jonahgao commented on code in PR #14102:
URL: https://github.com/apache/datafusion/pull/14102#discussion_r1912749696
##########
datafusion/expr/src/utils.rs:
##########
@@ -705,27 +711,20 @@ pub fn exprlist_to_fields<'a>(
.map(|e| match e {
Expr::Wildcard { qualifier, options } => match qualifier {
Review Comment:
Although we have moved wildcard expansions to the analyzer
https://github.com/apache/datafusion/pull/11681, it still does wildcard
expansions when computing plan schemas(in
[exprlist_to_fields](https://github.com/apache/datafusion/blob/f9cc3325cdb5891b7566a6f3503c1f7ac6ad51e0/datafusion/expr/src/utils.rs#L696)
and
[exprlist_len](https://github.com/apache/datafusion/blob/f9cc3325cdb5891b7566a6f3503c1f7ac6ad51e0/datafusion/expr/src/utils.rs#L797C8-L797C20)).
I wonder if performing wildcard expansions before computing schemas would be
simplier, at least it would avoid redundant work.
##########
datafusion/expr/src/utils.rs:
##########
@@ -379,14 +379,12 @@ fn get_exprs_except_skipped(
}
}
-/// Resolves an `Expr::Wildcard` to a collection of `Expr::Column`'s.
-pub fn expand_wildcard(
- schema: &DFSchema,
- plan: &LogicalPlan,
- wildcard_options: Option<&WildcardOptions>,
-) -> Result<Vec<Expr>> {
+/// For each column specified in the USING JOIN condition, the JOIN plan
outputs it twice
+/// (once for each join side), but an unqualified wildcard should include it
only once.
+/// This function returns the columns that should be excluded.
+fn exclude_using_columns(plan: &LogicalPlan) -> Result<HashSet<Column>> {
Review Comment:
Although we have moved wildcard expansions to the analyzer #11681, it still
does wildcard expansions when computing plan schemas(in
[exprlist_to_fields](https://github.com/apache/datafusion/blob/f9cc3325cdb5891b7566a6f3503c1f7ac6ad51e0/datafusion/expr/src/utils.rs#L696)
and
[exprlist_len](https://github.com/apache/datafusion/blob/f9cc3325cdb5891b7566a6f3503c1f7ac6ad51e0/datafusion/expr/src/utils.rs#L797C8-L797C20)).
I wonder if performing wildcard expansions before computing schemas would be
simplier, at least it would avoid redundant work.
--
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: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
