Append end-of-input sentinel token to drop range checks

Add a sentinel WP_Parser_Token with id EMPTY_RULE_ID (0) to the end of
the token array. Real MySQL tokens never have id 0 (WHITESPACE, the only
token with id 0, is stripped by the lexer before tokens reach the
parser), so the sentinel cannot match any real terminal.

This lets the hot path drop the 'position < token_count' range check
everywhere it reads the current token id: the selector lookup at method
entry, the inline terminal match inside the branch loop, and the
post-branch INTO negative lookahead for selectStatement. Any read past
the last real token falls naturally into the nullable-fallback or
branch-miss handling.

Also drop a few dead locals ($token_count, $fragment_ids) that no
longer appear in the hot path after the change.

End-to-end parser benchmark:
  Before: ~28,700 QPS (avg)   After: ~29,800 QPS (+4%).

Author: JanJakes

packages/mysql-on-sqlite/src/parser/class-wp-parser.php

@@ -25,8 +25,14 @@ class WP_Parser {
 
 	public function __construct( WP_Parser_Grammar $grammar, array $tokens ) {
 		$this->grammar             = $grammar;
-		$this->tokens              = $tokens;
 		$this->token_count         = count( $tokens );
+		// Append an end-of-input sentinel token whose id is EMPTY_RULE_ID
+		// (0). The hot path can then read $tokens[$pos]->id unconditionally
+		// when $pos is the current cursor, because the sentinel naturally
+		// fails to match any real grammar terminal while feeding the
+		// nullable-fallback branch of the selector check.
+		$tokens[]                  = new WP_Parser_Token( WP_Parser_Grammar::EMPTY_RULE_ID, 0, 0, '' );
+		$this->tokens              = $tokens;
 		$this->position            = 0;
 		$this->rules               = $grammar->rules;
 		$this->rule_names          = $grammar->rule_names;
@@ -51,15 +57,14 @@ public function parse() {
 	 * round trip per consumed token.
 	 */
 	private function parse_recursive( $rule_id ) {
-		$tokens      = $this->tokens;
-		$token_count = $this->token_count;
-		$position    = $this->position;
+		$tokens   = $this->tokens;
+		$position = $this->position;
 
 		// Narrow the set of branches worth trying using the precomputed FIRST
 		// sets. When no entry exists for the current token but the rule is
 		// nullable, all candidate branches would match empty, so we return
 		// immediately without entering any branch.
-		$tid = $position < $token_count ? $tokens[ $position ]->id : WP_Parser_Grammar::EMPTY_RULE_ID;
+		$tid = $tokens[ $position ]->id;
 		if ( isset( $this->branches_for_token[ $rule_id ][ $tid ] ) ) {
 			$candidate_branches = $this->branches_for_token[ $rule_id ][ $tid ];
 		} elseif ( isset( $this->nullable_branches[ $rule_id ] ) ) {
@@ -70,9 +75,8 @@ private function parse_recursive( $rule_id ) {
 
 		$highest_terminal_id = $this->highest_terminal_id;
 		$branches            = $this->rules[ $rule_id ];
-		$fragment_ids        = $this->fragment_ids;
 		$rule_name           = $this->rule_names[ $rule_id ];
-		$is_fragment         = isset( $fragment_ids[ $rule_id ] );
+		$is_fragment         = isset( $this->fragment_ids[ $rule_id ] );
 		$is_select_statement = 'selectStatement' === $rule_name;
 		$branch_matches      = false;
 		$children            = array();
@@ -83,10 +87,10 @@ private function parse_recursive( $rule_id ) {
 			$branch_matches = true;
 			foreach ( $branch as $subrule_id ) {
 				if ( $subrule_id <= $highest_terminal_id ) {
-					if (
-						$this->position < $token_count
-						&& $tokens[ $this->position ]->id === $subrule_id
-					) {
+					// The sentinel at $tokens[$token_count] has id 0 so it
+					// cannot match any real terminal, making the range check
+					// unnecessary here.
+					if ( $tokens[ $this->position ]->id === $subrule_id ) {
 						$children[] = $tokens[ $this->position ];
 						++$this->position;
 						continue;
@@ -125,7 +129,6 @@ private function parse_recursive( $rule_id ) {
 			if (
 				$branch_matches
 				&& $is_select_statement
-				&& $this->position < $token_count
 				&& WP_MySQL_Lexer::INTO_SYMBOL === $tokens[ $this->position ]->id
 			) {
 				$branch_matches = false;