You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@datafusion.apache.org by "rkrishn7 (via GitHub)" <gi...@apache.org> on 2025/09/22 00:40:45 UTC
[PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
rkrishn7 opened a new pull request, #17703:
URL: https://github.com/apache/datafusion/pull/17703
## Which issue does this PR close?
<!--
We generally require a GitHub issue to be filed for all bug fixes and enhancements and this helps us generate change logs for our releases. You can link an issue to this PR using the GitHub syntax. For example `Closes #123` indicates that this PR will close issue #123.
-->
- Closes https://github.com/apache/datafusion/issues/17581
## What changes are included in this PR?
<!--
There is no need to duplicate the description in the issue here but it is sometimes worth providing a summary of the individual changes in this PR.
-->
- Renames `reassign_predicate_columns` to `reassign_expr_columns`
- Updates usage
## Are these changes tested?
No, there should be no changes in functionality.
## Are there any user-facing changes?
<!--
If there are user-facing changes then we may require documentation to be updated before approving the PR.
-->
<!--
If there are any breaking changes to public APIs, please add the `api change` label.
-->
N/A
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
Posted by "rkrishn7 (via GitHub)" <gi...@apache.org>.
rkrishn7 commented on code in PR #17703:
URL: https://github.com/apache/datafusion/pull/17703#discussion_r2366473188
##########
datafusion/physical-plan/src/filter.rs:
##########
@@ -716,7 +716,7 @@ impl RecordBatchStream for FilterExecStream {
}
/// Return the equals Column-Pairs and Non-equals Column-Pairs
-pub fn collect_columns_from_predicate(
+fn collect_columns_from_predicate(
Review Comment:
I'm just realizing this was `pub` and not `pub(crate)` so technically this is an API change
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
Posted by "rkrishn7 (via GitHub)" <gi...@apache.org>.
rkrishn7 commented on code in PR #17703:
URL: https://github.com/apache/datafusion/pull/17703#discussion_r2384272903
##########
datafusion/datasource/src/file_scan_config.rs:
##########
@@ -764,24 +751,24 @@ impl FileScanConfig {
eq_properties: &mut EquivalenceProperties,
schema: &Schema,
) -> Result<()> {
- macro_rules! ignore_dangling_col {
- ($col:expr) => {
- if let Some(col) = $col.as_any().downcast_ref::<Column>() {
- if schema.index_of(col.name()).is_err() {
- continue;
+ // Gather valid equality pairs from the filter expression
+ let equal_pairs = split_conjunction(&filter).into_iter().filter_map(|expr| {
+ // Ignore any binary expressions that reference non-existent columns in the current schema
+ // (e.g. due to unnecessary projections being removed)
+ reassign_expr_columns(Arc::clone(expr), schema)
+ .ok()
Review Comment:
Yeah `reassign_expr_columns` returns a `Result` and we're filter mapping over the list of expressions so need to return an Option.
In this case it's okay to discard the result I think (per the comment)
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
Posted by "alamb (via GitHub)" <gi...@apache.org>.
alamb merged PR #17703:
URL: https://github.com/apache/datafusion/pull/17703
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
Posted by "rkrishn7 (via GitHub)" <gi...@apache.org>.
rkrishn7 commented on PR #17703:
URL: https://github.com/apache/datafusion/pull/17703#issuecomment-3339934947
Hey @alamb sorry I've been quite busy this week.
I plan to address your comments either today or tomorrow!
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
Posted by "alamb (via GitHub)" <gi...@apache.org>.
alamb commented on code in PR #17703:
URL: https://github.com/apache/datafusion/pull/17703#discussion_r2368556894
##########
datafusion/physical-plan/src/filter.rs:
##########
@@ -716,7 +716,7 @@ impl RecordBatchStream for FilterExecStream {
}
/// Return the equals Column-Pairs and Non-equals Column-Pairs
-pub fn collect_columns_from_predicate(
+fn collect_columns_from_predicate(
Review Comment:
maybe you can leave a `#deprecated` version of the old name (that just calles the new name) as described in https://datafusion.apache.org/contributor-guide/api-health.html#deprecation-guidelines
##########
datafusion/datasource/src/file_scan_config.rs:
##########
@@ -764,24 +751,24 @@ impl FileScanConfig {
eq_properties: &mut EquivalenceProperties,
schema: &Schema,
) -> Result<()> {
- macro_rules! ignore_dangling_col {
- ($col:expr) => {
- if let Some(col) = $col.as_any().downcast_ref::<Column>() {
- if schema.index_of(col.name()).is_err() {
- continue;
+ // Gather valid equality pairs from the filter expression
+ let equal_pairs = split_conjunction(&filter).into_iter().filter_map(|expr| {
+ // Ignore any binary expressions that reference non-existent columns in the current schema
+ // (e.g. due to unnecessary projections being removed)
+ reassign_expr_columns(Arc::clone(expr), schema)
+ .ok()
Review Comment:
is this `ok` actually required? Does anything fail if we remove it?
