Hisoka-X commented on code in PR #42220:
URL: https://github.com/apache/spark/pull/42220#discussion_r1297961931
##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/TableOutputResolver.scala:
##########
@@ -238,11 +238,17 @@ object TableOutputResolver {
if (reordered.length == expectedCols.length) {
if (matchedCols.size < inputCols.length) {
- val extraCols = inputCols.filterNot(col =>
matchedCols.contains(col.name))
- .map(col => s"${toSQLId(col.name)}").mkString(", ")
- throw
QueryCompilationErrors.incompatibleDataToTableExtraStructFieldsError(
- tableName, colPath.quoted, extraCols
- )
+ if (colPath.isEmpty) {
+ val cannotFindCol = expectedCols.filter(col =>
!matchedCols.contains(col.name)).head.name
+ throw
QueryCompilationErrors.incompatibleDataToTableCannotFindDataError(tableName,
Review Comment:
> if (reordered.length == expectedCols.length) I think this means there is
no missing col
Yep, but there are some special for V1 case, eg: table [x, y, z], inputcol:
[x, y, k], when we set `fillDefaultValue` as true, the `reordered` will be [x,
y, null as z]. Then `reordered.length == expectedCols.length` will be true.
This PR only change the error to `CANNOT_FIND_DATA`, but I'm not sure we
should support case like `table [x, y, z], inputcol: [x, y, k]`? Then just put
`z` as `null`.
In the future, when we support default value on DS V2, we will face this
problem again.
--
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]
