Online PHP editor | rfc for fKHcS

<?php class Trie { public $data = []; public $value = false; } $tokens = [ "apple" => "T_APPLE", "ape" => "T_APE", "cape" => "T_CAPE", ]; /* $root = new Trie; foreach ($tokens as $name => $value) { $node = $root; // For each character in the string for ($i = 0; $i < strlen($name); $i++) { $char = $name[$i]; if (!isset($node->data[$char])) { // If we don't have a child with that char // Create it $node->data[$char] = new Trie; } // Reset the node for the next character $node = $node->data[$char]; } // Finally, set the value on the final node $node->value = $value; } // "ape" is a valid token, so we expect T_APE var_dump($root->data['a']->data['p']->data['n']->value); // T_APE */ function lex($string, Trie $root) { $length = strlen($string); $i = 0; $tokens = []; $node = $root; $buffer = ''; // We want to iterate over the entire string. while ($i < $length) { // Get the current character $char = $string[$i]; if (isset($node->data[$char])) { // We have a valid next character $i++; // Save the character in the buffer $buffer .= $char; // Move to the next state $node = $node->data[$char]; } elseif ($node->value) { // We have a value and no valid next character // Emit the token $tokens[] = [$node->value, $buffer]; // Clear the buffer $buffer = ''; // Reset back to the root for the next token $node = $root; } else { // We can't continue parsing this node throw new Exception("Syntax error at offset $i"); } } if ($buffer !== '') { // We finished without flushing one token if ($node->value) { $tokens[] = [$node->value, $buffer]; } else { // Not a valid complete token throw new Exception("Syntax error at offset $i"); } } return $tokens; } foreach ($tokens as $token) { lex($token); }

3v4l.org

Active branches

Archived branches