@appthreat/atom-parsetools 1.0.0

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/AttributeEmulator.php

@@ -0,0 +1,56 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+final class AttributeEmulator extends TokenEmulator
+{
+    public function getPhpVersion(): string
+    {
+        return Emulative::PHP_8_0;
+    }
+
+    public function isEmulationNeeded(string $code) : bool
+    {
+        return strpos($code, '#[') !== false;
+    }
+
+    public function emulate(string $code, array $tokens): array
+    {
+        // We need to manually iterate and manage a count because we'll change
+        // the tokens array on the way.
+        $line = 1;
+        for ($i = 0, $c = count($tokens); $i < $c; ++$i) {
+            if ($tokens[$i] === '#' && isset($tokens[$i + 1]) && $tokens[$i + 1] === '[') {
+                array_splice($tokens, $i, 2, [
+                    [\T_ATTRIBUTE, '#[', $line]
+                ]);
+                $c--;
+                continue;
+            }
+            if (\is_array($tokens[$i])) {
+                $line += substr_count($tokens[$i][1], "\n");
+            }
+        }
+
+        return $tokens;
+    }
+
+    public function reverseEmulate(string $code, array $tokens): array
+    {
+        // TODO
+        return $tokens;
+    }
+
+    public function preprocessCode(string $code, array &$patches): string {
+        $pos = 0;
+        while (false !== $pos = strpos($code, '#[', $pos)) {
+            // Replace #[ with %[
+            $code[$pos] = '%';
+            $patches[] = [$pos, 'replace', '#'];
+            $pos += 2;
+        }
+        return $code;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/CoaleseEqualTokenEmulator.php

@@ -0,0 +1,47 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+final class CoaleseEqualTokenEmulator extends TokenEmulator
+{
+    public function getPhpVersion(): string
+    {
+        return Emulative::PHP_7_4;
+    }
+
+    public function isEmulationNeeded(string $code): bool
+    {
+        return strpos($code, '??=') !== false;
+    }
+
+    public function emulate(string $code, array $tokens): array
+    {
+        // We need to manually iterate and manage a count because we'll change
+        // the tokens array on the way
+        $line = 1;
+        for ($i = 0, $c = count($tokens); $i < $c; ++$i) {
+            if (isset($tokens[$i + 1])) {
+                if ($tokens[$i][0] === T_COALESCE && $tokens[$i + 1] === '=') {
+                    array_splice($tokens, $i, 2, [
+                        [\T_COALESCE_EQUAL, '??=', $line]
+                    ]);
+                    $c--;
+                    continue;
+                }
+            }
+            if (\is_array($tokens[$i])) {
+                $line += substr_count($tokens[$i][1], "\n");
+            }
+        }
+
+        return $tokens;
+    }
+
+    public function reverseEmulate(string $code, array $tokens): array
+    {
+        // ??= was not valid code previously, don't bother.
+        return $tokens;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/EnumTokenEmulator.php

@@ -0,0 +1,31 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+final class EnumTokenEmulator extends KeywordEmulator
+{
+    public function getPhpVersion(): string
+    {
+        return Emulative::PHP_8_1;
+    }
+
+    public function getKeywordString(): string
+    {
+        return 'enum';
+    }
+
+    public function getKeywordToken(): int
+    {
+        return \T_ENUM;
+    }
+
+    protected function isKeywordContext(array $tokens, int $pos): bool
+    {
+        return parent::isKeywordContext($tokens, $pos)
+            && isset($tokens[$pos + 2])
+            && $tokens[$pos + 1][0] === \T_WHITESPACE
+            && $tokens[$pos + 2][0] === \T_STRING;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/ExplicitOctalEmulator.php

