|
| 1 | +<?php |
| 2 | + |
| 3 | +declare(strict_types=1); |
| 4 | + |
| 5 | +namespace SimpleSAML\XML\Utils; |
| 6 | + |
| 7 | +use SimpleSAML\XML\Exception\RuntimeException; |
| 8 | + |
| 9 | +use function in_array; |
| 10 | +use function preg_match_all; |
| 11 | +use function preg_replace; |
| 12 | + |
| 13 | +/** |
| 14 | + * XPathFilter helper functions for the XML library. |
| 15 | + * |
| 16 | + * @package simplesamlphp/xml-common |
| 17 | + */ |
| 18 | +class XPathFilter |
| 19 | +{ |
| 20 | + /** |
| 21 | + * Remove the content from all single or double-quoted strings in $input, leaving only quotes. |
| 22 | + * |
| 23 | + * @param string $input |
| 24 | + * @return string |
| 25 | + * @throws \SimpleSAML\XML\Exception\RuntimeException |
| 26 | + */ |
| 27 | + public static function removeStringContents(string $input): string |
| 28 | + { |
| 29 | + /** |
| 30 | + * This regex should not be vulnerable to a ReDOS, because it uses possessive quantifiers |
| 31 | + * that prevent backtracking. |
| 32 | + * |
| 33 | + * @see https://www.owasp.org/index.php/Regular_expression_Denial_of_Service_-_ReDoS |
| 34 | + * |
| 35 | + * Use possessive quantifiers (i.e. *+ and ++ instead of * and + respectively) to prevent backtracking. |
| 36 | + * |
| 37 | + * '/(["\'])(?:(?!\1).)*+\1/' |
| 38 | + * (["\']) # Match a single or double quote and capture it in group 1 |
| 39 | + * (?: # Start a non-capturing group |
| 40 | + * (?! # Negative lookahead |
| 41 | + * \1 # Match the same quote as in group 1 |
| 42 | + * ) # End of negative lookahead |
| 43 | + * . # Match any character (that is not a quote, because of the negative lookahead) |
| 44 | + * )*+ # Repeat the non-capturing group zero or more times, possessively |
| 45 | + * \1 # Match the same quote as in group 1 |
| 46 | + */ |
| 47 | + $res = preg_replace( |
| 48 | + '/(["\'])(?:(?!\\1).)*+\\1/', |
| 49 | + "\\1\\1", // Replace the content with two of the quotes that were matched |
| 50 | + $input, |
| 51 | + ); |
| 52 | + |
| 53 | + if (null === $res) { |
| 54 | + throw new RuntimeException("Error in preg_replace"); |
| 55 | + } |
| 56 | + |
| 57 | + return $res; |
| 58 | + } |
| 59 | + |
| 60 | + |
| 61 | + /** |
| 62 | + * Check if the $xpath_expression uses an XPath function that is not in the list of allowed functions |
| 63 | + * |
| 64 | + * @param string $xpathExpression the expression to check. Should be a valid xpath expression |
| 65 | + * @param string[] $allowedFunctions array of string with a list of allowed function names |
| 66 | + * @throws \SimpleSAML\XML\Exception\RuntimeException |
| 67 | + */ |
| 68 | + public static function filterXPathFunction(string $xpathExpression, array $allowedFunctions): void |
| 69 | + { |
| 70 | + /** |
| 71 | + * Look for the function specifier '(' and look for a function name before it. |
| 72 | + * Ignoring whitespace before the '(' and the function name. |
| 73 | + * All functions must match a string on a list of allowed function names |
| 74 | + */ |
| 75 | + $matches = []; |
| 76 | + $res = preg_match_all( |
| 77 | + /** |
| 78 | + * Function names are lower-case alpha (i.e. [a-z]) and can contain one or more hyphens, |
| 79 | + * but cannot start or end with a hyphen. To match this, we start with matching one or more |
| 80 | + * lower-case alpha characters, followed by zero or more atomic groups that start with a hyphen |
| 81 | + * and then match one or more lower-case alpha characters. This ensures that the function name |
| 82 | + * cannot start or end with a hyphen, but can contain one or more hyphens. |
| 83 | + * More than one consecutive hyphen does not match. |
