Migrate physical plan tests to insta (Part-1) (#15313)

* Basic setup done for physical plan insta tests

* Migrated tests to insta in aggregate/mod.rs

* Migrated tests to insta in aggregate/topk/priority_map.rs

* Migrated tests to insta in joins

* Fallback to previous implementation in a test due to an issue (#15312)

* Fix formatting

* Update Cargo.toml

* Revert any Cargo.lock changes

* Clean workspace and attempt fixing failing build test

* format Cargo.toml

* Add insta as a dev-dependency instead

* Update datafusion/physical-plan/src/aggregates/mod.rs

Co-authored-by: Dmitrii Blaginin <[email protected]>

* Preserved comments from aggregates/mod.rs

* Resolved errors regarding dependencies and formatting

---------

Co-authored-by: Dmitrii Blaginin <[email protected]>

Author: Shreyaskr1409

Diff for: Cargo.lock

@@ -68,6 +68,7 @@ tokio = { workspace = true }
 criterion = { workspace = true, features = ["async_futures"] }
 datafusion-functions-aggregate = { workspace = true }
 datafusion-functions-window = { workspace = true }
+insta = { workspace = true }
 rand = { workspace = true }
 rstest = { workspace = true }
 rstest_reuse = "0.7.0"

Diff for: datafusion/physical-plan/Cargo.toml

@@ -113,6 +113,7 @@ mod tests {
     };
     use arrow::datatypes::{Field, Schema, SchemaRef};
     use arrow::util::pretty::pretty_format_batches;
+    use insta::assert_snapshot;
     use std::sync::Arc;
 
     #[test]
@@ -180,15 +181,15 @@ mod tests {
         let cols = agg.emit()?;
         let batch = RecordBatch::try_new(test_schema(), cols)?;
         let actual = format!("{}", pretty_format_batches(&[batch])?);
-        let expected = r#"
+
+        assert_snapshot!(actual, @r#"
 +----------+--------------+
 | trace_id | timestamp_ms |
 +----------+--------------+
 | 1        | 1            |
 +----------+--------------+
         "#
-        .trim();
-        assert_eq!(actual, expected);
+        );
 
         Ok(())
     }
@@ -205,15 +206,15 @@ mod tests {
         let cols = agg.emit()?;
         let batch = RecordBatch::try_new(test_schema(), cols)?;
         let actual = format!("{}", pretty_format_batches(&[batch])?);
-        let expected = r#"
+
+        assert_snapshot!(actual, @r#"
 +----------+--------------+
 | trace_id | timestamp_ms |
 +----------+--------------+
 | 1        | 1            |
 +----------+--------------+
         "#
-        .trim();
-        assert_eq!(actual, expected);
+        );
 
         Ok(())
     }
@@ -230,15 +231,14 @@ mod tests {
         let cols = agg.emit()?;
         let batch = RecordBatch::try_new(test_schema(), cols)?;
         let actual = format!("{}", pretty_format_batches(&[batch])?);
-        let expected = r#"
+        assert_snapshot!(actual, @r#"
 +----------+--------------+
 | trace_id | timestamp_ms |
 +----------+--------------+
 | 2        | 2            |
 +----------+--------------+
         "#
-        .trim();
-        assert_eq!(actual, expected);
+        );
 
         Ok(())
     }
@@ -255,15 +255,14 @@ mod tests {
         let cols = agg.emit()?;
         let batch = RecordBatch::try_new(test_schema(), cols)?;
         let actual = format!("{}", pretty_format_batches(&[batch])?);
-        let expected = r#"
+        assert_snapshot!(actual, @r#"
 +----------+--------------+
 | trace_id | timestamp_ms |
 +----------+--------------+
 | 1        | 1            |
 +----------+--------------+
         "#
-        .trim();
-        assert_eq!(actual, expected);
+        );
 
         Ok(())
     }
@@ -280,15 +279,14 @@ mod tests {
         let cols = agg.emit()?;
         let batch = RecordBatch::try_new(test_schema(), cols)?;
         let actual = format!("{}", pretty_format_batches(&[batch])?);
-        let expected = r#"
+        assert_snapshot!(actual, @r#"
 +----------+--------------+
 | trace_id | timestamp_ms |
 +----------+--------------+
 | 1        | 2            |
 +----------+--------------+
         "#
-        .trim();
-        assert_eq!(actual, expected);
+        );
 
         Ok(())
     }
@@ -305,15 +303,14 @@ mod tests {
         let cols = agg.emit()?;
         let batch = RecordBatch::try_new(test_schema(), cols)?;
         let actual = format!("{}", pretty_format_batches(&[batch])?);
-        let expected = r#"
+        assert_snapshot!(actual, @r#"
 +----------+--------------+
 | trace_id | timestamp_ms |
 +----------+--------------+
 | 1        | 1            |
 +----------+--------------+
         "#
-        .trim();
-        assert_eq!(actual, expected);
+        );
 
