alamb commented on code in PR #7243:
URL: https://github.com/apache/arrow-datafusion/pull/7243#discussion_r1288327485


##########
datafusion/core/src/physical_plan/coalesce_batches.rs:
##########
@@ -305,46 +305,10 @@ pub fn concat_batches(
 #[cfg(test)]
 mod tests {
     use super::*;
-    use crate::config::ConfigOptions;
-    use crate::datasource::MemTable;
-    use crate::physical_plan::filter::FilterExec;
     use crate::physical_plan::{memory::MemoryExec, 
repartition::RepartitionExec};
-    use crate::prelude::SessionContext;
     use crate::test::create_vec_batches;
     use arrow::datatypes::{DataType, Field, Schema};
 
-    #[tokio::test]

Review Comment:
   These are tests that a config option makes it through to planning SQL, so I 
moved them to `options.slt`



##########
datafusion/core/src/physical_plan/aggregates/mod.rs:
##########
@@ -1231,7 +1230,6 @@ mod tests {
         DisplayAs, ExecutionPlan, Partitioning, RecordBatchStream,
         SendableRecordBatchStream, Statistics,
     };
-    use crate::prelude::SessionContext;

Review Comment:
   The point of this PR is to remove these dependencies



##########
datafusion/core/src/physical_plan/joins/symmetric_hash_join.rs:
##########
@@ -1811,173 +1799,14 @@ mod tests {
         Ok(())
     }
 
-    #[tokio::test]
-    async fn join_change_in_planner() -> Result<()> {

Review Comment:
   these are SQL level tests that need to entire `SessionContext` so I moved it 
to the `core_integration` test



##########
datafusion/core/src/physical_plan/aggregates/mod.rs:
##########
@@ -1449,8 +1447,7 @@ mod tests {
             DataType::Int64,
         ))];
 
-        let session_ctx = SessionContext::new();
-        let task_ctx = session_ctx.task_ctx();
+        let task_ctx = Arc::new(TaskContext::default());

Review Comment:
   This pattern is 95% of this PR: create the `TaskContext` directly rather 
than via a `SessionContext`



-- 
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]

Reply via email to