@@ -0,0 +1,44 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+class ExplicitOctalEmulator extends TokenEmulator {
+    public function getPhpVersion(): string {
+        return Emulative::PHP_8_1;
+    }
+
+    public function isEmulationNeeded(string $code): bool {
+        return strpos($code, '0o') !== false || strpos($code, '0O') !== false;
+    }
+
+    public function emulate(string $code, array $tokens): array {
+        for ($i = 0, $c = count($tokens); $i < $c; ++$i) {
+            if ($tokens[$i][0] == \T_LNUMBER && $tokens[$i][1] === '0' &&
+                isset($tokens[$i + 1]) && $tokens[$i + 1][0] == \T_STRING &&
+                preg_match('/[oO][0-7]+(?:_[0-7]+)*/', $tokens[$i + 1][1])
+            ) {
+                $tokenKind = $this->resolveIntegerOrFloatToken($tokens[$i + 1][1]);
+                array_splice($tokens, $i, 2, [
+                    [$tokenKind, '0' . $tokens[$i + 1][1], $tokens[$i][2]],
+                ]);
+                $c--;
+            }
+        }
+        return $tokens;
+    }
+
+    private function resolveIntegerOrFloatToken(string $str): int
+    {
+        $str = substr($str, 1);
+        $str = str_replace('_', '', $str);
+        $num = octdec($str);
+        return is_float($num) ? \T_DNUMBER : \T_LNUMBER;
+    }
+
+    public function reverseEmulate(string $code, array $tokens): array {
+        // Explicit octals were not legal code previously, don't bother.
+        return $tokens;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/FlexibleDocStringEmulator.php

@@ -0,0 +1,76 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+final class FlexibleDocStringEmulator extends TokenEmulator
+{
+    const FLEXIBLE_DOC_STRING_REGEX = <<<'REGEX'
+/<<<[ \t]*(['"]?)([a-zA-Z_\x80-\xff][a-zA-Z0-9_\x80-\xff]*)\1\r?\n
+(?:.*\r?\n)*?
+(?<indentation>\h*)\2(?![a-zA-Z0-9_\x80-\xff])(?<separator>(?:;?[\r\n])?)/x
+REGEX;
+
+    public function getPhpVersion(): string
+    {
+        return Emulative::PHP_7_3;
+    }
+
+    public function isEmulationNeeded(string $code) : bool
+    {
+        return strpos($code, '<<<') !== false;
+    }
+
+    public function emulate(string $code, array $tokens): array
+    {
+        // Handled by preprocessing + fixup.
+        return $tokens;
+    }
+
+    public function reverseEmulate(string $code, array $tokens): array
+    {
+        // Not supported.
+        return $tokens;
+    }
+
+    public function preprocessCode(string $code, array &$patches): string {
+        if (!preg_match_all(self::FLEXIBLE_DOC_STRING_REGEX, $code, $matches, PREG_SET_ORDER|PREG_OFFSET_CAPTURE)) {
+            // No heredoc/nowdoc found
+            return $code;
+        }
+
+        // Keep track of how much we need to adjust string offsets due to the modifications we
+        // already made
+        $posDelta = 0;
+        foreach ($matches as $match) {
+            $indentation = $match['indentation'][0];
+            $indentationStart = $match['indentation'][1];
+
+            $separator = $match['separator'][0];
+            $separatorStart = $match['separator'][1];
+
+            if ($indentation === '' && $separator !== '') {
+                // Ordinary heredoc/nowdoc
+                continue;
+            }
+
+            if ($indentation !== '') {
+                // Remove indentation
+                $indentationLen = strlen($indentation);
+                $code = substr_replace($code, '', $indentationStart + $posDelta, $indentationLen);
+                $patches[] = [$indentationStart + $posDelta, 'add', $indentation];
+                $posDelta -= $indentationLen;
+            }
+
+            if ($separator === '') {
+                // Insert newline as separator
+                $code = substr_replace($code, "\n", $separatorStart + $posDelta, 0);
+                $patches[] = [$separatorStart + $posDelta, 'remove', "\n"];
+                $posDelta += 1;
+            }
+        }
+
+        return $code;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/FnTokenEmulator.php

@@ -0,0 +1,23 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+final class FnTokenEmulator extends KeywordEmulator
+{
+    public function getPhpVersion(): string
+    {
+        return Emulative::PHP_7_4;
+    }
+
+    public function getKeywordString(): string
+    {
+        return 'fn';
+    }
+
+    public function getKeywordToken(): int
+    {
+        return \T_FN;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/KeywordEmulator.php

