Merge pull request MaterializeInc#2063 from sploiselle/select-set-ope…

…rations

docs: add set operations to SELECT

doc/user/sql/join.md

@@ -32,7 +32,7 @@ Field | Use
 ------|-----
 _select_pred_ | The predicating [`SELECT`](../select) clauses you want to use, e.g. `SELECT col_ref FROM table_ref...`. The _table_ref_ from the _select_pred_ is the left-hand table.
 **NATURAL** | Join table expressions on all columns with the same names in both tables. This is similar to the `USING` clause naming all identically named columns in both tables.
-_join_type_ | The type of `JOIN` you want to use (_`INNER` is implied default_).
+_join_type_ | The type of `JOIN` you want to use _(`INNER` is implied default)_.
 _table_ref_ | The table expression you want to join, i.e. the right-hand table.
 **USING (** _col_ref..._ **)** | If the join condition does not require table-level qualification (i.e. joining tables on columns with the same name), the columns to join the tables on. For example, `USING (customer_id)`.
 **ON** _expression_ | The condition on which to join the tables. For example `ON purchase.customer_id = customer.id`.

doc/user/sql/select.md

@@ -31,11 +31,11 @@ Scenario | `SELECT` dataflow behavior
 
 ## Syntax
 
-{{< diagram "select.html" >}}
+{{< diagram "select_stmt.html" >}}
 
 Field | Use
 ------|-----
-**ALL** | Return all rows from query (_implied default_).
+**ALL** | Return all rows from query _(implied default)_.
 **DISTINCT** | Return only distinct values from query.
 **DISTINCT ON (** _col&lowbar;ref..._ **)**  | Return only the first row with a distinct value for _col&lowbar;ref_.
 _target&lowbar;elem_ | Return identified columns or functions.
@@ -47,6 +47,9 @@ _join&lowbar;expr_ | A join expression; for more details, see our [`JOIN` docume
 **ORDER BY** _col&lowbar;ref_ ... | Order results in either **ASC** or **DESC** order (_**ASC** is implied default_).<br/><br>**ORDER BY** does not work when creating views, but does work when reading from views.
 **LIMIT** | Limit the number of returned results to _expr_.
 **OFFSET** | Skip the first _expr_ number of rows.
+**UNION** | Records present in `select_stmt` or `another_select_stmt`.<br/><br/>**DISTINCT** returns only unique rows from these results _(implied default)_.<br/><br/>With **ALL** specified, each record occurs a number of times equal to the sum of the times it occurs in each input statement.
+**INTERSECT** | Records present in both `select_stmt` and `another_select_stmt`.<br/><br/>**DISTINCT** returns only unique rows from these results _(implied default)_.<br/><br/>With **ALL** specified, each record occurs a number of times equal to the lesser of the times it occurs in each input statement.
+**EXCEPT** | Records present in `select_stmt` but not in `another_select_stmt`.<br/><br/>**DISTINCT** returns only unique rows from these results _(implied default)_.<br/><br/>With **ALL** specified, each record occurs a number of times equal to the times it occurs in `select_stmt` less the times it occurs in `another_select_stmt`, or not at all if the former is greater than latter.
 
 ## Details
 

src/sql-parser/src/parser.rs

@@ -2446,10 +2446,22 @@ impl Parser {
                 break;
             }
             self.next_token(); // skip past the set operator
+
+            let all = self.parse_keyword("ALL");
+            let all_range = self.peek_prev_range();
+            let distinct = self.parse_keyword("DISTINCT");
+            let distinct_range = self.peek_prev_range();
+            if all && distinct {
+                return parser_err!(
+                    self,
+                    all_range.start..distinct_range.end,
+                    "Cannot specify both ALL and DISTINCT in set operation"
+                );
+            }
             expr = SetExpr::SetOperation {
                 left: Box::new(expr),
                 op: op.unwrap(),
-                all: self.parse_keyword("ALL"),
+                all,
                 right: Box::new(self.parse_query_body(next_precedence)?),
             };
         }

