Fix regular expression parsing #3244
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -7,6 +7,7 @@ | |
| use Hoa\Compiler\Llk\TreeNode; | ||
| use Hoa\Exception\Exception; | ||
| use Hoa\File\Read; | ||
| use Nette\Utils\RegexpException; | ||
| use Nette\Utils\Strings; | ||
| use PhpParser\Node\Expr; | ||
| use PhpParser\Node\Name; | ||
|
|
@@ -31,7 +32,11 @@ | |
| use function in_array; | ||
| use function is_int; | ||
| use function is_string; | ||
| use function rtrim; | ||
| use function sscanf; | ||
| use function str_replace; | ||
| use function strlen; | ||
| use function substr; | ||
| use const PREG_OFFSET_CAPTURE; | ||
| use const PREG_UNMATCHED_AS_NULL; | ||
|
|
||
|
|
@@ -375,6 +380,13 @@ private function parseGroups(string $regex): ?array | |
| self::$parser = Llk::load(new Read(__DIR__ . '/../../../resources/RegexGrammar.pp')); | ||
|
There was a problem hiding this comment. @ondrejmirtes Are we ok loading the file from here instead? Not sure what this does in terms of licensing, nor if there's a better place to put it. If so I'll delete the patch file and drop the hoa/regex requirement. There was a problem hiding this comment. I am not sure this path will work in the packaged phpstan.phar - while I agree that having the There was a problem hiding this comment. I completely agree moving from the patching as the patching is very hard to understand and the changes needed are more and more complex. There was a problem hiding this comment. Ok I got rid of it, we'll see what @ondrejmirtes says :) There was a problem hiding this comment. I propose to add the whole and unpatched |
||
| } | ||
|
|
||
| try { | ||
| Strings::match('', $regex); | ||
| } catch (RegexpException) { | ||
| // pattern is invalid, so let the RegularExpressionPatternRule report it | ||
| return null; | ||
| } | ||
Seldaek marked this conversation as resolved.
Show resolved
Hide resolved
|
||
|
|
||
| try { | ||
| $ast = self::$parser->parse($regex); | ||
Seldaek marked this conversation as resolved.
Show resolved
Hide resolved
|
||
| } catch (Exception) { | ||
|
|
@@ -516,25 +528,37 @@ private function getQuantificationRange(TreeNode $node): array | |
| $lastChild = $node->getChild($node->getChildrenNumber() - 1); | ||
| $value = $lastChild->getValue(); | ||
|
|
||
| // normalize away possessive and lazy quantifier-modifiers | ||
| $token = str_replace(['_possessive', '_lazy'], '', $value['token']); | ||
| $value = rtrim($value['value'], '+?'); | ||
|
|
||
| if ($token === 'n_to_m') { | ||
| if (sscanf($value, '{%d,%d}', $n, $m) !== 2 || !is_int($n) || !is_int($m)) { | ||
| throw new ShouldNotHappenException(); | ||
| } | ||
|
|
||
| $min = $n; | ||
| $max = $m; | ||
| } elseif ($token === 'n_or_more') { | ||
| if (sscanf($value, '{%d,}', $n) !== 1 || !is_int($n)) { | ||
| throw new ShouldNotHappenException(); | ||
| } | ||
|
|
||
| $min = $n; | ||
| } elseif ($token === 'exactly_n') { | ||
| if (sscanf($value, '{%d}', $n) !== 1 || !is_int($n)) { | ||
| throw new ShouldNotHappenException(); | ||
| } | ||
|
|
||
| $min = $n; | ||
| $max = $n; | ||
| } elseif ($token === 'zero_or_one') { | ||
| $min = 0; | ||
| $max = 1; | ||
| } elseif ($token === 'zero_or_more') { | ||
| $min = 0; | ||
| } elseif ($token === 'one_or_more') { | ||
| $min = 1; | ||
| } | ||
|
|
||
| return [$min, $max]; | ||
|
|
@@ -591,20 +615,8 @@ private function walkGroupAst(TreeNode $ast, TrinaryLogic &$isNonEmpty, TrinaryL | |
| if ($literalValue !== null) { | ||
| if (Strings::match($literalValue, '/^\d+$/') === null) { | ||
| $isNumeric = TrinaryLogic::createNo(); | ||
| } elseif ($isNumeric->maybe()) { | ||
| $isNumeric = TrinaryLogic::createYes(); | ||
Seldaek marked this conversation as resolved.
