Merge branch 'dev'

* dev:
  Bump version number to 0.2.12
  Fail immediately in consume() if error code is 400. #57
  Remove importFrom(utils, data)
  Remove vignette artefacts
  Rebuild vignettes
  Implement workarounds to pass R CMD check in wrapper.R
  Suppress warnings in download.datasets() to deal with some csv samples containing missing final line, i.e. missing line break. #115
  Rebuild documentation using roxygen2 v5
  Document how to use different regions. #105
  Rename arguments from tries to .retry #116
  Fixed typo in URL. #105
  Update date
  Refactor, and catch error with NA values. #115

DESCRIPTION

@@ -4,8 +4,8 @@ Title: Interface with Azure Machine Learning Datasets, Experiments and Web Servi
 Description: Functions and datasets to support Azure Machine Learning. This
     allows you to interact with datasets, as well as publish and consume R functions
     as API services.
-Version: 0.2.11
-Date: 2016-07-01
+Version: 0.2.12
+Date: 2016-07-13
 Authors@R: c(
       person("Andrie", "de Vries", role=c("aut", "cre"), email="[email protected]"),
       person(family="Microsoft Corporation", role="cph"),

NAMESPACE

@@ -1,4 +1,4 @@
-# Generated by roxygen2 (4.1.1): do not edit by hand
+# Generated by roxygen2: do not edit by hand
 
 S3method(print,Datasets)
 S3method(print,Experiments)

R/consume.R

@@ -29,6 +29,7 @@
 #' @export
 #'
 #' @inheritParams refresh
+#' @inheritParams publishWebService
 #' @param endpoint Either an AzureML web service endpoint returned by \code{\link{publishWebService}}, \code{\link{endpoints}}, or simply an AzureML web service from \code{\link{services}}; in the latter case the default endpoint for the service will be used.
 #' @param ... variable number of requests entered as lists in key-value format; optionally a single data frame argument.
 #' @param globalParam global parameters entered as a list, default value is an empty list
@@ -43,7 +44,7 @@
 #' @family consumption functions
 #' @importFrom jsonlite fromJSON
 #' @example inst/examples/example_publish.R
-consume <- function(endpoint, ..., globalParam, retryDelay = 10, output = "output1", tries = 5)
+consume <- function(endpoint, ..., globalParam, retryDelay = 10, output = "output1", .retry = 5)
 {
   if(is.Service(endpoint))
   {
@@ -73,7 +74,7 @@ consume <- function(endpoint, ..., globalParam, retryDelay = 10, output = "outpu
     }
   }
   # Make API call with parameters
-  result <- callAPI(apiKey, requestUrl, requestsLists,  globalParam, retryDelay, tries = tries)
+  result <- callAPI(apiKey, requestUrl, requestsLists,  globalParam, retryDelay, .retry = .retry)
   if(inherits(result, "error")) stop("AzureML returned an error code")
 
   # Access output by converting from JSON into list and indexing into Results
@@ -104,14 +105,14 @@ consume <- function(endpoint, ..., globalParam, retryDelay = 10, output = "outpu
 # @param keyvalues the data to be passed to the web service
 # @param globalParam the global parameters for the web service
 # @param retryDelay number of seconds to wait after failing (max 3 tries) to try again
-# @param tries the number of retry attempts
+# @param .retry the number of retry attempts
 # @return result the response
 #
 # @importFrom jsonlite toJSON
 # @importFrom curl handle_setheaders new_handle handle_setopt curl_fetch_memory
 # @keywords internal
 callAPI <- function(apiKey, requestUrl, keyvalues, globalParam, 
-                    retryDelay=10, tries = 5) {
+                    retryDelay=10, .retry = 5) {
   # Set number of tries and HTTP status to 0
   result <- NULL
   # Construct request payload
@@ -136,7 +137,7 @@ callAPI <- function(apiKey, requestUrl, keyvalues, globalParam,
                   postfields = body
                 )
   )
-  r <- try_fetch(requestUrl, h, delay = retryDelay, tries = tries)
+  r <- try_fetch(requestUrl, h, no_retry_on = 400, delay = retryDelay, .retry = .retry)
   result <- fromJSON(rawToChar(r$content))
   if(r$status_code >= 400)  {
     stop(paste(capture.output(result), collapse="\n"))

