Separate h2o.jar from h2o R package except for developer build number…

… 99999. For official releases, the user must obtain h2o.jar separately from our S3 site and place it in the correct R subdirectory listed in the h2o.init.Rd docs. The method h2o.init will attempt to do this automatically if it cannot find the file when the user tries to initialize H2O locally.

Makefile

@@ -110,7 +110,7 @@ build:
 	@echo
 	@echo "PHASE: Building R package..."
 	@echo
-	$(MAKE) -C R PROJECT_VERSION=$(PROJECT_VERSION) 1> target/logs/r_build.log
+	$(MAKE) -C R PROJECT_VERSION=$(PROJECT_VERSION) BUILD_NUMBER=$(BUILD_NUMBER) 1> target/logs/r_build.log
 
 	@echo
 	@echo "PHASE: Building zip package..."

R/Makefile

@@ -1,5 +1,6 @@
 
 PROJECT_VERSION ?= 99.70
+BUILD_NUMBER ?= 99999
 BUILD_BRANCH = $(shell git branch | grep '*' | sed 's/* //')
 R_VERSION = 3.1
 REPO_PATH_WIN = bin/windows/contrib/$(R_VERSION)
@@ -46,15 +47,19 @@ else
 	cp -p h2o_package.pdf ../target/R
 endif
 	rm -rf h2o-package/inst/java
-	mkdir -p h2o-package/inst/java/tmp
-	cp -f ../target/h2o.jar h2o-package/inst/java/tmp/h2o_full.jar
-	(cd h2o-package/inst/java/tmp && jar xf h2o_full.jar)
-	(cd h2o-package/inst/java/tmp && rm -fr hadoop/0.* hadoop/1.* hadoop/cdh[35]* hadoop/cdh4_yarn)
-	(cd h2o-package/inst/java/tmp && rm -f h2o_full.jar)
-	(cd h2o-package/inst/java/tmp && cp META-INF/MANIFEST.MF ..)
-	(cd h2o-package/inst/java/tmp && rm -fr META-INF)
-	(cd h2o-package/inst/java/tmp && jar cfm ../h2o.jar ../MANIFEST.MF *)
-	rm -rf h2o-package/inst/java/tmp
+	# mkdir -p h2o-package/inst/java/tmp
+	# cp -f ../target/h2o.jar h2o-package/inst/java/tmp/h2o_full.jar
+	# (cd h2o-package/inst/java/tmp && jar xf h2o_full.jar)
+	# (cd h2o-package/inst/java/tmp && rm -fr hadoop/0.* hadoop/1.* hadoop/cdh[35]* hadoop/cdh4_yarn)
+	# (cd h2o-package/inst/java/tmp && rm -f h2o_full.jar)
+	# (cd h2o-package/inst/java/tmp && cp META-INF/MANIFEST.MF ..)
+	# (cd h2o-package/inst/java/tmp && rm -fr META-INF)
+	# (cd h2o-package/inst/java/tmp && jar cfm ../h2o.jar ../MANIFEST.MF *)
+	# rm -rf h2o-package/inst/java/tmp
+	mkdir -p h2o-package/inst/java
+ifeq ($(BUILD_NUMBER),99999)
+	cp -f ../target/h2o.jar h2o-package/inst/java
+endif
 	R CMD build h2o-package
 
 	# Create file with description of all packages

R/h2o-package/R/Wrapper.R

@@ -2,7 +2,7 @@
 # 1) If can't connect and user doesn't want to start H2O, stop immediately
 # 2) If user does want to start H2O and running locally, attempt to bring up H2O launcher
 # 3) If user does want to start H2O, but running non-locally, print an error
-h2o.init <- function(ip = "127.0.0.1", port = 54321, startH2O = TRUE, Xmx = "1g", beta = FALSE) {
+h2o.init <- function(ip = "127.0.0.1", port = 54321, startH2O = TRUE, forceDL = FALSE, Xmx = "1g", beta = FALSE) {
   if(!is.character(ip)) stop("ip must be of class character")
   if(!is.numeric(port)) stop("port must be of class numeric")
   if(!is.logical(startH2O)) stop("startH2O must be of class logical")
@@ -16,7 +16,7 @@ h2o.init <- function(ip = "127.0.0.1", port = 54321, startH2O = TRUE, Xmx = "1g"
       stop(paste("Cannot connect to H2O server. Please check that H2O is running at", myURL))
     else if(ip == "localhost" || ip == "127.0.0.1") {
       cat("\nH2O is not running yet, starting it now...\n")
-      .h2o.startJar(Xmx, beta)
+      .h2o.startJar(Xmx, beta, forceDL)
       count = 0; while(!url.exists(myURL) && count < 60) { Sys.sleep(1); count = count + 1 }
       if(!url.exists(myURL)) stop("H2O failed to start, stopping execution.")
     } else stop("Can only start H2O launcher if IP address is localhost.")
@@ -158,7 +158,7 @@ h2o.clusterStatus <- function(client) {
 #     h2o.shutdown(new("H2OClient", ip=ip, port=port), FALSE)
 # }
 
-.h2o.startJar <- function(memory = "1g", beta = FALSE) {
+.h2o.startJar <- function(memory = "1g", beta = FALSE, forceDL = FALSE) {
   command <- .h2o.checkJava()
 
   #
@@ -185,7 +185,8 @@ h2o.clusterStatus <- function(client) {
     stderr <- paste("/tmp/h2o", usr, "started_from_r.err", sep="_")
   }
 
