Skip to content

[Spark] Only attempt server-side planning for Delta tables in loadTable #7039

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
chenwang-databricks wants to merge 2 commits into
base: master
Choose a base branch
from
+11 −7
Open

[Spark] Only attempt server-side planning for Delta tables in loadTable #7039

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f5bfd5e
[Spark] Only attempt server-side planning for Delta tables in loadTable
chenwang-databricks Jun 16, 2026
59c766d
[Spark] Use Delta tables in ServerSidePlannedTableSuite fixtures
chenwang-databricks Jun 22, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 9 additions & 5 deletions spark/src/main/scala/org/apache/spark/sql/delta/catalog/AbstractDeltaCatalog.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -374,13 +374,17 @@ class AbstractDeltaCatalog extends DelegatingCatalogExtension
.map(_.loadTable(ident))
.getOrElse(super.loadTable(ident))

ServerSidePlannedTable.tryCreate(spark, ident, table, isUnityCatalog).foreach { sspt =>
return sspt
}

table match {
case v1: V1Table if DeltaTableUtils.isDeltaTable(v1.catalogTable) =>
loadCatalogTable(ident, v1.catalogTable)
// Server-side planning only applies to Delta tables. Attempt it here, inside the
// Delta-`V1Table` branch, rather than on every loaded table: a non-Delta table or a
// catalog-specific shape (e.g. Unity Catalog's `MetadataTable` wrapping a `ViewInfo`
// for a metric view) must flow through the `case o => o` fallthrough unchanged so the
// resolver can route it correctly. Capturing those in `ServerSidePlannedTable.tryCreate`
// would short-circuit view loading.
ServerSidePlannedTable
.tryCreate(spark, ident, table, isUnityCatalog)
.getOrElse(loadCatalogTable(ident, v1.catalogTable))
case o => o
}
} catch {
Expand Down
Loading