@@ -0,0 +1,62 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+abstract class KeywordEmulator extends TokenEmulator
+{
+    abstract function getKeywordString(): string;
+    abstract function getKeywordToken(): int;
+
+    public function isEmulationNeeded(string $code): bool
+    {
+        return strpos(strtolower($code), $this->getKeywordString()) !== false;
+    }
+
+    protected function isKeywordContext(array $tokens, int $pos): bool
+    {
+        $previousNonSpaceToken = $this->getPreviousNonSpaceToken($tokens, $pos);
+        return $previousNonSpaceToken === null || $previousNonSpaceToken[0] !== \T_OBJECT_OPERATOR;
+    }
+
+    public function emulate(string $code, array $tokens): array
+    {
+        $keywordString = $this->getKeywordString();
+        foreach ($tokens as $i => $token) {
+            if ($token[0] === T_STRING && strtolower($token[1]) === $keywordString
+                    && $this->isKeywordContext($tokens, $i)) {
+                $tokens[$i][0] = $this->getKeywordToken();
+            }
+        }
+
+        return $tokens;
+    }
+
+    /**
+     * @param mixed[] $tokens
+     * @return array|string|null
+     */
+    private function getPreviousNonSpaceToken(array $tokens, int $start)
+    {
+        for ($i = $start - 1; $i >= 0; --$i) {
+            if ($tokens[$i][0] === T_WHITESPACE) {
+                continue;
+            }
+
+            return $tokens[$i];
+        }
+
+        return null;
+    }
+
+    public function reverseEmulate(string $code, array $tokens): array
+    {
+        $keywordToken = $this->getKeywordToken();
+        foreach ($tokens as $i => $token) {
+            if ($token[0] === $keywordToken) {
+                $tokens[$i][0] = \T_STRING;
+            }
+        }
+
+        return $tokens;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/MatchTokenEmulator.php

@@ -0,0 +1,23 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+final class MatchTokenEmulator extends KeywordEmulator
+{
+    public function getPhpVersion(): string
+    {
+        return Emulative::PHP_8_0;
+    }
+
+    public function getKeywordString(): string
+    {
+        return 'match';
+    }
+
+    public function getKeywordToken(): int
+    {
+        return \T_MATCH;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/NullsafeTokenEmulator.php

@@ -0,0 +1,67 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+final class NullsafeTokenEmulator extends TokenEmulator
+{
+    public function getPhpVersion(): string
+    {
+        return Emulative::PHP_8_0;
+    }
+
+    public function isEmulationNeeded(string $code): bool
+    {
+        return strpos($code, '?->') !== false;
+    }
+
+    public function emulate(string $code, array $tokens): array
+    {
+        // We need to manually iterate and manage a count because we'll change
+        // the tokens array on the way
+        $line = 1;
+        for ($i = 0, $c = count($tokens); $i < $c; ++$i) {
+            if ($tokens[$i] === '?' && isset($tokens[$i + 1]) && $tokens[$i + 1][0] === \T_OBJECT_OPERATOR) {
+                array_splice($tokens, $i, 2, [
+                    [\T_NULLSAFE_OBJECT_OPERATOR, '?->', $line]
+                ]);
+                $c--;
+                continue;
+            }
+
+            // Handle ?-> inside encapsed string.
+            if ($tokens[$i][0] === \T_ENCAPSED_AND_WHITESPACE && isset($tokens[$i - 1])
+                && $tokens[$i - 1][0] === \T_VARIABLE
+                && preg_match('/^\?->([a-zA-Z_\x80-\xff][a-zA-Z0-9_\x80-\xff]*)/', $tokens[$i][1], $matches)
+            ) {
+                $replacement = [
+                    [\T_NULLSAFE_OBJECT_OPERATOR, '?->', $line],
+                    [\T_STRING, $matches[1], $line],
+                ];
+                if (\strlen($matches[0]) !== \strlen($tokens[$i][1])) {
+                    $replacement[] = [
+                        \T_ENCAPSED_AND_WHITESPACE,
+                        \substr($tokens[$i][1], \strlen($matches[0])),
+                        $line
+                    ];
+                }
+                array_splice($tokens, $i, 1, $replacement);
+                $c += \count($replacement) - 1;
+                continue;
+            }
+
+            if (\is_array($tokens[$i])) {
+                $line += substr_count($tokens[$i][1], "\n");
+            }
+        }
+
+        return $tokens;
+    }
+
+    public function reverseEmulate(string $code, array $tokens): array
+    {
+        // ?-> was not valid code previously, don't bother.
+        return $tokens;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/NumericLiteralSeparatorEmulator.php

