ORC-1607: Fix testDoubleNaNAndInfinite to use `TestFileDump.checkOu…

…tput` ### What changes were proposed in this pull request? This PR aims to fix a unit test failure. ### Why are the changes needed? ``` [ERROR] Failures: [ERROR] TestJsonFileDump.testDoubleNaNAndInfinite:167 expected: <{"fileName":"TestFileDump.testDump.orc","fileVersion":"0.12","writerVersion":"ORC_14","softwareVersion":"ORC Java unknown","numberOfRows":3,"compression":"ZSTD","compressionBufferSize":262144,"schemaString":"struct<x:double>","schema":{"columnId":0,"columnType":"STRUCT","children":{"x":{"columnId":1,"columnType":"DOUBLE"}}},"calendar":"Julian/Gregorian","stripeStatistics":[{"stripeNumber":1,"columnStatistics":[{"columnId":0,"count":3,"hasNull":false},{"columnId":1,"count":3,"hasNull":false,"bytesOnDisk":27,"min":NaN,"max":NaN,"sum":NaN,"type":"DOUBLE"}]}],"fileStatistics":[{"columnId":0,"count":3,"hasNull":false},{"columnId":1,"count":3,"hasNull":false,"bytesOnDisk":27,"min":NaN,"max":NaN,"sum":NaN,"type":"DOUBLE"}],"stripes":[{"stripeNumber":1,"stripeInformation":{"offset":3,"indexLength":55,"dataLength":27,"footerLength":35,"rowCount":3},"streams":[{"columnId":0,"section":"ROW_INDEX","startOffset":3,"length":11},{"columnId":1,"section":"ROW_INDEX","startOffset":14,"length":44},{"columnId":1,"section":"DATA","startOffset":58,"length":27}],"encodings":[{"columnId":0,"kind":"DIRECT"},{"columnId":1,"kind":"DIRECT"}]}],"fileLength":286,"rawDataSize":36,"paddingLength":0,"paddingRatio":0.0,"status":"OK"}> but was: <{"fileName":"TestFileDump.testDump.orc","fileVersion":"0.12","writerVersion":"ORC_14","softwareVersion":"ORC Java 2.1.0-SNAPSHOT","numberOfRows":3,"compression":"ZSTD","compressionBufferSize":262144,"schemaString":"struct<x:double>","schema":{"columnId":0,"columnType":"STRUCT","children":{"x":{"columnId":1,"columnType":"DOUBLE"}}},"calendar":"Julian/Gregorian","stripeStatistics":[{"stripeNumber":1,"columnStatistics":[{"columnId":0,"count":3,"hasNull":false},{"columnId":1,"count":3,"hasNull":false,"bytesOnDisk":27,"min":NaN,"max":NaN,"sum":NaN,"type":"DOUBLE"}]}],"fileStatistics":[{"columnId":0,"count":3,"hasNull":false},{"columnId":1,"count":3,"hasNull":false,"bytesOnDisk":27,"min":NaN,"max":NaN,"sum":NaN,"type":"DOUBLE"}],"stripes":[{"stripeNumber":1,"stripeInformation":{"offset":3,"indexLength":55,"dataLength":27,"footerLength":35,"rowCount":3},"streams":[{"columnId":0,"section":"ROW_INDEX","startOffset":3,"length":11},{"columnId":1,"section":"ROW_INDEX","startOffset":14,"length":44},{"columnId":1,"section":"DATA","startOffset":58,"length":27}],"encodings":[{"columnId":0,"kind":"DIRECT"},{"columnId":1,"kind":"DIRECT"}]}],"fileLength":293,"rawDataSize":36,"paddingLength":0,"paddingRatio":0.0,"status":"OK"}> [INFO] [ERROR] Tests run: 51, Failures: 1, Errors: 0, Skipped: 0 ``` ### How was this patch tested? Manually check on Apple Silicon MacOS. ### Was this patch authored or co-authored using generative AI tooling? No. Closes #1781 from dongjoon-hyun/ORC-1607. Authored-by: Dongjoon Hyun <dongjoon@apache.org> Signed-off-by: Dongjoon Hyun <dongjoon@apache.org>
apache · Feb 3, 2024 · 0dee947 · 0dee947
1 parent 6c3c451
commit 0dee947
Show file tree

Hide file tree

Showing 2 changed files with 112 additions and 6 deletions.
diff --git a/java/tools/src/test/org/apache/orc/tools/TestJsonFileDump.java b/java/tools/src/test/org/apache/orc/tools/TestJsonFileDump.java
@@ -33,7 +33,6 @@
 import org.junit.jupiter.api.BeforeEach;
 import org.junit.jupiter.api.Test;
 
-import java.io.ByteArrayOutputStream;
 import java.io.File;
 import java.io.FileOutputStream;
 import java.io.PrintStream;
@@ -156,14 +155,14 @@ public void testDoubleNaNAndInfinite() throws Exception {
     assertEquals(3, writer.getNumberOfRows());
 
     PrintStream origOut = System.out;
-    ByteArrayOutputStream myOut = new ByteArrayOutputStream();
+    String outputFilename = "orc-file-dump-nan-and-infinite.json";
+    FileOutputStream myOut = new FileOutputStream(workDir + File.separator + outputFilename);
 
