[arrow-datafusion] branch master updated: collect table stats by default for listing table (#1347)

[houqp]

This is an automated email from the ASF dual-hosted git repository.

houqp pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/arrow-datafusion.git


The following commit(s) were added to refs/heads/master by this push:
     new cf637e7  collect table stats by default for listing table (#1347)
cf637e7 is described below

commit cf637e7611a88cf89de1ba7fb93d21c687d30be6
Author: QP Hou <q...@scribd.com>
AuthorDate: Mon Nov 22 16:10:27 2021 -0800

    collect table stats by default for listing table (#1347)
    
    * collect table stats by default for listing table
    * add test
---
 datafusion/src/datasource/listing/table.rs | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/datafusion/src/datasource/listing/table.rs 
b/datafusion/src/datasource/listing/table.rs
index aadc340..22e3f75 100644
--- a/datafusion/src/datasource/listing/table.rs
+++ b/datafusion/src/datasource/listing/table.rs
@@ -76,7 +76,7 @@ impl ListingOptions {
             file_extension: String::new(),
             format,
             table_partition_cols: vec![],
-            collect_stat: false,
+            collect_stat: true,
             target_partitions: 1,
         }
     }
@@ -304,6 +304,23 @@ mod tests {
     }
 
     #[tokio::test]
+    async fn load_table_stats_by_default() -> Result<()> {
+        let testdata = crate::test_util::parquet_test_data();
+        let filename = format!("{}/{}", testdata, "alltypes_plain.parquet");
+        let opt = ListingOptions::new(Arc::new(ParquetFormat::default()));
+        let schema = opt
+            .infer_schema(Arc::new(LocalFileSystem {}), &filename)
+            .await?;
+        let table =
+            ListingTable::new(Arc::new(LocalFileSystem {}), filename, schema, 
opt);
+        let exec = table.scan(&None, 1024, &[], None).await?;
+        assert_eq!(exec.statistics().num_rows, Some(8));
+        assert_eq!(exec.statistics().total_byte_size, Some(671));
+
+        Ok(())
+    }
+
+    #[tokio::test]
     async fn read_empty_table() -> Result<()> {
         let store = TestObjectStore::new_arc(&[("table/p1=v1/file.avro", 
100)]);