R/datasets.R

@@ -268,7 +268,7 @@ delete.datasets <- function(ws, name, host){
                    curl_escape(ws$id),
                    curl_escape(familyId)
     )
-    z <- try_fetch(uri, h, tries = 3, delay = 2)
+    z <- try_fetch(uri, h, .retry = 3, delay = 2)
     z$status_code
   }
   status_code <- vapply(datasets$FamilyId, delete_one, FUN.VALUE = numeric(1), USE.NAMES = FALSE)

R/discover.R

@@ -86,43 +86,15 @@ discoverSchema <- function(helpURL, scheme = "https",
 
   # Accesses the names of the columns in the example
   # and stores it in a list of column names
-#   columnNames <- vector("list", length = length(inputExample))
-#   columnNames <- list()
-#   for(i in seq_along(inputExample)) {
-#     columnNames[[i]] = names(inputExample)[i]
-#   }
   columnNames <- lapply(seq_along(inputExample), function(i)names(inputExample[i]))
 
-    # Uses multiple nested loops to access the various paths in the 
-  # swagger document and find the execution path
-  foundExecPath = FALSE
-  pathNo = 0
-  execPathNo = -1
-  for(execPath in swagger$paths) {
-    pathNo = pathNo + 1
-    for(operationpath in execPath) {
-      for(operation in operationpath) {
-        # Goes through the characteristcs in every operation e.g. operationId
-        for(charac in operation) {
-          # Finds the path in which the 
-          # operationId (characteristic of the path) == execute 
-          # and sets the execution path number
-          if(charac[1] == "execute")
-          {
-            #Sets found execution path to true
-            foundExecPath = TRUE
-            execPathNo = pathNo
-            break
-          }
-        }
-      }
-    }
+  execPathNo <- grep("/execute\\?", names(swagger$paths))
+  if(is.numeric(execPathNo)) {
+    executePath <- names(swagger$paths)[[execPathNo]]
+  } else {
+    "Path not found"
   }
 
-  # Stores the execution path
-  executePath <- if(foundExecPath) names(swagger$paths)[[execPathNo]] 
-  else "Path not found"
-
   # Constructs the request URL with the parameters as well as execution path found. 
   # The separator is set to an empty string
   requestUrl <- paste0(scheme,"://", host, 

R/fetch.R

@@ -67,7 +67,7 @@ validate_response <- function(r){
 # @param uri the uri to fetch
 # @param handle a curl handle
 # @param retry_on HTTP status codes that result in retry
-# @param tries number of tries before failing
+# @param .retry number of tries before failing
 # @param delay in seconds between retries, subject to exponent
 # @param exponent increment each successive delay by delay^exponent
 # @param no_message_threshold Only show messages if delay is greater than this limit
@@ -76,20 +76,26 @@ validate_response <- function(r){
 # @return the result of curl_fetch_memory(uri, handle)
 # 
 try_fetch <- function(uri, handle, 
-                      retry_on = c(400, 401, 440, 503, 504, 509), 
-                      tries = 6, 
+                      retry_on = c(400, 401, 440, 503, 504, 509),
+                      no_retry_on,
+                      .retry = 6, 
                       delay = 1, exponent = 2,
                       no_message_threshold = 1)
 {
   r = curl_fetch_memory(uri, handle)
-  # if(r$status_code == 400) validate_response(r)
+  # if(r$status_code == 400){
+  #   validate_response(r)
+  # }
+  if(!missing(no_retry_on) && !is.null(no_retry_on)){
+    retry_on <- setdiff(retry_on, no_retry_on)
+  }
   if(!(r$status_code %in% retry_on)) {
     validate_response(r)
     return(r)
   }
   collisions = 1
   printed_message <- FALSE
-  while(collisions < (tries)) {
+  while(collisions < (.retry)) {
     r = curl_fetch_memory(uri, handle)
     if(!(r$status_code %in% retry_on)) {
       validate_response(r)

