shujingyang-db commented on code in PR #44571:
URL: https://github.com/apache/spark/pull/44571#discussion_r1442456803
##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/xml/StaxXmlParser.scala:
##########
@@ -208,56 +202,36 @@ class StaxXmlParser(
(parser.peek, dataType) match {
case (_: StartElement, dt: DataType) => convertComplicatedType(dt,
attributes)
case (_: EndElement, _: StringType) =>
+ StaxXmlParserUtils.consumeNextEndElement(parser)
// Empty. It's null if "" is the null value
if (options.nullValue == "") {
null
} else {
UTF8String.fromString("")
}
- case (_: EndElement, _: DataType) => null
+ case (_: EndElement, _: DataType) =>
+ StaxXmlParserUtils.consumeNextEndElement(parser)
+ null
case (c: Characters, ArrayType(st, _)) =>
// For `ArrayType`, it needs to return the type of element. The values
are merged later.
parser.next
- convertTo(c.getData, st)
- case (c: Characters, st: StructType) =>
- parser.next
- parser.peek match {
- case _: EndElement =>
- // It couldn't be an array of value tags
- // as the opening tag is immediately followed by a closing tag.
- if (c.isWhiteSpace) {
- return null
- }
- val indexOpt = getFieldNameToIndex(st).get(options.valueTag)
- indexOpt match {
- case Some(index) =>
- convertTo(c.getData, st.fields(index).dataType)
- case None => null
- }
- case _ =>
- val row = convertObject(parser, st)
- if (!c.isWhiteSpace) {
- addOrUpdate(row.toSeq(st).toArray, st, options.valueTag,
c.getData, addToTail = false)
- } else {
- row
- }
- }
+ val value = convertTo(c.getData, st)
+ StaxXmlParserUtils.consumeNextEndElement(parser)
Review Comment:
Got you! I guess if the next event is Comment, Cdata, etc., we wanted to
skip those events as well and also the end element.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