| 84 | + * |
| 85 | + * Use possessive quantifiers (i.e. *+ and ++ instead of * and + respectively) to prevent backtracking |
| 86 | + * and thus prevent a ReDOS. |
| 87 | +
|
| 88 | + * '/([a-z]++(?>-[a-z]++)*+)\s*+\(/' |
| 89 | + * ( # Start a capturing group |
| 90 | + * [a-z]++ # Match one or more lower-case alpha characters |
| 91 | + * (?> # Start an atomic group (no capturing) |
| 92 | + * - # Match a hyphen |
| 93 | + * [a-z]++ # Match one or more lower-case alpha characters, possessively |
| 94 | + * )*+ # Repeat the atomic group zero or more times, |
| 95 | + * ) # End of the capturing group |
| 96 | + * \s*+ # Match zero or more whitespace characters, possessively |
| 97 | + * \( # Match an opening parenthesis |
| 98 | + */ |
| 99 | + |
| 100 | + '/([a-z]++(?>-[a-z]++)*+)\\s*+\\(/', |
| 101 | + $xpathExpression, |
| 102 | + $matches, |
| 103 | + ); |
| 104 | + |
| 105 | + // Check that all the function names we found are in the list of allowed function names |
| 106 | + foreach ($matches[1] as $match) { |
| 107 | + if (!in_array($match, $allowedFunctions)) { |
| 108 | + throw new RuntimeException("Invalid function: '" . $match . "'"); |
| 109 | + } |
| 110 | + } |
| 111 | + } |
| 112 | + |
| 113 | + |
| 114 | + /** |
| 115 | + * Check if the $xpath_expression uses an XPath axis that is not in the list of allowed axes |
| 116 | + * |
| 117 | + * @param string $xpathExpression the expression to check. Should be a valid xpath expression |
| 118 | + * @param string[] $allowedAxes array of string with a list of allowed axes names |
| 119 | + * @throws \SimpleSAML\XML\Exception\RuntimeException |
| 120 | + */ |
| 121 | + public static function filterXPathAxis(string $xpathExpression, array $allowedAxes): void |
| 122 | + { |
| 123 | + /** |
| 124 | + * Look for the axis specifier '::' and look for a function name before it. |
| 125 | + * Ignoring whitespace before the '::' and the axis name. |
| 126 | + * All axes must match a string on a list of allowed axis names |
| 127 | + */ |
| 128 | + $matches = []; |
| 129 | + $res = preg_match_all( |
| 130 | + /** |
| 131 | + * We use the same rules for matching Axis names as we do for function names. |
| 132 | + * The only difference is that we match the '::' instead of the '(' |
| 133 | + * so everything that was said about the regular expression for function names |
| 134 | + * applies here as well. |
| 135 | + * |
| 136 | + * Use possessive quantifiers (i.e. *+ and ++ instead of * and + respectively) to prevent backtracking |
| 137 | + * and thus prevent a ReDOS. |
| 138 | + * |
| 139 | + * '/([a-z]++(?>-[a-z]++)*+)\s*+::' |
| 140 | + * ( # Start a capturing group |
| 141 | + * [a-z]++ # Match one or more lower-case alpha characters |
| 142 | + * (?> # Start an atomic group (no capturing) |
| 143 | + * - # Match a hyphen |
| 144 | + * [a-z]++ # Match one or more lower-case alpha characters, possessively |
| 145 | + * )*+ # Repeat the atomic group zero or more times, |
| 146 | + * ) # End of the capturing group |
| 147 | + * \s*+ # Match zero or more whitespace characters, possessively |
| 148 | + * \( # Match an opening parenthesis |
| 149 | + */ |
| 150 | + |
| 151 | + '/([a-z]++(?>-[a-z]++)*+)\\s*+::/', |
| 152 | + $xpathExpression, |
| 153 | + $matches, |
| 154 | + ); |
| 155 | + |
| 156 | + // Check that all the axes names we found are in the list of allowed axes names |
| 157 | + foreach ($matches[1] as $match) { |
| 158 | + if (!in_array($match, $allowedAxes)) { |
| 159 | + throw new RuntimeException("Invalid axis: '" . $match . "'"); |
| 160 | + } |
| 161 | + } |
| 162 | + } |
| 163 | +} |
0 commit comments