opt: Inline constant values

Inline constants in expressions like:

  SELECT x, x+1 FROM (VALUES (1)) AS t(x) ;

with the new inlining rules, this becomes:

  VALUES (1, 2)

The new inlining rules are useful for mutation expressions (e.g. UPDATE),
which can nest multiple Project and Values expressions that often use
constant values. For example:

  CREATE TABLE ab (a INT PRIMARY KEY, b INT AS (a + 1) STORED);
  UPDATE ab SET a=1

This now gets mapped by the optimizer to this internal equivalent:

  UPDATE ab SET a=1, b=2

Release note: None

pkg/sql/opt/exec/execbuilder/testdata/explain

@@ -601,11 +601,9 @@ values  ·              ·                (a int)  ·
 query TTTTT
 EXPLAIN (TYPES) SELECT x[1] FROM (SELECT ARRAY[1,2,3] AS x)
 ----
-render       ·              ·                                           (x int)    ·
- │           render 0       ((x)[int[]][(1)[int]])[int]                 ·          ·
- └── values  ·              ·                                           (x int[])  ·
-·            size           1 column, 1 row                             ·          ·
-·            row 0, expr 0  (ARRAY[(1)[int],(2)[int],(3)[int]])[int[]]  ·          ·
+values  ·              ·                (x int)  ·
+·       size           1 column, 1 row  ·        ·
+·       row 0, expr 0  (1)[int]         ·        ·
 
 query T
 EXPLAIN (OPT) SELECT 1 AS r

pkg/sql/opt/exec/execbuilder/testdata/orderby

@@ -730,17 +730,24 @@ render     ·         ·       (k)     ·
 #
 
 query TTTTT
-EXPLAIN (VERBOSE) SELECT k FROM kv JOIN (VALUES (1,2)) AS z(a,b) ON kv.k = z.a ORDER BY INDEX kv@foo
-----
-render            ·              ·                (k)           ·
- │                render 0       k                ·             ·
- └── lookup-join  ·              ·                (column1, k)  ·
-      │           type           inner            ·             ·
-      ├── values  ·              ·                (column1)     ·
-      │           size           1 column, 1 row  ·             ·
-      │           row 0, expr 0  1                ·             ·
-      └── scan    ·              ·                (k)           ·
-·                 table          kv@primary       ·             ·
+EXPLAIN (VERBOSE)
+SELECT k FROM kv JOIN (VALUES (1,2), (3,4)) AS z(a,b) ON kv.k = z.a ORDER BY INDEX kv@foo
+----
+render                      ·              ·                 (k)              ·
+ │                          render 0       k                 ·                ·
+ └── sort                   ·              ·                 (k, v)           -v,+k
+      │                     order          -v,+k             ·                ·
+      └── render            ·              ·                 (k, v)           ·
+           │                render 0       k                 ·                ·
+           │                render 1       v                 ·                ·
+           └── lookup-join  ·              ·                 (column1, k, v)  ·
+                │           type           inner             ·                ·
+                ├── values  ·              ·                 (column1)        ·
+                │           size           1 column, 2 rows  ·                ·
+                │           row 0, expr 0  1                 ·                ·
+                │           row 1, expr 0  3                 ·                ·
+                └── scan    ·              ·                 (k, v)           ·
+·                           table          kv@primary        ·                ·
 
 query TTTTT
 EXPLAIN (VERBOSE) SELECT k FROM kv a NATURAL JOIN kv ORDER BY INDEX kv@foo

pkg/sql/opt/memo/testdata/logprops/join

@@ -838,21 +838,19 @@ semi-join-apply
 
 # Calculate anti-join cardinality when left side has non-zero cardinality.
 opt
-SELECT * FROM (SELECT * FROM (VALUES (1))) WHERE NOT EXISTS(SELECT * FROM uv WHERE u=column1)
+SELECT * FROM (SELECT * FROM (VALUES (1), (2))) WHERE NOT EXISTS(SELECT * FROM uv WHERE u=column1)
 ----
 anti-join
  ├── columns: column1:1(int)
- ├── cardinality: [0 - 1]
- ├── key: ()
- ├── fd: ()-->(1)
+ ├── cardinality: [0 - 2]
  ├── values
  │    ├── columns: column1:1(int)
- │    ├── cardinality: [1 - 1]
- │    ├── key: ()
- │    ├── fd: ()-->(1)
+ │    ├── cardinality: [2 - 2]
  │    ├── prune: (1)
+ │    ├── tuple [type=tuple{int}]
+ │    │    └── const: 1 [type=int]
  │    └── tuple [type=tuple{int}]
- │         └── const: 1 [type=int]
+ │         └── const: 2 [type=int]
  ├── scan uv
  │    ├── columns: u:2(int) v:3(int!null)
  │    └── prune: (2,3)

pkg/sql/opt/norm/custom_funcs.go

@@ -277,6 +277,11 @@ func (c *CustomFuncs) CanHaveZeroRows(input memo.RelExpr) bool {
 	return input.Relational().Cardinality.CanBeZero()
 }
 
