[SPARK-39910][SQL] Delegate path qualification to filesystem during DataSource file path globbing

dev/.rat-excludes

@@ -138,3 +138,4 @@ people.xml
 ui-test/package.json
 ui-test/package-lock.json
 core/src/main/resources/org/apache/spark/ui/static/package.json
+.*\.har

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

@@ -760,7 +760,7 @@ object DataSource extends Logging {
     val qualifiedPaths = pathStrings.map { pathString =>
       val path = new Path(pathString)
       val fs = path.getFileSystem(hadoopConf)
-      path.makeQualified(fs.getUri, fs.getWorkingDirectory)
+      fs.makeQualified(path)
     }
 
     // Split the paths into glob and non glob paths, because we don't need to do an existence check

sql/core/src/test/resources/test-data/test-archive.har/_index

@@ -0,0 +1,2 @@
+%2F dir 1707380620211+493+tigrulya+hadoop 0 0 test.csv 
+%2Ftest.csv file part-0 0 6 1707380620197+420+tigrulya+hadoop 

sql/core/src/test/resources/test-data/test-archive.har/_masterindex

@@ -0,0 +1,2 @@
+3 
+0 1948547033 0 119 

sql/core/src/test/resources/test-data/test-archive.har/part-0

@@ -0,0 +1,3 @@
+1
+2
+3

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

@@ -17,6 +17,8 @@
 
 package org.apache.spark.sql.execution.datasources
 
+import java.net.URI
+
 import org.apache.hadoop.conf.Configuration
 import org.apache.hadoop.fs.{FileStatus, Path, RawLocalFileSystem}
 import org.scalatest.PrivateMethodTester
@@ -214,4 +216,6 @@ class MockFileSystem extends RawLocalFileSystem {
   override def globStatus(pathPattern: Path): Array[FileStatus] = {
     mockGlobResults.getOrElse(pathPattern, Array())
   }
+
+  override def getUri: URI = URI.create("mockFs://mockFs/")
 }

sql/core/src/test/scala/org/apache/spark/sql/test/DataFrameReaderWriterSuite.scala

@@ -1363,4 +1363,12 @@ class DataFrameReaderWriterSuite extends QueryTest with SharedSparkSession with
       }
     }
   }
+
+  test("SPARK-39910: read files from Hadoop archives") {
+    val fileSchema = new StructType().add("str", StringType)
+    val harPath = testFile("test-data/test-archive.har")
+      .replaceFirst("file:/", "har:/")
+
+    testRead(spark.read.schema(fileSchema).csv(s"$harPath/test.csv"), data, fileSchema)
+  }
 }