@@ -0,0 +1,105 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+final class NumericLiteralSeparatorEmulator extends TokenEmulator
+{
+    const BIN = '(?:0b[01]+(?:_[01]+)*)';
+    const HEX = '(?:0x[0-9a-f]+(?:_[0-9a-f]+)*)';
+    const DEC = '(?:[0-9]+(?:_[0-9]+)*)';
+    const SIMPLE_FLOAT = '(?:' . self::DEC . '\.' . self::DEC . '?|\.' . self::DEC . ')';
+    const EXP = '(?:e[+-]?' . self::DEC . ')';
+    const FLOAT = '(?:' . self::SIMPLE_FLOAT . self::EXP . '?|' . self::DEC . self::EXP . ')';
+    const NUMBER = '~' . self::FLOAT . '|' . self::BIN . '|' . self::HEX . '|' . self::DEC . '~iA';
+
+    public function getPhpVersion(): string
+    {
+        return Emulative::PHP_7_4;
+    }
+
+    public function isEmulationNeeded(string $code) : bool
+    {
+        return preg_match('~[0-9]_[0-9]~', $code)
+            || preg_match('~0x[0-9a-f]+_[0-9a-f]~i', $code);
+    }
+
+    public function emulate(string $code, array $tokens): array
+    {
+        // We need to manually iterate and manage a count because we'll change
+        // the tokens array on the way
+        $codeOffset = 0;
+        for ($i = 0, $c = count($tokens); $i < $c; ++$i) {
+            $token = $tokens[$i];
+            $tokenLen = \strlen(\is_array($token) ? $token[1] : $token);
+
+            if ($token[0] !== T_LNUMBER && $token[0] !== T_DNUMBER) {
+                $codeOffset += $tokenLen;
+                continue;
+            }
+
+            $res = preg_match(self::NUMBER, $code, $matches, 0, $codeOffset);
+            assert($res, "No number at number token position");
+
+            $match = $matches[0];
+            $matchLen = \strlen($match);
+            if ($matchLen === $tokenLen) {
+                // Original token already holds the full number.
+                $codeOffset += $tokenLen;
+                continue;
+            }
+
+            $tokenKind = $this->resolveIntegerOrFloatToken($match);
+            $newTokens = [[$tokenKind, $match, $token[2]]];
+
+            $numTokens = 1;
+            $len = $tokenLen;
+            while ($matchLen > $len) {
+                $nextToken = $tokens[$i + $numTokens];
+                $nextTokenText = \is_array($nextToken) ? $nextToken[1] : $nextToken;
+                $nextTokenLen = \strlen($nextTokenText);
+
+                $numTokens++;
+                if ($matchLen < $len + $nextTokenLen) {
+                    // Split trailing characters into a partial token.
+                    assert(is_array($nextToken), "Partial token should be an array token");
+                    $partialText = substr($nextTokenText, $matchLen - $len);
+                    $newTokens[] = [$nextToken[0], $partialText, $nextToken[2]];
+                    break;
+                }
+
+                $len += $nextTokenLen;
+            }
+
+            array_splice($tokens, $i, $numTokens, $newTokens);
+            $c -= $numTokens - \count($newTokens);
+            $codeOffset += $matchLen;
+        }
+
+        return $tokens;
+    }
+
+    private function resolveIntegerOrFloatToken(string $str): int
+    {
+        $str = str_replace('_', '', $str);
+
+        if (stripos($str, '0b') === 0) {
+            $num = bindec($str);
+        } elseif (stripos($str, '0x') === 0) {
+            $num = hexdec($str);
+        } elseif (stripos($str, '0') === 0 && ctype_digit($str)) {
+            $num = octdec($str);
+        } else {
+            $num = +$str;
+        }
+
+        return is_float($num) ? T_DNUMBER : T_LNUMBER;
+    }
+
+    public function reverseEmulate(string $code, array $tokens): array
+    {
+        // Numeric separators were not legal code previously, don't bother.
+        return $tokens;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/ReadonlyFunctionTokenEmulator.php

