/** * Parse a reference */ public function evaluate(Lexer $lexer) { switch (true) { case $lexer->isNextToken(Lexer::T_SHORT_P): throw new ParserException('Property \\p (Unicode Property) not supported use \\x to specify unicode character or range'); break; case $lexer->isNextToken(Lexer::T_SHORT_UNICODE_X): $lexer->moveNext(); if ($lexer->lookahead['value'] !== '{') { throw new ParserException('Expecting character { after \\X none found'); } $tokens = array(); while ($lexer->moveNext() && $lexer->lookahead['value'] !== '}') { # check if we nested eg.{ddd{d} if ($lexer->lookahead['value'] === '{') { throw new ParserException('Nesting hex value ranges is not allowed'); } if ($lexer->lookahead['value'] !== " " && ctype_xdigit($lexer->lookahead['value']) === false) { throw new ParserException(sprintf('Character %s is not a hexdeciaml digit', $lexer->lookahead['value'])); } $tokens[] = $lexer->lookahead['value']; } # check that current lookahead is a closing character as it's possible to iterate to end of string (i.e. lookahead === null) if ($lexer->lookahead['value'] !== '}') { throw new ParserException('Closing quantifier token `}` not found'); } if (count($tokens) === 0) { throw new ParserException('No hex number found inside the range'); } $number = trim(implode('', $tokens)); return Utf8::chr(hexdec($number)); break; case $lexer->isNextToken(Lexer::T_SHORT_X): // only allow another 2 hex characters $glimpse = $lexer->glimpse(); if ($glimpse['value'] === '{') { throw new ParserException('Braces not supported here'); } $tokens = array(); $count = 2; while ($count > 0 && $lexer->moveNext()) { $tokens[] = $lexer->lookahead['value']; --$count; } $value = trim(implode('', $tokens)); return Utf8::chr(hexdec($value)); break; default: throw new ParserException('No Unicode expression to evaluate'); } }
/** * Parse the current token for new Quantifiers * * @access public * @return ReverseRegex\Generator\Scope a new head * @param ReverseRegex\Generator\Scope $head * @param ReverseRegex\Generator\Scope $set * @param ReverseRegex\Lexer $lexer */ public function parse(Scope $head, Scope $set, Lexer $lexer) { if ($lexer->lookahead['type'] !== Lexer::T_SET_OPEN) { throw new ParserException('Opening character set token not found'); } $peek = $lexer->glimpse(); if ($peek['type'] === Lexer::T_SET_NEGATED) { throw new ParserException('Negated Character Set ranges not supported at this time'); } $normal_lexer = new Lexer($this->normalize($head, $set, $lexer)); while ($normal_lexer->moveNext() && !$normal_lexer->isNextToken(Lexer::T_SET_CLOSE)) { $glimpse = $normal_lexer->glimpse(); if ($glimpse['type'] === Lexer::T_SET_RANGE) { continue; //value be included in range when `-` character is passed } switch (true) { case $normal_lexer->isNextToken(Lexer::T_SET_RANGE): $range_start = $normal_lexer->token['value']; $normal_lexer->moveNext(); if ($normal_lexer->isNextToken(Lexer::T_ESCAPE_CHAR)) { $normal_lexer->moveNext(); } $range_end = $normal_lexer->lookahead['value']; $this->fillRange($head, $range_start, $range_end); break; case $normal_lexer->isNextToken(Lexer::T_LITERAL_NUMERIC) || $normal_lexer->isNextToken(Lexer::T_LITERAL_CHAR): $index = (int) Utf8::ord($normal_lexer->lookahead['value']); $head->setLiteral($index, $normal_lexer->lookahead['value']); break; default: # ignore } } $head->getLiterals()->sort(); return $head; }