Github user vanzin commented on a diff in the pull request: https://github.com/apache/spark/pull/2516#discussion_r18067446 --- Diff: core/src/main/scala/org/apache/spark/deploy/SparkSubmitArguments.scala --- @@ -406,22 +412,173 @@ private[spark] class SparkSubmitArguments(args: Seq[String]) { } } -object SparkSubmitArguments { - /** Load properties present in the given file. */ - def getPropertiesFromFile(file: File): Seq[(String, String)] = { - require(file.exists(), s"Properties file $file does not exist") - require(file.isFile(), s"Properties file $file is not a normal file") - val inputStream = new FileInputStream(file) +private[spark] object SparkSubmitArguments { + /** + * Resolves Configuration sources in order of highest to lowest + * 1. Each map passed in as additionalConfig from first to last + * 2. Environment variables (including legacy variable mappings) + * 3. System config variables (eg by using -Dspark.var.name) + * 4 SPARK_DEFAULT_CONF/spark-defaults.conf or SPARK_HOME/conf/spark-defaults.conf + * 5. hard coded defaults in class path at spark-submit-defaults.prop + * + * A property file specified by one of the means listed above gets read in and the properties are + * considered to be at the priority of the method that specified the files. + * A property specified in a property file will not override an existing + * config value at that same level + * + * @param additionalConfigs Seq of additional Map[ConfigName->ConfigValue] in order of highest + * priority to lowest this will have priority over internal sources + * @return Map[propName->propFile] containing values merged from all sources in order of priority + */ + def mergeSparkProperties(additionalConfigs: Seq [Map[String,String]]) = { + // Configuration read in from spark-submit-defaults.prop file found on the classpath + var hardCodedDefaultConfig: Option[Map[String,String]] = None + var is: InputStream = null + var isr: Option[InputStreamReader] = None try { - val properties = new Properties() - properties.load(inputStream) - properties.stringPropertyNames().toSeq.map(k => (k, properties(k).trim)) - } catch { - case e: IOException => - val message = s"Failed when loading Spark properties file $file" - throw new SparkException(message, e) + is = Thread.currentThread().getContextClassLoader.getResourceAsStream(ClassPathSparkSubmitDefaults) + + // only open InputStreamReader if InputStream was successfully opened + isr = Option(is).map{is: InputStream => + new InputStreamReader(is, CharEncoding.UTF_8) + } + + hardCodedDefaultConfig = isr.map( defaultValueStream => + SparkSubmitArguments.getPropertyValuesFromStream(defaultValueStream)) } finally { - inputStream.close() + Option(is).foreach(_.close) + isr.foreach(_.close) } + + if (hardCodedDefaultConfig.isEmpty || (hardCodedDefaultConfig.get.size == 0)) { + throw new IllegalStateException(s"Default values not found at classpath $ClassPathSparkSubmitDefaults") + } + + // Configuration read in from defaults file if it exists + var sparkDefaultConfig = SparkSubmitArguments.getSparkDefaultFileConfig --- End diff -- `sys.props`?
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org