adding more docs

Author: Jiayu Liu

datafusion/src/logical_plan/expr.rs

@@ -191,6 +191,7 @@ pub enum Expr {
         /// Whether this is a DISTINCT aggregation or not
         distinct: bool,
     },
+    /// Represents the call of a window function with arguments.
     WindowFunction {
         /// Name of the function
         fun: windows::WindowFunction,

datafusion/src/optimizer/hash_build_probe_order.rs

@@ -54,7 +54,8 @@ fn get_num_rows(logical_plan: &LogicalPlan) -> Option<usize> {
             let num_rows_input = get_num_rows(input);
             num_rows_input.map(|rows| std::cmp::min(*limit, rows))
         }
-        LogicalPlan::Window { .. } | LogicalPlan::Aggregate { .. } => {
+        LogicalPlan::Window { input, .. } => get_num_rows(input),
+        LogicalPlan::Aggregate { .. } => {
             // we cannot yet predict how many rows will be produced by an aggregate or a window because
             // we do not know the cardinality of the grouping keys
             None

datafusion/src/physical_plan/windows.rs

@@ -15,17 +15,19 @@
 // specific language governing permissions and limitations
 // under the License.
 
+//! Window functions provide the ability to perform calculations across
+//! sets of rows that are related to the current query row.
+//!
+//! see also https://www.postgresql.org/docs/current/functions-window.html
+
 use super::expressions::{avg_return_type, sum_return_type};
 use super::{functions::Signature, type_coercion::data_types};
 use crate::error::{DataFusionError, Result};
 use crate::physical_plan::aggregates::AggregateFunction;
-use arrow::datatypes::{DataType, Schema, TimeUnit};
+use arrow::datatypes::{DataType, TimeUnit};
 use std::{fmt, str::FromStr};
 
-/// Window functions provide the ability to perform calculations across sets of rows that are
-/// related to the current query row
-///
-/// see also https://www.postgresql.org/docs/current/functions-window.html
+/// WindowFunction
 #[derive(Debug, Clone, PartialEq, Eq)]
 pub enum WindowFunction {
     /// window function that leverages an aggregate function
@@ -60,14 +62,22 @@ impl fmt::Display for WindowFunction {
     }
 }
 
+/// An aggregate function that is part of a built-in window function
 #[derive(Debug, Clone, PartialEq, Eq)]
 pub enum BuiltInWindowFunction {
+    /// row number
     RowNumber,
+    /// rank
     Rank,
+    /// dense rank
     DenseRank,
+    /// lag
     Lag,
+    /// lead
     Lead,
+    /// first value
     FirstValue,
+    /// last value
     LastValue,
 }