Minor: consolidate test data (#6217)

* Minor: consolidate test data directories

* move json data

* move parquet data

Author: alamb

datafusion/core/src/datasource/file_format/json.rs

@@ -256,7 +256,7 @@ mod tests {
         projection: Option<Vec<usize>>,
         limit: Option<usize>,
     ) -> Result<Arc<dyn ExecutionPlan>> {
-        let filename = "tests/jsons/2.json";
+        let filename = "tests/data/2.json";
         let format = JsonFormat::default();
         scan_format(state, &format, ".", filename, projection, limit).await
     }
@@ -266,7 +266,7 @@ mod tests {
         let session = SessionContext::new();
         let ctx = session.state();
         let store = Arc::new(LocalFileSystem::new()) as _;
-        let filename = "tests/jsons/schema_infer_limit.json";
+        let filename = "tests/data/schema_infer_limit.json";
         let format = JsonFormat::default().with_schema_infer_max_rec(Some(3));
 
         let file_schema = format

datafusion/core/src/physical_plan/file_format/csv.rs

@@ -785,7 +785,7 @@ mod tests {
         let options = CsvReadOptions::default()
             .schema_infer_max_records(2)
             .has_header(true);
-        let df = ctx.read_csv("tests/csv/corrupt.csv", options).await?;
+        let df = ctx.read_csv("tests/data/corrupt.csv", options).await?;
         let tmp_dir = TempDir::new()?;
         let out_dir = tmp_dir.as_ref().to_str().unwrap().to_string() + "/out";
         let e = df

datafusion/core/src/physical_plan/file_format/json.rs

@@ -309,7 +309,7 @@ mod tests {
 
     use super::*;
 
-    const TEST_DATA_BASE: &str = "tests/jsons";
+    const TEST_DATA_BASE: &str = "tests/data";
 
     async fn prepare_store(
         state: &SessionState,
@@ -707,7 +707,7 @@ mod tests {
         let options = CsvReadOptions::default()
             .schema_infer_max_records(2)
             .has_header(true);
-        let df = ctx.read_csv("tests/csv/corrupt.csv", options).await?;
+        let df = ctx.read_csv("tests/data/corrupt.csv", options).await?;
         let tmp_dir = TempDir::new()?;
         let out_dir = tmp_dir.as_ref().to_str().unwrap().to_string() + "/out";
         let e = df

datafusion/core/src/physical_plan/file_format/parquet.rs

@@ -940,7 +940,7 @@ mod tests {
         let options = CsvReadOptions::default()
             .schema_infer_max_records(2)
             .has_header(true);
-        let df = ctx.read_csv("tests/csv/corrupt.csv", options).await?;
+        let df = ctx.read_csv("tests/data/corrupt.csv", options).await?;
         let tmp_dir = TempDir::new()?;
         let out_dir = tmp_dir.as_ref().to_str().unwrap().to_string() + "/out";
         let e = df

datafusion/core/tests/jsons/1.json renamed to datafusion/core/tests/data/1.json

@@ -17,7 +17,7 @@
 
 use super::*;
 
-const TEST_DATA_BASE: &str = "tests/jsons";
+const TEST_DATA_BASE: &str = "tests/data";
 
 #[tokio::test]
 async fn json_query() {
@@ -92,7 +92,7 @@ async fn json_explain() {
             \n  CoalescePartitionsExec\
             \n    AggregateExec: mode=Partial, gby=[], aggr=[COUNT(UInt8(1))]\
             \n      RepartitionExec: partitioning=RoundRobinBatch(NUM_CORES), input_partitions=1\
-            \n        JsonExec: file_groups={1 group: [[WORKING_DIR/tests/jsons/2.json]]}, projection=[a]\n",
+            \n        JsonExec: file_groups={1 group: [[WORKING_DIR/tests/data/2.json]]}, projection=[a]\n",
         ],
     ];
     assert_eq!(expected, actual);

datafusion/core/tests/jsons/2.json renamed to datafusion/core/tests/data/2.json

@@ -25,7 +25,7 @@ use test_utils::{batches_to_vec, partitions_to_sorted_vec};
 #[tokio::test]
 async fn sort_with_lots_of_repetition_values() -> Result<()> {
     let ctx = SessionContext::new();
-    let filename = "tests/parquet/data/repeat_much.snappy.parquet";
+    let filename = "tests/data/repeat_much.snappy.parquet";
 
     ctx.register_parquet("rep", filename, ParquetReadOptions::default())
         .await?;

datafusion/core/tests/jsons/3.json renamed to datafusion/core/tests/data/3.json

@@ -151,7 +151,7 @@ async fn fixed_size_binary_columns() {
     let ctx = SessionContext::new();
     ctx.register_parquet(
         "t0",
-        "tests/parquet/data/test_binary.parquet",
+        "tests/data/test_binary.parquet",
         ParquetReadOptions::default(),
     )
     .await
@@ -170,7 +170,7 @@ async fn window_fn_timestamp_tz() {
     let ctx = SessionContext::new();
     ctx.register_parquet(
         "t0",
-        "tests/parquet/data/timestamp_with_tz.parquet",
+        "tests/data/timestamp_with_tz.parquet",
         ParquetReadOptions::default(),
     )
     .await