Remove sparkr and related loading code, partially re-open SparkInterp…

…reterLauncherTest
StrongestNumber9 · Feb 10, 2025 · 3e1d946 · 3e1d946
1 parent 055c870
commit 3e1d946
Show file tree

Hide file tree

Showing 7 changed files with 7 additions and 217 deletions.
diff --git a/spark/interpreter/src/main/java/org/apache/zeppelin/spark/ZeppelinRContext.java b/spark/interpreter/src/main/java/org/apache/zeppelin/spark/ZeppelinRContext.java
diff --git a/spark/interpreter/src/test/resources/spark_server.R b/spark/interpreter/src/test/resources/spark_server.R
diff --git a/spark/interpreter/src/test/resources/spark_ui.R b/spark/interpreter/src/test/resources/spark_ui.R
diff --git a/...er-integration/src/test/java/org/apache/zeppelin/integration/ZSessionIntegrationTest.java b/...er-integration/src/test/java/org/apache/zeppelin/integration/ZSessionIntegrationTest.java
@@ -193,14 +193,6 @@ public void testZSession_Spark() throws Exception {
               "+---+---+", result.getResults().get(0).getData().trim());
       assertTrue(result.getJobUrls().size() > 0);
 
-      // sparkr
-      result = session.execute("r", "df <- as.DataFrame(faithful)\nhead(df)");
-      assertEquals(Status.FINISHED, result.getStatus());
-      assertEquals(1, result.getResults().size());
-      assertEquals("TEXT", result.getResults().get(0).getType());
-      assertTrue(result.getResults().get(0).getData(), result.getResults().get(0).getData().contains("eruptions waiting"));
-      assertEquals(2, result.getJobUrls().size());
-
       // spark sql
       result = session.execute("sql", "select * from df");
       assertEquals(Status.FINISHED, result.getStatus());
@@ -264,15 +256,6 @@ public void testZSession_Spark_Submit() throws Exception {
               "+---+---+", result.getResults().get(0).getData().trim());
       assertTrue(result.getJobUrls().size() > 0);
 
-      // sparkr
-      result = session.submit("r", "df <- as.DataFrame(faithful)\nhead(df)");
-      result = session.waitUntilFinished(result.getStatementId());
-      assertEquals(Status.FINISHED, result.getStatus());
-      assertEquals(1, result.getResults().size());
-      assertEquals("TEXT", result.getResults().get(0).getType());
-      assertTrue(result.getResults().get(0).getData(), result.getResults().get(0).getData().contains("eruptions waiting"));
-      assertEquals(2, result.getJobUrls().size());
-
       // spark sql
       result = session.submit("sql", "select * from df");
       result = session.waitUntilFinished(result.getStatementId());

diff --git a/...r-integration/src/test/java/org/apache/zeppelin/integration/ZeppelinSparkClusterTest.java b/...r-integration/src/test/java/org/apache/zeppelin/integration/ZeppelinSparkClusterTest.java
@@ -295,22 +295,6 @@ public void sparkSQLTest() throws IOException {
     assertEquals("_1\t_2\nhello\t20\n", p.getReturn().message().get(0).getData());
   }
 
-  @Ignore(value="Tests Spark R which we do not support")
-  @Test
-  public void sparkRTest() throws IOException {
-    Note note = TestUtils.getInstance(Notebook.class).createNote("note1", anonymous);
-    Paragraph p = note.addNewParagraph(anonymous);
-
-    p.setText("%spark.r localDF <- data.frame(name=c(\"a\", \"b\", \"c\"), age=c(19, 23, 18))\n" +
-            "df <- createDataFrame(localDF)\n" +
-            "count(df)"
-    );
-
-    note.run(p.getId(), true);
-    assertEquals(Status.FINISHED, p.getStatus());
-    assertEquals("[1] 3", p.getReturn().message().get(0).getData().trim());
-  }
-
   @Test
   public void pySparkTest() throws IOException {
     // create new note

diff --git a/...gine/src/main/java/org/apache/zeppelin/interpreter/launcher/SparkInterpreterLauncher.java b/...gine/src/main/java/org/apache/zeppelin/interpreter/launcher/SparkInterpreterLauncher.java
@@ -89,7 +89,6 @@ public Map<String, String> buildEnvFromProperties(InterpreterLaunchContext conte
     }
 
     setupPropertiesForPySpark(sparkProperties);
-    setupPropertiesForSparkR(sparkProperties);
 
     String condaEnvName = context.getProperties().getProperty("zeppelin.interpreter.conda.env.name");
     if (StringUtils.isNotBlank(condaEnvName)) {
@@ -343,33 +342,6 @@ private void mergeSparkProperty(Properties sparkProperties, String propertyName,
     }
   }
 
-  private void setupPropertiesForSparkR(Properties sparkProperties) {
-    if (isYarnMode()) {
-      String sparkHome = getEnv("SPARK_HOME");
-      File sparkRBasePath = null;
-      if (sparkHome == null) {
-        if (!getSparkMaster().startsWith("local")) {
-          throw new RuntimeException("SPARK_HOME is not specified in interpreter-setting" +
-                  " for non-local mode, if you specify it in zeppelin-env.sh, please move that into " +
-                  " interpreter setting");
-        }
-        String zeppelinHome = zConf.getString(ZeppelinConfiguration.ConfVars.ZEPPELIN_HOME);
-        sparkRBasePath = new File(zeppelinHome,
-                "interpreter" + File.separator + "spark" + File.separator + "R");
-      } else {
-        sparkRBasePath = new File(sparkHome, "R" + File.separator + "lib");
-      }
-
-      File sparkRPath = new File(sparkRBasePath, "sparkr.zip");
-      if (sparkRPath.exists() && sparkRPath.isFile()) {
-        mergeSparkProperty(sparkProperties, "spark.yarn.dist.archives",
-                sparkRPath.getAbsolutePath() + "#sparkr");
-      } else {
-        LOGGER.warn("sparkr.zip is not found, SparkR may not work.");
-      }
-    }
-  }
-
   /**
    * Returns cached Spark Master value if it's present, or calculate it
    *