+// ColsAreEmpty returns true if the column set is empty.
+func (c *CustomFuncs) ColsAreEmpty(cols opt.ColSet) bool {
+	return cols.Empty()
+}
+
 // ColsAreSubset returns true if the left columns are a subset of the right
 // columns.
 func (c *CustomFuncs) ColsAreSubset(left, right opt.ColSet) bool {
@@ -288,6 +293,12 @@ func (c *CustomFuncs) ColsAreEqual(left, right opt.ColSet) bool {
 	return left.Equals(right)
 }
 
+// ColsIntersect returns true if at least one column appears in both the left
+// and right sets.
+func (c *CustomFuncs) ColsIntersect(left, right opt.ColSet) bool {
+	return left.Intersects(right)
+}
+
 // UnionCols returns the union of the left and right column sets.
 func (c *CustomFuncs) UnionCols(left, right opt.ColSet) opt.ColSet {
 	return left.Union(right)

pkg/sql/opt/norm/inline.go

@@ -19,6 +19,100 @@ import (
 	"github.com/cockroachdb/cockroach/pkg/sql/opt/memo"
 )
 
+// FindInlinableConstants returns the set of input columns that are synthesized
+// constant value expressions: ConstOp, TrueOp, FalseOp, or NullOp. Constant
+// value expressions can often be inlined into referencing expressions. Only
+// Project and Values operators synthesize constant value expressions.
+func (c *CustomFuncs) FindInlinableConstants(input memo.RelExpr) opt.ColSet {
+	var cols opt.ColSet
+	if project, ok := input.(*memo.ProjectExpr); ok {
+		for i := range project.Projections {
+			item := &project.Projections[i]
+			if opt.IsConstValueOp(item.Element) {
+				cols.Add(int(item.Col))
+			}
+		}
+	} else if values, ok := input.(*memo.ValuesExpr); ok && len(values.Rows) == 1 {
+		tup := values.Rows[0].(*memo.TupleExpr)
+		for i, scalar := range tup.Elems {
+			if opt.IsConstValueOp(scalar) {
+				cols.Add(int(values.Cols[i]))
+			}
+		}
+	}
+	return cols
+}
+
+// InlineProjectionConstants recursively searches each projection expression and
+// replaces any references to input columns that are constant. It returns a new
+// Projections list containing the replaced expressions.
+func (c *CustomFuncs) InlineProjectionConstants(
+	projections memo.ProjectionsExpr, input memo.RelExpr, constCols opt.ColSet,
+) memo.ProjectionsExpr {
+	newProjections := make(memo.ProjectionsExpr, len(projections))
+	for i := range projections {
+		item := &projections[i]
+		newProjections[i].Col = item.Col
+		newProjections[i].Element = c.inlineConstants(item.Element, input, constCols).(opt.ScalarExpr)
+	}
+	return newProjections
+}
+
+// InlineFilterConstants recursively searches each filter expression and
+// replaces any references to input columns that are constant. It returns a new
+// Filters list containing the replaced expressions.
+func (c *CustomFuncs) InlineFilterConstants(
+	filters memo.FiltersExpr, input memo.RelExpr, constCols opt.ColSet,
+) memo.FiltersExpr {
+	newFilters := make(memo.FiltersExpr, len(filters))
+	for i := range filters {
+		item := &filters[i]
+		newFilters[i].Condition = c.inlineConstants(item.Condition, input, constCols).(opt.ScalarExpr)
+	}
+	return newFilters
+}
+
+// inlineConstants recursively searches the given expression and replaces any
+// references to input columns that are constant. It returns the replaced
+// expression.
+func (c *CustomFuncs) inlineConstants(
+	e opt.Expr, input memo.RelExpr, constCols opt.ColSet,
+) opt.Expr {
+	var replace ReconstructFunc
+	replace = func(e opt.Expr) opt.Expr {
+		switch t := e.(type) {
+		case *memo.VariableExpr:
+			if constCols.Contains(int(t.Col)) {
+				return c.extractColumn(input, t.Col)
+			}
+			return t
+		}
+		return c.f.Reconstruct(e, replace)
+	}
+	return replace(e)
+}
+
+// extractColumn searches a Project or Values input expression for the column
+// having the given id. It returns the expression for that column.
+func (c *CustomFuncs) extractColumn(input memo.RelExpr, col opt.ColumnID) opt.ScalarExpr {
+	if project, ok := input.(*memo.ProjectExpr); ok {
+		for i := range project.Projections {
+			item := &project.Projections[i]
+			if item.Col == col {
+				return item.Element
+			}
+		}
+	} else if values, ok := input.(*memo.ValuesExpr); ok && len(values.Rows) == 1 {
+		tup := values.Rows[0].(*memo.TupleExpr)
+		for i, scalar := range tup.Elems {
+			if values.Cols[i] == col {
+				return scalar
+			}
+		}
+	}
+	panic("could not find column to extract")
+}
+
 // HasDuplicateRefs returns true if the target projection expressions or
 // passthrough columns reference any outer column more than one time, or if the
 // projection expressions contain a correlated subquery. For example:

pkg/sql/opt/norm/rules/inline.opt

@@ -8,13 +8,82 @@
 #
 #   SELECT (x+1)+1 FROM a
 #
-# Inlining variables can result in the complete elimination of Project
-# operators, or at least in the ability to more freely reorder them within the
-# larger relational expression tree. This allows pushing filters further down
-# the tree, as well as with pulling them up in the decorrelation case.
+# Inlining variables can result in the simplification or even complete
+# elimination of operators, or at least in the ability to more freely reorder
+# them within the larger relational expression tree. This allows pushing filters
+# further down the tree, as well as with pulling them up in the decorrelation
+# case.
 # =============================================================================
 
 
+# InlineProjectConstants finds variable references in Projections expressions
+# that refer to constant input values, and then inlines those constant values
+# in place of the corresponding variable references. This sometimes allows
+# further simplifications such as constant folding or Project merging.
+[InlineProjectConstants, Normalize]
+(Project
+    $input:* & ^(ColsAreEmpty $constCols:(FindInlinableConstants $input))
+    $projections:[ ... $item:* & (ColsIntersect (OuterCols $item) $constCols) ... ]
+    $passthrough:*
+)
+=>
+(Project
+    $input
+    (InlineProjectionConstants $projections $input $constCols)
+    $passthrough
+)
+
+# InlineSelectConstants finds variable references in Filters expressions that
+# refer to constant input values, and then inlines those constant values in
+# place of the corresponding variable references. This sometimes allows further
+# simplifications such as constant folding or generation of constrained scans.
+[InlineSelectConstants, Normalize]
+(Select
+    $input:* & ^(ColsAreEmpty $constCols:(FindInlinableConstants $input))
+    $filters:[ ... $item:* & (ColsIntersect (OuterCols $item) $constCols) ... ]
+)
+=>
+(Select
+    $input
+    (InlineFilterConstants $filters $input $constCols)
+)
+
+# InlineJoinConstantsLeft finds variable references in a join condition that
+# refers to constant values projected by the left input. It then inlines those
+# constant values in place of the corresponding variable references. This
+# sometimes allows further simplifications such as constant folding or filter
+# pushdown.
+[InlineJoinConstantsLeft, Normalize]
+(Join
+    $left:* & ^(ColsAreEmpty $constCols:(FindInlinableConstants $left))
+    $right:*
+    $on:[ ... $item:* & (ColsIntersect (OuterCols $item) $constCols) ... ]
+)
+=>
+((OpName)
+    $left
+    $right
+    (InlineFilterConstants $on $left $constCols)
+)
+
+# InlineJoinConstantsRight finds variable references in a join condition that
+# refers to constant values projected by the right input. It then inlines those
+# constant values in place of the corresponding variable references. This
+# sometimes allows further simplifications such as constant folding or filter
+# pushdown.
+[InlineJoinConstantsRight, Normalize]
+(Join
+    $left:*
+    $right:* & ^(ColsAreEmpty $constCols:(FindInlinableConstants $right))
+    $on:[ ... $item:* & (ColsIntersect (OuterCols $item) $constCols) ... ]
+)
+=>
+((OpName)
+    $left
+    $right
+    (InlineFilterConstants $on $right $constCols)
+)
+
 # PushSelectIntoInlinableProject pushes the Select operator into a Project, even
 # though the filter references it. This is made possible by inlining the
 # references to projected columns so that the Select becomes independent of the