www/assets/sass/_docs_body.scss

@@ -201,6 +201,10 @@ main {
     padding-bottom: 0.9rem;
   }
 
+  td {
+    vertical-align: top;
+  }
+
   svg {
     margin-top: 2rem;
   }

www/layouts/partials/sql-grammar/bnf/select_stmt.bnf

@@ -1,4 +1,7 @@
 select_stmt ::=
-	'SELECT' ( 'ALL' )? ( ( target_elem ) ( ( ',' target_elem ) )* ) ( 'FROM' ( ( table_expr ) ( ( ',' table_expr ) )* )  )? ( ( 'WHERE' expr ) )? ( 'GROUP' 'BY' ( ( col_ref ) ( ( ',' col_ref ) )* ) )? ( 'HAVING' expr )? ( 'ORDER' 'BY' ( ( col_ref ( 'ASC' | 'DESC' )? ) ( ( ',' col_ref ( 'ASC' | 'DESC' )? ) )* ) )? ( 'LIMIT' expr )? ( 'OFFSET' expr )?
-	| 'SELECT' ( 'DISTINCT' ) ( ( target_elem ) ( ( ',' target_elem ) )* ) ( 'FROM' ( ( table_expr ) ( ( ',' table_expr ) )* )  )? ( ( 'WHERE' expr ) )? ( 'GROUP' 'BY' ( ( col_ref ) ( ( ',' col_ref ) )* ) )? ( 'HAVING' expr )? ( 'ORDER' 'BY' ( ( col_ref ( 'ASC' | 'DESC' )? ) ( ( ',' col_ref ( 'ASC' | 'DESC' )? ) )* ) )? ( 'LIMIT' expr )? ( 'OFFSET' expr )?
-	| 'SELECT' ( 'DISTINCT' 'ON' '(' ( ( col_ref ) ( ( ',' col_ref ) )* ) ')' ) ( ( target_elem ) ( ( ',' target_elem ) )* ) ( 'FROM' ( ( table_expr ) ( ( ',' table_expr ) )* )  )? ( ( 'WHERE' expr ) )? ( 'GROUP' 'BY' ( ( col_ref ) ( ( ',' col_ref ) )* ) )? ( 'HAVING' expr )? ( 'ORDER' 'BY' ( ( col_ref ( 'ASC' | 'DESC' )? ) ( ( ',' col_ref ( 'ASC' | 'DESC' )? ) )* ) )?  ( 'LIMIT' expr )? ( 'OFFSET' expr )?
+    'SELECT'  (
+        ( 'ALL' )? ( ( target_elem ) ( ( ',' target_elem ) )* ) ( 'FROM' ( ( table_expr ) ( ( ',' table_expr ) )* )  )? join_expr? ( ( 'WHERE' expr ) )? ( 'GROUP' 'BY' ( ( col_ref ) ( ( ',' col_ref ) )* ) )? ( 'HAVING' expr )? ( 'ORDER' 'BY' ( ( col_ref ( 'ASC' | 'DESC' )? ) ( ( ',' col_ref ( 'ASC' | 'DESC' )? ) )* ) )? ( 'LIMIT' expr )? ( 'OFFSET' expr )?
+	    | ( 'DISTINCT' ) ( ( target_elem ) ( ( ',' target_elem ) )* ) ( 'FROM' ( ( table_expr ) ( ( ',' table_expr ) )* )  )? join_expr? ( ( 'WHERE' expr ) )? ( 'GROUP' 'BY' ( ( col_ref ) ( ( ',' col_ref ) )* ) )? ( 'HAVING' expr )? ( 'ORDER' 'BY' ( ( col_ref ( 'ASC' | 'DESC' )? ) ( ( ',' col_ref ( 'ASC' | 'DESC' )? ) )* ) )? ( 'LIMIT' expr )? ( 'OFFSET' expr )?
+        | ( 'DISTINCT' 'ON' '(' ( ( col_ref ) ( ( ',' col_ref ) )* ) ')' ) ( ( target_elem ) ( ( ',' target_elem ) )* ) ( 'FROM' ( ( table_expr ) ( ( ',' table_expr ) )* )  )? join_expr? ( ( 'WHERE' expr ) )? ( 'GROUP' 'BY' ( ( col_ref ) ( ( ',' col_ref ) )* ) )? ( 'HAVING' expr )? ( 'ORDER' 'BY' ( ( col_ref ( 'ASC' | 'DESC' )? ) ( ( ',' col_ref ( 'ASC' | 'DESC' )? ) )* ) )?  ( 'LIMIT' expr )? ( 'OFFSET' expr )?
+    )
+    ( ( 'UNION' | 'INTERSECT' | 'EXCEPT' ) ('ALL' | 'DISTINCT')? another_select_stmt)?