It seems suspicious to me (though I see it is how the old code works too)
##########
datafusion/physical-expr/src/utils/mod.rs:
##########
@@ -238,22 +238,26 @@ pub fn collect_columns(expr: &Arc<dyn PhysicalExpr>) -> HashSet<Column> {
columns
}
-/// Re-assign column indices referenced in predicate according to given schema.
-/// This may be helpful when dealing with projections.
-pub fn reassign_predicate_columns(
- pred: Arc<dyn PhysicalExpr>,
+/// Re-assign indices of [`Column`]s within the given [`PhysicalExpr`] according to
+/// the provided [`Schema`].
+///
+/// This can be useful when attempting to map an expression onto a different schema.
+///
+/// # Errors
+///
+/// This function will return an error if any column in the expression cannot be found
+/// in the provided schema.
+pub fn reassign_expr_columns(
+ expr: Arc<dyn PhysicalExpr>,
schema: &Schema,
- ignore_not_found: bool,
) -> Result<Arc<dyn PhysicalExpr>> {
- pred.transform_down(|expr| {
- let expr_any = expr.as_any();
-
- if let Some(column) = expr_any.downcast_ref::<Column>() {
+ expr.transform_down(|expr| {
+ if let Some(column) = expr.as_any().downcast_ref::<Column>() {
let index = match schema.index_of(column.name()) {
Ok(idx) => idx,
- Err(_) if ignore_not_found => usize::MAX,
Err(e) => return Err(e.into()),
};
Review Comment:
You could further simplify this to
```suggestion
let index = match schema.index_of(column.name())?;
```
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
Posted by "rkrishn7 (via GitHub)" <gi...@apache.org>.
rkrishn7 commented on PR #17703:
URL: https://github.com/apache/datafusion/pull/17703#issuecomment-3341968990
@alamb Updated w/ your suggestions!
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
Posted by "alamb (via GitHub)" <gi...@apache.org>.
alamb commented on PR #17703:
URL: https://github.com/apache/datafusion/pull/17703#issuecomment-3339930122
@rkrishn7 are we good to merge this PR ? Or do you plan to address the comments prior to merge?
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
Posted by "alamb (via GitHub)" <gi...@apache.org>.
alamb commented on PR #17703:
URL: https://github.com/apache/datafusion/pull/17703#issuecomment-3343303262
Thank you @rkrishn7
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
Posted by "alamb (via GitHub)" <gi...@apache.org>.
alamb commented on PR #17703:
URL: https://github.com/apache/datafusion/pull/17703#issuecomment-3340194143
> Hey @alamb sorry I've been quite busy this week.
>
> I plan to address your comments either today or tomorrow!
Ok, no problem! We can also do it as a follow on PR too if you prefer
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] chore: refactor usage of `reassign_predicate_columns` [datafusion]
Posted by "rkrishn7 (via GitHub)" <gi...@apache.org>.
rkrishn7 commented on code in PR #17703:
URL: https://github.com/apache/datafusion/pull/17703#discussion_r2384274536
##########
datafusion/physical-expr/src/utils/mod.rs:
##########
@@ -238,22 +238,26 @@ pub fn collect_columns(expr: &Arc<dyn PhysicalExpr>) -> HashSet<Column> {
columns
}
-/// Re-assign column indices referenced in predicate according to given schema.
-/// This may be helpful when dealing with projections.
-pub fn reassign_predicate_columns(
- pred: Arc<dyn PhysicalExpr>,
+/// Re-assign indices of [`Column`]s within the given [`PhysicalExpr`] according to
+/// the provided [`Schema`].
+///
+/// This can be useful when attempting to map an expression onto a different schema.
+///
+/// # Errors
+///
+/// This function will return an error if any column in the expression cannot be found
+/// in the provided schema.
+pub fn reassign_expr_columns(
+ expr: Arc<dyn PhysicalExpr>,
schema: &Schema,
- ignore_not_found: bool,
) -> Result<Arc<dyn PhysicalExpr>> {
- pred.transform_down(|expr| {
- let expr_any = expr.as_any();
-
- if let Some(column) = expr_any.downcast_ref::<Column>() {
+ expr.transform_down(|expr| {
+ if let Some(column) = expr.as_any().downcast_ref::<Column>() {
let index = match schema.index_of(column.name()) {
Ok(idx) => idx,
- Err(_) if ignore_not_found => usize::MAX,
Err(e) => return Err(e.into()),
};
Review Comment:
Yes good catch!
--
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-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org