class InlineHTMLSniff

class InlineHTMLSniff implements Sniff {
    
    /**
     * List of supported BOM definitions.
     *
     * Use encoding names as keys and hex BOM representations as values.
     *
     * @var array
     */
    protected $bomDefinitions = [
        'UTF-8' => 'efbbbf',
        'UTF-16 (BE)' => 'feff',
        'UTF-16 (LE)' => 'fffe',
    ];
    
    /**
     * Returns an array of tokens this test wants to listen for.
     *
     * @return array<int|string>
     */
    public function register() {
        return [
            T_INLINE_HTML,
        ];
    }
    
    //end register()
    
    /**
     * Processes this test, when one of its tokens is encountered.
     *
     * @param \PHP_CodeSniffer\Files\File $phpcsFile The file being scanned.
     * @param int                         $stackPtr  The position of the current token in
     *                                               the stack passed in $tokens.
     *
     * @return int|void
     */
    public function process(File $phpcsFile, $stackPtr) {
        // Allow a byte-order mark.
        $tokens = $phpcsFile->getTokens();
        foreach ($this->bomDefinitions as $expectedBomHex) {
            $bomByteLength = strlen($expectedBomHex) / 2;
            $htmlBomHex = bin2hex(substr($tokens[0]['content'], 0, $bomByteLength));
            if ($htmlBomHex === $expectedBomHex && strlen($tokens[0]['content']) === $bomByteLength) {
                return;
            }
        }
        // Ignore shebang lines.
        $tokens = $phpcsFile->getTokens();
        if (substr($tokens[$stackPtr]['content'], 0, 2) === '#!') {
            return;
        }
        $error = 'PHP files must only contain PHP code';
        $phpcsFile->addError($error, $stackPtr, 'Found');
        return $phpcsFile->numTokens;
    }
    
    //end process()

}