Show resolved
Hide resolved
|
||
| } | ||
|
|
||
| if (!$inOptionalQuantification) { | ||
|
|
@@ -613,32 +625,11 @@ private function walkGroupAst(TreeNode $ast, TrinaryLogic &$isNonEmpty, TrinaryL | |
| } | ||
| } | ||
|
|
||
Seldaek marked this conversation as resolved.
Show resolved
Hide resolved
|
||
| // [^0-9] should not parse as numeric-string, and [^list-everything-but-numbers] is technically | ||
| // doable but really silly compared to just \d so we can safely assume the string is not numeric | ||
| // for negative classes | ||
| if ($ast->getId() === '#negativeclass') { | ||
| $isNumeric = TrinaryLogic::createNo(); | ||
| } | ||
|
|
||
| foreach ($children as $child) { | ||
|
|
@@ -653,13 +644,65 @@ private function walkGroupAst(TreeNode $ast, TrinaryLogic &$isNonEmpty, TrinaryL | |
|
|
||
| private function getLiteralValue(TreeNode $node): ?string | ||
| { | ||
| if ($node->getId() !== 'token') { | ||
| return null; | ||
| } | ||
|
|
||
| // token is the token name from grammar without the namespace so literal and class:literal are both called literal here | ||
| $token = $node->getValueToken(); | ||
| $value = $node->getValueValue(); | ||
|
|
||
| if (in_array($token, ['literal', 'escaped_end_class'], true)) { | ||
| if (strlen($node->getValueValue()) > 1 && $value[0] === '\\') { | ||
| return substr($value, 1); | ||
| } | ||
|
|
||
| return $value; | ||
| } | ||
|
|
||
| // literal "-" in front/back of a character class like '[-a-z]' or '[abc-]', not forming a range | ||
| if ($token === 'range') { | ||
| return $value; | ||
| } | ||
|
|
||
| // literal "[" or "]" inside character classes '[[]' or '[]]' | ||
| if (in_array($token, ['class_', '_class_literal'], true)) { | ||
| return $value; | ||
| } | ||
|
|
||
| // character escape sequences, just return a fixed string | ||
| if (in_array($token, ['character', 'dynamic_character', 'character_type'], true)) { | ||
| if ($token === 'character_type' && $value === '\d') { | ||
| return '0'; | ||
| } | ||
|
|
||
| return $value; | ||
| } | ||
|
|
||
| // [:digit:] and the like, more support coming later | ||
| if ($token === 'posix_class') { | ||
| if ($value === '[:digit:]') { | ||
| return '0'; | ||
| } | ||
| if (in_array($value, ['[:alpha:]', '[:alnum:]', '[:upper:]', '[:lower:]', '[:word:]', '[:ascii:]', '[:print:]', '[:xdigit:]', '[:graph:]'], true)) { | ||
| return 'a'; | ||
| } | ||
| if ($value === '[:blank:]') { | ||
| return " \t"; | ||
| } | ||
| if ($value === '[:cntrl:]') { | ||
| return "\x00\x1F"; | ||
| } | ||
| if ($value === '[:space:]') { | ||
| return " \t\r\n\v\f"; | ||
| } | ||
| if ($value === '[:punct:]') { | ||
| return '!"#$%&\'()*+,\-./:;<=>?@[\]^_`{|}~'; | ||
| } | ||
| } | ||
|
|
||
| if ($token === 'anchor' || $token === 'match_point_reset') { | ||
| return ''; | ||
| } | ||
|
|
||
| return null; | ||
|
|
||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
\ncan be in pcre class as wellThere was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes I am unsure why this is here, I haven't experimented much, but it is rare to have actual newlines in regexes as they're usually escaped as \n, while this here matches a real newline.
I'll create a new issue for this to look at it later, together with proper support for the x modifier, because it's very much related. Edit: phpstan/phpstan#11360