         Ok(())
     }
@@ -330,15 +327,14 @@ mod tests {
         let cols = agg.emit()?;
         let batch = RecordBatch::try_new(test_schema(), cols)?;
         let actual = format!("{}", pretty_format_batches(&[batch])?);
-        let expected = r#"
+        assert_snapshot!(actual, @r#"
 +----------+--------------+
 | trace_id | timestamp_ms |
 +----------+--------------+
 | 2        | 2            |
 +----------+--------------+
         "#
-        .trim();
-        assert_eq!(actual, expected);
+        );
 
         Ok(())
     }
@@ -355,15 +351,14 @@ mod tests {
         let cols = agg.emit()?;
         let batch = RecordBatch::try_new(test_schema(), cols)?;
         let actual = format!("{}", pretty_format_batches(&[batch])?);
-        let expected = r#"
+        assert_snapshot!(actual, @r#"
 +----------+--------------+
 | trace_id | timestamp_ms |
 +----------+--------------+
 | 1        | 1            |
 +----------+--------------+
         "#
-        .trim();
-        assert_eq!(actual, expected);
+        );
 
         Ok(())
     }
@@ -380,15 +375,14 @@ mod tests {
         let cols = agg.emit()?;
         let batch = RecordBatch::try_new(test_schema(), cols)?;
         let actual = format!("{}", pretty_format_batches(&[batch])?);
-        let expected = r#"
+        assert_snapshot!(actual, @r#"
 +----------+--------------+
 | trace_id | timestamp_ms |
 +----------+--------------+
 | 1        | 2            |
 +----------+--------------+
         "#
-        .trim();
-        assert_eq!(actual, expected);
+        );
 
         Ok(())
     }
@@ -406,16 +400,15 @@ mod tests {
         let cols = agg.emit()?;
         let batch = RecordBatch::try_new(test_schema(), cols)?;
         let actual = format!("{}", pretty_format_batches(&[batch])?);
-        let expected = r#"
+        assert_snapshot!(actual, @r#"
 +----------+--------------+
 | trace_id | timestamp_ms |
 +----------+--------------+
 |          | 3            |
 | 1        | 1            |
 +----------+--------------+
         "#
-        .trim();
-        assert_eq!(actual, expected);
+        );
 
         Ok(())
     }

Diff for: datafusion/physical-plan/src/aggregates/mod.rs

@@ -645,8 +645,9 @@ mod tests {
     use crate::common;
     use crate::test::build_table_scan_i32;
 
-    use datafusion_common::{assert_batches_sorted_eq, assert_contains};
+    use datafusion_common::{assert_contains, test_util::batches_to_sort_string};
     use datafusion_execution::runtime_env::RuntimeEnvBuilder;
+    use insta::assert_snapshot;
 
     async fn join_collect(
         left: Arc<dyn ExecutionPlan>,
@@ -829,20 +830,19 @@ mod tests {
         let (columns, batches) = join_collect(left, right, task_ctx).await?;
 
         assert_eq!(columns, vec!["a1", "b1", "c1", "a2", "b2", "c2"]);
-        let expected = [
-            "+----+----+----+----+----+----+",
-            "| a1 | b1 | c1 | a2 | b2 | c2 |",
-            "+----+----+----+----+----+----+",
-            "| 1  | 4  | 7  | 10 | 12 | 14 |",
-            "| 1  | 4  | 7  | 11 | 13 | 15 |",
-            "| 2  | 5  | 8  | 10 | 12 | 14 |",
-            "| 2  | 5  | 8  | 11 | 13 | 15 |",
-            "| 3  | 6  | 9  | 10 | 12 | 14 |",
-            "| 3  | 6  | 9  | 11 | 13 | 15 |",
-            "+----+----+----+----+----+----+",
-        ];
-
-        assert_batches_sorted_eq!(expected, &batches);
+
+        assert_snapshot!(batches_to_sort_string(&batches), @r#"
+            +----+----+----+----+----+----+
+            | a1 | b1 | c1 | a2 | b2 | c2 |
+            +----+----+----+----+----+----+
+            | 1  | 4  | 7  | 10 | 12 | 14 |
+            | 1  | 4  | 7  | 11 | 13 | 15 |
+            | 2  | 5  | 8  | 10 | 12 | 14 |
+            | 2  | 5  | 8  | 11 | 13 | 15 |
+            | 3  | 6  | 9  | 10 | 12 | 14 |
+            | 3  | 6  | 9  | 11 | 13 | 15 |
+            +----+----+----+----+----+----+
+            "#);
 
         Ok(())
     }