[SPARK-26339][SQL]Throws better exception when reading files that start with underscore

sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/DataSource.scala

@@ -542,7 +542,7 @@ case class DataSource(
       checkFilesExist: Boolean): Seq[Path] = {
     val allPaths = caseInsensitiveOptions.get("path") ++ paths
     val hadoopConf = sparkSession.sessionState.newHadoopConf()
-    allPaths.flatMap { path =>
+    val allGlobPath = allPaths.flatMap { path =>
       val hdfsPath = new Path(path)
       val fs = hdfsPath.getFileSystem(hadoopConf)
       val qualified = hdfsPath.makeQualified(fs.getUri, fs.getWorkingDirectory)
@@ -559,6 +559,21 @@ case class DataSource(
       }
       globPath
     }.toSeq
+
+    val (filteredOut, filteredIn) = allGlobPath.partition { path =>
+      InMemoryFileIndex.shouldFilterOut(path.getName)
+    }
+    if (filteredOut.nonEmpty) {
+      if (filteredIn.isEmpty) {
+        throw new AnalysisException(
+          s"All paths were ignored:\n${filteredOut.mkString("\n  ")}")
+      } else {
+        logDebug(
+          s"Some paths were ignored:\n${filteredOut.mkString("\n  ")}")
+      }
+    }
+
+    allGlobPath
   }
 }
 

sql/core/src/test/resources/test-data/_cars.csv

@@ -0,0 +1,7 @@
+
+year,make,model,comment,blank
+"2012","Tesla","S","No comment",
+
+1997,Ford,E350,"Go get one now they are going fast",
+2015,Chevy,Volt
+

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

@@ -53,6 +53,7 @@ class CSVSuite extends QueryTest with SharedSQLContext with SQLTestUtils with Te
   private val carsEmptyValueFile = "test-data/cars-empty-value.csv"
   private val carsBlankColName = "test-data/cars-blank-column-name.csv"
   private val carsCrlf = "test-data/cars-crlf.csv"
+  private val carsFilteredOutFile = "test-data/_cars.csv"
   private val emptyFile = "test-data/empty.csv"
   private val commentsFile = "test-data/comments.csv"
   private val disableCommentsFile = "test-data/disable_comments.csv"
@@ -345,6 +346,25 @@ class CSVSuite extends QueryTest with SharedSQLContext with SQLTestUtils with Te
     assert(result.schema.fieldNames.size === 1)
   }
 
+  test("SPARK-26339 Not throw an exception if some of specified paths are filtered in") {
+    val cars = spark
+      .read
+      .option("header", "false")
+      .csv(testFile(carsFile), testFile(carsFilteredOutFile))
+
+    verifyCars(cars, withHeader = false, checkTypes = false)
+  }
+
+  test("SPARK-26339 Throw an exception only if all of the specified paths are filtered out") {
+    val e = intercept[AnalysisException] {
+      val cars = spark
+        .read
+        .option("header", "false")
+        .csv(testFile(carsFilteredOutFile))
+    }.getMessage
+    assert(e.contains("All paths were ignored:"))
+  }
+
   test("DDL test with empty file") {
     withView("carsTable") {
       spark.sql(