No longer write Parquet column metadata after column chunks *and* in the footer

parquet/src/column/writer/mod.rs

@@ -298,7 +298,7 @@ impl<'a, E: ColumnValueEncoder> GenericColumnWriter<'a, E> {
                 num_page_nulls: 0,
             },
             column_metrics: ColumnMetrics::<E::T>::new(),
-            column_index_builder: ColumnIndexBuilder::new(),
+            column_index_builder,
             offset_index_builder: OffsetIndexBuilder::new(),
             encodings,
             data_page_boundary_ascending: true,

parquet/src/file/metadata/mod.rs

@@ -72,14 +72,16 @@ use crate::schema::types::{
 /// [PageIndex documentation]: https://github.com/apache/parquet-format/blob/master/PageIndex.md
 pub type ParquetColumnIndex = Vec<Vec<Index>>;
 
-/// [`OffsetIndexMetaData`] for each row group of each column.
+/// [`OffsetIndexMetaData`] for each data page of each row group of each column
+///
+/// This structure is the parsed representation of the [`OffsetIndex`] from the
+/// Parquet file footer, as described in the Parquet [PageIndex documentation].
 ///
 /// `offset_index[row_group_number][column_number]` holds
 /// the [`OffsetIndexMetaData`] corresponding to column
 /// `column_number`of row group `row_group_number`.
 ///
-/// For example `offset_index[2][3]` holds the [`OffsetIndexMetaData`] for
-/// the fourth column in the third row group of the parquet file.
+/// [PageIndex documentation]: https://github.com/apache/parquet-format/blob/master/PageIndex.md
 pub type ParquetOffsetIndex = Vec<Vec<OffsetIndexMetaData>>;
 
 /// Parsed metadata for a single Parquet file
@@ -1374,8 +1376,29 @@ mod tests {
             schema_descr.clone(),
             column_orders,
         );
-        let parquet_meta = ParquetMetaData::new(file_metadata.clone(), row_group_meta.clone());
-        let base_expected_size = 1352;
+
+        // Now, add in Exact Statistics
+        let columns_with_stats = schema_descr
+            .columns()
+            .iter()
+            .map(|column_descr| {
+                ColumnChunkMetaData::builder(column_descr.clone())
+                    .set_statistics(Statistics::new::<i32>(Some(0), Some(100), None, 0, false))
+                    .build()
+            })
+            .collect::<Result<Vec<_>>>()
+            .unwrap();
+
+        let row_group_meta_with_stats = RowGroupMetaData::builder(schema_descr)
+            .set_num_rows(1000)
+            .set_column_metadata(columns_with_stats)
+            .build()
+            .unwrap();
+        let row_group_meta_with_stats = vec![row_group_meta_with_stats];
+
+        let parquet_meta = ParquetMetaData::new(file_metadata.clone(), row_group_meta_with_stats);
+        let base_expected_size = 2088;
+
         assert_eq!(parquet_meta.memory_size(), base_expected_size);
 
         let mut column_index = ColumnIndexBuilder::new();