-  jar_file <- paste(.h2o.pkg.path, "java", "h2o.jar", sep = .Platform$file.sep)
+  # jar_file <- paste(.h2o.pkg.path, "java", "h2o.jar", sep = .Platform$file.sep)
+  jar_file <- .h2o.downloadJar(overwrite = forceDL)
   jar_file <- paste('"', jar_file, '"', sep = "")
   args <- c(paste("-Xms", memory, sep=""),
             paste("-Xmx", memory, sep=""),
@@ -246,14 +247,26 @@ h2o.clusterStatus <- function(client) {
   stop("Cannot find Java. Please install the latest JDK from http://www.oracle.com/technetwork/java/javase/downloads/index.html")
 }
 
-.h2o.downloadJar <- function(branch = "rel-kahan", version) {
-  if(missing(version))
-    version <- packageVersion("h2o")[1,4]
-  dest_file <- paste(.h2o.pkg.path, "java", "h2o.jar", sep = .Platform$file.sep)
-  if(!file.exists(dest_file)) {
-    if(version == '99999') stop("Cannot find ", dest_file, "\nPlease check that Makefile copied the jar correctly.")
+.h2o.downloadJar <- function(branch, version, overwrite = FALSE) {
+  dest_folder <- paste(.h2o.pkg.path, "java", sep = .Platform$file.sep)
+  if(!file.exists(dest_folder)) dir.create(dest_folder)
+  dest_file <- paste(dest_folder, "h2o.jar", sep = .Platform$file.sep)
+
+  # Download if h2o.jar doesn't already exist or user specifies force overwrite
+  if(overwrite || !file.exists(dest_file)) {
+    if(missing(branch)) branch <- packageDescription("h2o")$Branch
+    if(missing(version)) version <- packageVersion("h2o")[1,4]
+
+    # Production version must exist on local drive
+    if(version == '99999' && !file.exists(dest_file))
+      stop("Cannot find ", dest_file, "\nPlease check that Makefile copied the jar correctly.")
+
     h2o_url <- paste("http://s3.amazonaws.com/h2o-release/h2o", branch, version, "h2o.jar", sep = "/")
     download.file(h2o_url, dest_file, mode = "wb")
+
+    # TODO: Check file integrity to ensure download okay
+    if(!file.exists(dest_file))
+      stop("Error: Transfer failed. Please download ", h2o_url, " and place h2o.jar in ", dest_folder)
   }
   return(dest_file)
 }

R/h2o-package/man/h2o.init.Rd

@@ -4,20 +4,26 @@
 \description{
 Connects to a running H2O instance and checks the local H2O R package is the correct version (i.e. that the version of the R package and the version of H2O are the same).}
 \usage{
-h2o.init(ip = "127.0.0.1", port = 54321, startH2O = TRUE, Xmx = "1g", beta = FALSE)}
+h2o.init(ip = "127.0.0.1", port = 54321, startH2O = TRUE, forceDL = FALSE, Xmx = "1g", 
+        beta = FALSE)}
 
 \arguments{
   \item{ip}{Object of class \code{"character"} representing the IP address of the server where H2O is running.}
   \item{port}{Object of class \code{"numeric"} representing the port number of the H2O server.}
-  \item{startH2O}{ (Optional) A logical value indicating whether to start the H2O launcher GUI if no connection with H2O is detected. This is only possible if H2O was installed from the InstallBuilder executable and \code{ip = "localhost"} or \code{ip = "127.0.0.1"}.}
+  \item{startH2O}{(Optional) A logical value indicating whether to start the H2O launcher GUI if no connection with H2O is detected. This is only possible if \code{ip = "localhost"} or \code{ip = "127.0.0.1"}.}
+  \item{forceDL}{(Optional) A logical value indicating whether to force download of the H2O executable. Defaults to FALSE, so the executable will only be downloaded if it does not already exist in the h2o R library resources directory \code{h2o/java/h2o.jar}.}
   \item{Xmx}{(Optional) A string specifying the maximum size, in bytes, of the memory allocation pool to H2O. This value must a multiple of 1024 greater than 2MB. Append the letter m or M to indicate megabytes, or g or G to indicate gigabytes.}
   \item{beta}{(Optional) A logical value indicating whether H2O should be launch in beta mode.}
 }
 \details{
-This method first checks if H2O is connectable. If it cannot connect and \code{startH2O = TRUE} with IP of localhost, it will attempt to start an instance of H2O with IP = localhost, port = 54321. Otherwise, it stops immediately with an error. Once connected, the method checks to see if the local H2O R package version matches the version of H2O running on the server. If there is a mismatch and the user indicates she wishes to upgrade, it will remove the local H2O R package and download/install the H2O R package from the server.
+This method first checks if H2O is connectable. If it cannot connect and \code{startH2O = TRUE} with IP of localhost, it will attempt to start an instance of H2O with IP = localhost, port = 54321. Otherwise, it stops immediately with an error.
+
+When initializing H2O locally, this method searches for h2o.jar in the R library resources (\code{system.file("java", "h2o.jar", package = "h2o")}), and if the file does not exist, it will automatically attempt to download the correct version from Amazon S3. The user must have Internet access for this process to be successful.
+
+Once connected, the method checks to see if the local H2O R package version matches the version of H2O running on the server. If there is a mismatch and the user indicates she wishes to upgrade, it will remove the local H2O R package and download/install the H2O R package from the server.
 }
 \value{
-Once the package is successfully installed, this method will load it and return a \code{H2OClient} object containing the IP address and port number of the H2O server. See the \href{http://docs.0xdata.com/userguide/topR.html}{H2O R package documentation} for more details, or type \code{??h2o}.
+Once the package is successfully installed, this method will load it and return a \code{H2OClient} object containing the IP address and port number of the H2O server. See the \href{http://docs.0xdata.com/userguide/topR.html}{H2O R package documentation} for more details, or type \code{??h2o} in the R console.
 }
 \note{ Users may wish to manually upgrade their package (rather than waiting 
 	until being prompted), which requires that they fully uninstall and reinstall