R/internal.R

@@ -39,7 +39,7 @@ get_datasets <- function(ws) {
   h = new_handle()
   handle_setheaders(h, .list = ws$.headers)
   uri <- sprintf("%s/workspaces/%s/datasources", ws$.studioapi, ws$id)
-  r <- try_fetch(uri = uri, handle = h, delay = 0.25, tries = 3)
+  r <- try_fetch(uri = uri, handle = h, delay = 0.25, .retry = 3)
 
   msg <- paste("No results returned from datasets(ws).", 
                "Please check your workspace credentials and api_endpoint are correct.")
@@ -89,7 +89,7 @@ get_experiments <- function(ws) {
   h = new_handle()
   handle_setheaders(h, .list=ws$.headers)
   uri = sprintf("%s/workspaces/%s/experiments", ws$.studioapi, ws$id)
-  r <- try_fetch(uri = uri, handle = h, delay = 0.25, tries = 3)
+  r <- try_fetch(uri = uri, handle = h, delay = 0.25, .retry = 3)
 
   msg <- paste("No results returned from experiments(ws).", 
                "Please check your workspace credentials and api_endpoint are correct.")

R/publish.R

@@ -29,6 +29,7 @@
 #' @export
 #'
 #' @inheritParams refresh
+#' @inheritParams workspace
 #' @param fun a function to publish; the function must have at least one argument.
 #' @param name name of the new web service; ignored when \code{serviceId} is specified (when updating an existing web service).
 #' 
@@ -43,6 +44,7 @@
 #' @param serviceId optional Azure web service ID; use to update an existing service (see Note below).
 #' @param host optional Azure regional host, defaulting to the global \code{management_endpoint} set in \code{\link{workspace}}
 #' @param data.frame \code{TRUE} indicates that the function \code{fun} accepts a data frame as input and returns a data frame output; automatically set to \code{TRUE} when \code{inputSchema} is a data frame.
+#' @param .retry number of tries before failing
 #' 
 #' @return A data.frame describing the new service endpoints, cf. \code{\link{endpoints}}. The output can be directly used by the \code{\link{consume}} function.
 #'  
@@ -83,11 +85,12 @@
 #' @importFrom uuid UUIDgenerate
 #' @importFrom curl new_handle handle_setheaders handle_setopt
 publishWebService <- function(ws, fun, name,
-                              inputSchema, outputSchema, `data.frame`=FALSE,
-                              export=character(0), 
-                              noexport=character(0), 
+                              inputSchema, outputSchema, 
+                              `data.frame` = FALSE,
+                              export = character(0), 
+                              noexport = character(0), 
                               packages,
-                              version="3.1.0", 
+                              version = "3.1.0", 
                               serviceId, 
                               host = ws$.management_endpoint,
                               .retry = 3)
@@ -130,18 +133,7 @@ publishWebService <- function(ws, fun, name,
     }
 
   } 
-  # inputSchema <- azureSchema(inputSchema)
-  # outputSchema <- azureSchema(outputSchema)
-  # if(`data.frame`) {
-  #   if(length(formals(fun)) != 1) {
-  #     stop("when data.frame=TRUE fun must only take one data.frame argument")
-  #   }
-  # } else {
-  # if(length(formals(fun)) != length(inputSchema)) {
-  #   stop("length(inputSchema) does not match the number of function arguments")
-  # }
-  # }
-
+
   ### Get and encode the dependencies
 
   if(missing(packages)) packages=NULL
@@ -191,7 +183,7 @@ publishWebService <- function(ws, fun, name,
   )
   handle_setheaders(h, .list = httpheader)
   handle_setopt(h, .list = opts)
-  r = try_fetch(publishURL, handle = h, tries = .retry)
+  r = try_fetch(publishURL, handle = h, .retry = .retry)
   result = rawToChar(r$content)
   if(r$status_code >= 400) stop(result)
   newService = fromJSON(result)