@@ -0,0 +1,31 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+/*
+ * In PHP 8.1, "readonly(" was special cased in the lexer in order to support functions with
+ * name readonly. In PHP 8.2, this may conflict with readonly properties having a DNF type. For
+ * this reason, PHP 8.2 instead treats this as T_READONLY and then handles it specially in the
+ * parser. This emulator only exists to handle this special case, which is skipped by the
+ * PHP 8.1 ReadonlyTokenEmulator.
+ */
+class ReadonlyFunctionTokenEmulator extends KeywordEmulator {
+    public function getKeywordString(): string {
+        return 'readonly';
+    }
+
+    public function getKeywordToken(): int {
+        return \T_READONLY;
+    }
+
+    public function getPhpVersion(): string {
+        return Emulative::PHP_8_2;
+    }
+
+    public function reverseEmulate(string $code, array $tokens): array {
+        // Don't bother
+        return $tokens;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/ReadonlyTokenEmulator.php

@@ -0,0 +1,36 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+use PhpParser\Lexer\Emulative;
+
+final class ReadonlyTokenEmulator extends KeywordEmulator
+{
+    public function getPhpVersion(): string
+    {
+        return Emulative::PHP_8_1;
+    }
+
+    public function getKeywordString(): string
+    {
+        return 'readonly';
+    }
+
+    public function getKeywordToken(): int
+    {
+        return \T_READONLY;
+    }
+
+    protected function isKeywordContext(array $tokens, int $pos): bool
+    {
+        if (!parent::isKeywordContext($tokens, $pos)) {
+            return false;
+        }
+        // Support "function readonly("
+        return !(isset($tokens[$pos + 1]) &&
+                 ($tokens[$pos + 1][0] === '(' ||
+                  ($tokens[$pos + 1][0] === \T_WHITESPACE &&
+                   isset($tokens[$pos + 2]) &&
+                   $tokens[$pos + 2][0] === '(')));
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/ReverseEmulator.php

@@ -0,0 +1,36 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+/**
+ * Reverses emulation direction of the inner emulator.
+ */
+final class ReverseEmulator extends TokenEmulator
+{
+    /** @var TokenEmulator Inner emulator */
+    private $emulator;
+
+    public function __construct(TokenEmulator $emulator) {
+        $this->emulator = $emulator;
+    }
+
+    public function getPhpVersion(): string {
+        return $this->emulator->getPhpVersion();
+    }
+
+    public function isEmulationNeeded(string $code): bool {
+        return $this->emulator->isEmulationNeeded($code);
+    }
+
+    public function emulate(string $code, array $tokens): array {
+        return $this->emulator->reverseEmulate($code, $tokens);
+    }
+
+    public function reverseEmulate(string $code, array $tokens): array {
+        return $this->emulator->emulate($code, $tokens);
+    }
+
+    public function preprocessCode(string $code, array &$patches): string {
+        return $code;
+    }
+}

package/plugins/nikic/php-parser/lib/PhpParser/Lexer/TokenEmulator/TokenEmulator.php

@@ -0,0 +1,25 @@
+<?php declare(strict_types=1);
+
+namespace PhpParser\Lexer\TokenEmulator;
+
+/** @internal */
+abstract class TokenEmulator
+{
+    abstract public function getPhpVersion(): string;
+
+    abstract public function isEmulationNeeded(string $code): bool;
+
+    /**
+     * @return array Modified Tokens
+     */
+    abstract public function emulate(string $code, array $tokens): array;
+
+    /**
+     * @return array Modified Tokens
+     */
+    abstract public function reverseEmulate(string $code, array $tokens): array;
+
+    public function preprocessCode(string $code, array &$patches): string {
+        return $code;
+    }
+}