[SPARK-17007][SQL] Move test data files into a test-data folder

sql/core/src/test/java/test/org/apache/spark/sql/JavaDataFrameSuite.java

@@ -327,23 +327,23 @@ private String getResource(String resource) {
 
   @Test
   public void testGenericLoad() {
-    Dataset<Row> df1 = spark.read().format("text").load(getResource("text-suite.txt"));
+    Dataset<Row> df1 = spark.read().format("text").load(getResource("test-data/text-suite.txt"));
     Assert.assertEquals(4L, df1.count());
 
     Dataset<Row> df2 = spark.read().format("text").load(
-      getResource("text-suite.txt"),
-      getResource("text-suite2.txt"));
+      getResource("test-data/text-suite.txt"),
+      getResource("test-data/text-suite2.txt"));
     Assert.assertEquals(5L, df2.count());
   }
 
   @Test
   public void testTextLoad() {
-    Dataset<String> ds1 = spark.read().textFile(getResource("text-suite.txt"));
+    Dataset<String> ds1 = spark.read().textFile(getResource("test-data/text-suite.txt"));
     Assert.assertEquals(4L, ds1.count());
 
     Dataset<String> ds2 = spark.read().textFile(
-      getResource("text-suite.txt"),
-      getResource("text-suite2.txt"));
+      getResource("test-data/text-suite.txt"),
+      getResource("test-data/text-suite2.txt"));
     Assert.assertEquals(5L, ds2.count());
   }
 

sql/core/src/test/scala/org/apache/spark/sql/execution/command/DDLSuite.scala

@@ -677,7 +677,8 @@ class DDLSuite extends QueryTest with SharedSQLContext with BeforeAndAfterEach {
   }
 
   test("create temporary view using") {
-    val csvFile = Thread.currentThread().getContextClassLoader.getResource("cars.csv").toString()
+    val csvFile =
+      Thread.currentThread().getContextClassLoader.getResource("test-data/cars.csv").toString
     withView("testview") {
       sql(s"CREATE OR REPLACE TEMPORARY VIEW testview (c1: String, c2: String)  USING " +
         "org.apache.spark.sql.execution.datasources.csv.CSVFileFormat  " +

sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVSuite.scala

@@ -33,23 +33,23 @@ import org.apache.spark.sql.types._
 class CSVSuite extends QueryTest with SharedSQLContext with SQLTestUtils {
   import testImplicits._
 
-  private val carsFile = "cars.csv"
-  private val carsMalformedFile = "cars-malformed.csv"
-  private val carsFile8859 = "cars_iso-8859-1.csv"
-  private val carsTsvFile = "cars.tsv"
-  private val carsAltFile = "cars-alternative.csv"
-  private val carsUnbalancedQuotesFile = "cars-unbalanced-quotes.csv"
-  private val carsNullFile = "cars-null.csv"
-  private val carsBlankColName = "cars-blank-column-name.csv"
-  private val emptyFile = "empty.csv"
-  private val commentsFile = "comments.csv"
-  private val disableCommentsFile = "disable_comments.csv"
-  private val boolFile = "bool.csv"
-  private val decimalFile = "decimal.csv"
-  private val simpleSparseFile = "simple_sparse.csv"
-  private val numbersFile = "numbers.csv"
-  private val datesFile = "dates.csv"
-  private val unescapedQuotesFile = "unescaped-quotes.csv"
+  private val carsFile = "test-data/cars.csv"
+  private val carsMalformedFile = "test-data/cars-malformed.csv"
+  private val carsFile8859 = "test-data/cars_iso-8859-1.csv"
+  private val carsTsvFile = "test-data/cars.tsv"
+  private val carsAltFile = "test-data/cars-alternative.csv"
+  private val carsUnbalancedQuotesFile = "test-data/cars-unbalanced-quotes.csv"
+  private val carsNullFile = "test-data/cars-null.csv"
+  private val carsBlankColName = "test-data/cars-blank-column-name.csv"
+  private val emptyFile = "test-data/empty.csv"
+  private val commentsFile = "test-data/comments.csv"
+  private val disableCommentsFile = "test-data/disable_comments.csv"
+  private val boolFile = "test-data/bool.csv"
+  private val decimalFile = "test-data/decimal.csv"
+  private val simpleSparseFile = "test-data/simple_sparse.csv"
+  private val numbersFile = "test-data/numbers.csv"
+  private val datesFile = "test-data/dates.csv"
+  private val unescapedQuotesFile = "test-data/unescaped-quotes.csv"
 
   private def testFile(fileName: String): String = {
     Thread.currentThread().getContextClassLoader.getResource(fileName).toString

sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetIOSuite.scala

@@ -568,7 +568,7 @@ class ParquetIOSuite extends QueryTest with ParquetTest with SharedSQLContext {
       withSQLConf(SQLConf.PARQUET_VECTORIZED_READER_ENABLED.key -> vectorized) {
         checkAnswer(
           // Decimal column in this file is encoded using plain dictionary
-          readResourceParquetFile("dec-in-i32.parquet"),
+          readResourceParquetFile("test-data/dec-in-i32.parquet"),
           spark.range(1 << 4).select('id % 10 cast DecimalType(5, 2) as 'i32_dec))
       }
     }
@@ -579,7 +579,7 @@ class ParquetIOSuite extends QueryTest with ParquetTest with SharedSQLContext {
       withSQLConf(SQLConf.PARQUET_VECTORIZED_READER_ENABLED.key -> vectorized) {
         checkAnswer(
           // Decimal column in this file is encoded using plain dictionary
-          readResourceParquetFile("dec-in-i64.parquet"),
+          readResourceParquetFile("test-data/dec-in-i64.parquet"),
           spark.range(1 << 4).select('id % 10 cast DecimalType(10, 2) as 'i64_dec))
       }
     }
@@ -590,7 +590,7 @@ class ParquetIOSuite extends QueryTest with ParquetTest with SharedSQLContext {
       withSQLConf(SQLConf.PARQUET_VECTORIZED_READER_ENABLED.key -> vectorized) {
         checkAnswer(
           // Decimal column in this file is encoded using plain dictionary
-          readResourceParquetFile("dec-in-fixed-len.parquet"),
+          readResourceParquetFile("test-data/dec-in-fixed-len.parquet"),
           spark.range(1 << 4).select('id % 10 cast DecimalType(10, 2) as 'fixed_len_dec))
       }
     }

sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetProtobufCompatibilitySuite.scala

@@ -22,27 +22,27 @@ import org.apache.spark.sql.test.SharedSQLContext
 
 class ParquetProtobufCompatibilitySuite extends ParquetCompatibilityTest with SharedSQLContext {
   test("unannotated array of primitive type") {
-    checkAnswer(readResourceParquetFile("old-repeated-int.parquet"), Row(Seq(1, 2, 3)))
+    checkAnswer(readResourceParquetFile("test-data/old-repeated-int.parquet"), Row(Seq(1, 2, 3)))
   }
 
   test("unannotated array of struct") {
     checkAnswer(
-      readResourceParquetFile("old-repeated-message.parquet"),
+      readResourceParquetFile("test-data/old-repeated-message.parquet"),
       Row(
         Seq(
           Row("First inner", null, null),
           Row(null, "Second inner", null),
           Row(null, null, "Third inner"))))
 
     checkAnswer(
-      readResourceParquetFile("proto-repeated-struct.parquet"),
+      readResourceParquetFile("test-data/proto-repeated-struct.parquet"),
       Row(
         Seq(
           Row("0 - 1", "0 - 2", "0 - 3"),
           Row("1 - 1", "1 - 2", "1 - 3"))))
 
     checkAnswer(
-      readResourceParquetFile("proto-struct-with-array-many.parquet"),
+      readResourceParquetFile("test-data/proto-struct-with-array-many.parquet"),
       Seq(
         Row(
           Seq(
@@ -60,13 +60,13 @@ class ParquetProtobufCompatibilitySuite extends ParquetCompatibilityTest with Sh
 
   test("struct with unannotated array") {
     checkAnswer(
-      readResourceParquetFile("proto-struct-with-array.parquet"),
+      readResourceParquetFile("test-data/proto-struct-with-array.parquet"),
       Row(10, 9, Seq.empty, null, Row(9), Seq(Row(9), Row(10))))
   }
 
   test("unannotated array of struct with unannotated array") {
     checkAnswer(
-      readResourceParquetFile("nested-array-struct.parquet"),
+      readResourceParquetFile("test-data/nested-array-struct.parquet"),
       Seq(
         Row(2, Seq(Row(1, Seq(Row(3))))),
         Row(5, Seq(Row(4, Seq(Row(6))))),
@@ -75,7 +75,7 @@ class ParquetProtobufCompatibilitySuite extends ParquetCompatibilityTest with Sh
 
   test("unannotated array of string") {
     checkAnswer(
-      readResourceParquetFile("proto-repeated-string.parquet"),
+      readResourceParquetFile("test-data/proto-repeated-string.parquet"),
       Seq(
         Row(Seq("hello", "world")),
         Row(Seq("good", "bye")),

sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetThriftCompatibilitySuite.scala

@@ -23,8 +23,8 @@ import org.apache.spark.sql.test.SharedSQLContext
 class ParquetThriftCompatibilitySuite extends ParquetCompatibilityTest with SharedSQLContext {
   import ParquetCompatibilityTest._
 
-  private val parquetFilePath =
-    Thread.currentThread().getContextClassLoader.getResource("parquet-thrift-compat.snappy.parquet")
+  private val parquetFilePath = Thread.currentThread().getContextClassLoader.getResource(
+    "test-data/parquet-thrift-compat.snappy.parquet")
 
   test("Read Parquet file generated by parquet-thrift") {
     logInfo(

sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/text/TextSuite.scala

@@ -66,7 +66,7 @@ class TextSuite extends QueryTest with SharedSQLContext {
 
   test("reading partitioned data using read.textFile()") {
     val partitionedData = Thread.currentThread().getContextClassLoader
-      .getResource("text-partitioned").toString
+      .getResource("test-data/text-partitioned").toString
     val ds = spark.read.textFile(partitionedData)
     val data = ds.collect()
 
@@ -76,7 +76,7 @@ class TextSuite extends QueryTest with SharedSQLContext {
 
   test("support for partitioned reading using read.text()") {
     val partitionedData = Thread.currentThread().getContextClassLoader
-      .getResource("text-partitioned").toString
+      .getResource("test-data/text-partitioned").toString
     val df = spark.read.text(partitionedData)
     val data = df.filter("year = '2015'").select("value").collect()
 
@@ -155,7 +155,7 @@ class TextSuite extends QueryTest with SharedSQLContext {
   }
 
   private def testFile: String = {
-    Thread.currentThread().getContextClassLoader.getResource("text-suite.txt").toString
+    Thread.currentThread().getContextClassLoader.getResource("test-data/text-suite.txt").toString
   }
 
   /** Verifies data and schema. */