material, a diffusely reflecting material or "matte"; but first, install using `remotes::install_github("tylermorganwall/rayrender")`, make sure to also disconnect and reconnect from Spark.
+Let's use this package to render a simple scene that includes a few spheres with using a lambertian material, a diffusely reflecting material or "matte"; but first, install using `install.packages("rayrender")`, make sure to also disconnect and reconnect from Spark.
 
 ```{r distributed-sim-render, eval=FALSE, echo=FALSE}
 library(rayrender)
@@ -404,7 +404,7 @@ scene <- generate_ground(material = lambertian()) %>%
 
 render_scene(scene, width = 1920, height = 1080, lookfrom = c(10, 5, 0), parallel = TRUE, filename = "images/distributed-r-rayrender.png")
 ```
-```{r distributed-sim-rayrender, eval=FALSE, exercise=TRUE}
+```{r distributed-sim-rayrender}
 library(rayrender)
 
 scene <- generate_ground(material = lambertian()) %>%