     // replace stdout and run command
-    System.setOut(new PrintStream(myOut, false, StandardCharsets.UTF_8));
-    FileDump.main(new String[]{testFilePath.toString(), "-j"});
+    System.setOut(new PrintStream(myOut, true, StandardCharsets.UTF_8));
+    FileDump.main(new String[]{testFilePath.toString(), "-j", "-p"});
     System.out.flush();
     System.setOut(origOut);
-    String[] lines = myOut.toString(StandardCharsets.UTF_8).split("\n");
-    assertEquals("{\"fileName\":\"TestFileDump.testDump.orc\",\"fileVersion\":\"0.12\",\"writerVersion\":\"ORC_14\",\"softwareVersion\":\"ORC Java unknown\",\"numberOfRows\":3,\"compression\":\"ZSTD\",\"compressionBufferSize\":262144,\"schemaString\":\"struct<x:double>\",\"schema\":{\"columnId\":0,\"columnType\":\"STRUCT\",\"children\":{\"x\":{\"columnId\":1,\"columnType\":\"DOUBLE\"}}},\"calendar\":\"Julian/Gregorian\",\"stripeStatistics\":[{\"stripeNumber\":1,\"columnStatistics\":[{\"columnId\":0,\"count\":3,\"hasNull\":false},{\"columnId\":1,\"count\":3,\"hasNull\":false,\"bytesOnDisk\":27,\"min\":NaN,\"max\":NaN,\"sum\":NaN,\"type\":\"DOUBLE\"}]}],\"fileStatistics\":[{\"columnId\":0,\"count\":3,\"hasNull\":false},{\"columnId\":1,\"count\":3,\"hasNull\":false,\"bytesOnDisk\":27,\"min\":NaN,\"max\":NaN,\"sum\":NaN,\"type\":\"DOUBLE\"}],\"stripes\":[{\"stripeNumber\":1,\"stripeInformation\":{\"offset\":3,\"indexLength\":55,\"dataLength\":27,\"footerLength\":35,\"rowCount\":3},\"streams\":[{\"columnId\":0,\"section\":\"ROW_INDEX\",\"startOffset\":3,\"length\":11},{\"columnId\":1,\"section\":\"ROW_INDEX\",\"startOffset\":14,\"length\":44},{\"columnId\":1,\"section\":\"DATA\",\"startOffset\":58,\"length\":27}],\"encodings\":[{\"columnId\":0,\"kind\":\"DIRECT\"},{\"columnId\":1,\"kind\":\"DIRECT\"}]}],\"fileLength\":286,\"rawDataSize\":36,\"paddingLength\":0,\"paddingRatio\":0.0,\"status\":\"OK\"}", lines[0]);
+    TestFileDump.checkOutput(outputFilename, workDir + File.separator + outputFilename);
   }
 }
diff --git a/java/tools/src/test/resources/orc-file-dump-nan-and-infinite.json b/java/tools/src/test/resources/orc-file-dump-nan-and-infinite.json
@@ -0,0 +1,107 @@
+{
+  "fileName": "TestFileDump.testDump.orc",
+  "fileVersion": "0.12",
+  "writerVersion": "ORC_14",
+  "softwareVersion": "ORC Java 2.1.0-SNAPSHOT",
+  "numberOfRows": 3,
+  "compression": "ZSTD",
+  "compressionBufferSize": 262144,
+  "schemaString": "struct<x:double>",
+  "schema": {
+    "columnId": 0,
+    "columnType": "STRUCT",
+    "children": {
+      "x": {
+        "columnId": 1,
+        "columnType": "DOUBLE"
+      }
+    }
+  },
+  "calendar": "Julian/Gregorian",
+  "stripeStatistics": [
+    {
+      "stripeNumber": 1,
+      "columnStatistics": [
+        {
+          "columnId": 0,
+          "count": 3,
+          "hasNull": false
+        },
+        {
+          "columnId": 1,
+          "count": 3,
+          "hasNull": false,
+          "bytesOnDisk": 27,
+          "min": NaN,
+          "max": NaN,
+          "sum": NaN,
+          "type": "DOUBLE"
+        }
+      ]
+    }
+  ],
+  "fileStatistics": [
+    {
+      "columnId": 0,
+      "count": 3,
+      "hasNull": false
+    },
+    {
+      "columnId": 1,
+      "count": 3,
+      "hasNull": false,
+      "bytesOnDisk": 27,
+      "min": NaN,
+      "max": NaN,
+      "sum": NaN,
+      "type": "DOUBLE"
+    }
+  ],
+  "stripes": [
+    {
+      "stripeNumber": 1,
+      "stripeInformation": {
+        "offset": 3,
+        "indexLength": 55,
+        "dataLength": 27,
+        "footerLength": 35,
+        "rowCount": 3
+      },
+      "streams": [
+        {
+          "columnId": 0,
+          "section": "ROW_INDEX",
+          "startOffset": 3,
+          "length": 11
+        },
+        {
+          "columnId": 1,
+          "section": "ROW_INDEX",
+          "startOffset": 14,
+          "length": 44
+        },
+        {
+          "columnId": 1,
+          "section": "DATA",
+          "startOffset": 58,
+          "length": 27
+        }
+      ],
+      "encodings": [
+        {
+          "columnId": 0,
+          "kind": "DIRECT"
+        },
+        {
+          "columnId": 1,
+          "kind": "DIRECT"
+        }
+      ]
+    }
+  ],
+  "fileLength": 293,
+  "rawDataSize": 36,
+  "paddingLength": 0,
+  "paddingRatio": 0.0,
+  "status": "OK"
+}