diff --git a/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/CSVOptions.scala b/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/CSVOptions.scala index bea8e97a9a94fc5c196d4e93ffe1854812881045..38aa2dd80a4f13d0a102d1ec2102dd9d8a0dce04 100644 --- a/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/CSVOptions.scala +++ b/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/CSVOptions.scala @@ -75,9 +75,6 @@ private[sql] class CSVOptions( val ignoreLeadingWhiteSpaceFlag = getBool("ignoreLeadingWhiteSpace") val ignoreTrailingWhiteSpaceFlag = getBool("ignoreTrailingWhiteSpace") - // Limit the number of lines we'll search for a header row that isn't comment-prefixed - val MAX_COMMENT_LINES_IN_HEADER = 10 - // Parse mode flags if (!ParseModes.isValidMode(parseMode)) { logWarning(s"$parseMode is not a valid parse mode. Using ${ParseModes.DEFAULT}.") diff --git a/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/CSVRelation.scala b/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/CSVRelation.scala index f8e3a1b6d46ae4c2999cfad5df62394bd8c054fe..471ed0d5604b232baeb843af8ca9c7185f42baf1 100644 --- a/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/CSVRelation.scala +++ b/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/CSVRelation.scala @@ -154,12 +154,14 @@ private[csv] class CSVRelation( */ private def findFirstLine(rdd: RDD[String]): String = { if (params.isCommentSet) { - rdd.take(params.MAX_COMMENT_LINES_IN_HEADER) - .find(!_.startsWith(params.comment.toString)) - .getOrElse(sys.error(s"No uncommented header line in " + - s"first ${params.MAX_COMMENT_LINES_IN_HEADER} lines")) + val comment = params.comment.toString + rdd.filter { line => + line.trim.nonEmpty && !line.startsWith(comment) + }.first() } else { - rdd.first() + rdd.filter { line => + line.trim.nonEmpty + }.first() } } } diff --git a/sql/core/src/test/resources/cars.csv b/sql/core/src/test/resources/cars.csv index 2b9d74ca607ad70d6496d30533c5f01184784da1..40ded573ade5c69c3c5e087d5ab95434250c0c12 100644 --- a/sql/core/src/test/resources/cars.csv +++ b/sql/core/src/test/resources/cars.csv @@ -1,3 +1,4 @@ + year,make,model,comment,blank "2012","Tesla","S","No comment",