R/services.R

@@ -75,7 +75,7 @@ services <- function(ws, service_id, name, host = ws$.management_endpoint)
   }
 
   uri <- sprintf("%s/workspaces/%s/webservices%s", host, ws$id, service_id)
-  r <- try_fetch(uri = uri, handle = h, delay = 0.25, tries = 3)
+  r <- try_fetch(uri = uri, handle = h, delay = 0.25, .retry = 3)
   #   if(inherits(r, "error")){
   #     msg <- paste("No results returned from datasets(ws).", 
   #                  "Please check your workspace credentials and api_endpoint are correct.")

R/workspace.R

@@ -73,22 +73,51 @@ default_api <- function(api_endpoint = "https://studioapi.azureml.net"){
 #' \if{html}{\figure{authorization_token.png}{options: width="60\%" alt="Figure: authorization_token.png"}}
 #' \if{latex}{\figure{authorizationToken.pdf}{options: width=7cm}}
 #'
-#' @section Using a \code{settings.json} file:
+#'
+#' @section Using a settings.json file:
 #'  If any of the \code{id}, \code{auth}, \code{api_endpoint} or \code{management_endpoint} arguments are missing, the function attempts to read values from the \code{config} file with JSON format:
 #'  \preformatted{
 #'  {"workspace":{
 #'    "id": "enter your AzureML workspace id here",
 #'    "authorization_token": "enter your AzureML authorization token here",
 #'    "api_endpoint": "https://studioapi.azureml.net",
+#'  }}
+#' }
+#' 
+#' To explicitly add the management endpoint in the JSON file, use:
+#'  \preformatted{
+#'  {"workspace":{
+#'    "id": "enter your AzureML workspace id here",
+#'    "authorization_token": "enter your AzureML authorization token here",
+#'    "api_endpoint": "https://studioapi.azureml.net",
 #'    "management_endpoint": "https://management.azureml.net"
 #'  }}
 #' }
+#'    
+#' @section Using a workspace in different Azure Machine Learning regions:
+#' 
+#' By default, the Azure Machine Learning workspace is located in US South Central, but it is possible to create a workspace in different regions, including Europe West and Asia Southeast.
+#' 
+#' To use a workspace in Asia Southeast, you can modify the api endpoint line in the JSON file:
+#'  \preformatted{
+#'   {"workspace": {
+#'     "api_endpoint": ["https://asiasoutheast.studio.azureml.net"]
+#'   }}
+#' }
+#'
+#' Similarly, for a workspace in Europe West:
+#'  \preformatted{
+#'   {"workspace": {
+#'     "api_endpoint": ["https://europewest.studio.azureml.net"]
+#'   }}
+#' }
+#' 
 
 #' 
 #' 
 #' @param id Optional workspace id from ML studio -> settings -> WORKSPACE ID. See the section "Finding your AzureML credentials" for more details.
 #' @param auth Optional authorization token from ML studio -> settings -> AUTHORIZATION TOKENS. See the section "Finding your AzureML credentials" for more details.
-#' @param api_endpoint Optional AzureML API web service URI. Defaults to \code{https://studioap.azureml.net} if not provided and not specified in config.  See note.
+#' @param api_endpoint Optional AzureML API web service URI. Defaults to \code{https://studioapi.azureml.net} if not provided and not specified in config.  See note.
 #' @param management_endpoint Optional AzureML management web service URI. Defaults to \code{https://management.azureml.net} if not provided and not specified in config.  See note.
 #' @param config Optional settings file containing id and authorization info. Used if any of the other arguments are missing. The default config file is \code{~/.azureml/settings.json}, but you can change this location by setting \code{options(AzureML.config = "newlocation")}. See the section "Using a settings.json file" for more details.
 #' @param ... ignored