first commit

This commit is contained in:
Sampanna Rimal
2024-08-27 17:48:06 +05:45
commit 53c0140f58
10839 changed files with 1125847 additions and 0 deletions

View File

@ -0,0 +1,31 @@
<?php declare(strict_types=1);
namespace PhpParser\Internal;
/**
* @internal
*/
class DiffElem {
public const TYPE_KEEP = 0;
public const TYPE_REMOVE = 1;
public const TYPE_ADD = 2;
public const TYPE_REPLACE = 3;
/** @var int One of the TYPE_* constants */
public int $type;
/** @var mixed Is null for add operations */
public $old;
/** @var mixed Is null for remove operations */
public $new;
/**
* @param int $type One of the TYPE_* constants
* @param mixed $old Is null for add operations
* @param mixed $new Is null for remove operations
*/
public function __construct(int $type, $old, $new) {
$this->type = $type;
$this->old = $old;
$this->new = $new;
}
}

View File

@ -0,0 +1,178 @@
<?php declare(strict_types=1);
namespace PhpParser\Internal;
/**
* Implements the Myers diff algorithm.
*
* Myers, Eugene W. "An O (ND) difference algorithm and its variations."
* Algorithmica 1.1 (1986): 251-266.
*
* @template T
* @internal
*/
class Differ {
/** @var callable(T, T): bool */
private $isEqual;
/**
* Create differ over the given equality relation.
*
* @param callable(T, T): bool $isEqual Equality relation
*/
public function __construct(callable $isEqual) {
$this->isEqual = $isEqual;
}
/**
* Calculate diff (edit script) from $old to $new.
*
* @param T[] $old Original array
* @param T[] $new New array
*
* @return DiffElem[] Diff (edit script)
*/
public function diff(array $old, array $new): array {
$old = \array_values($old);
$new = \array_values($new);
list($trace, $x, $y) = $this->calculateTrace($old, $new);
return $this->extractDiff($trace, $x, $y, $old, $new);
}
/**
* Calculate diff, including "replace" operations.
*
* If a sequence of remove operations is followed by the same number of add operations, these
* will be coalesced into replace operations.
*
* @param T[] $old Original array
* @param T[] $new New array
*
* @return DiffElem[] Diff (edit script), including replace operations
*/
public function diffWithReplacements(array $old, array $new): array {
return $this->coalesceReplacements($this->diff($old, $new));
}
/**
* @param T[] $old
* @param T[] $new
* @return array{array<int, array<int, int>>, int, int}
*/
private function calculateTrace(array $old, array $new): array {
$n = \count($old);
$m = \count($new);
$max = $n + $m;
$v = [1 => 0];
$trace = [];
for ($d = 0; $d <= $max; $d++) {
$trace[] = $v;
for ($k = -$d; $k <= $d; $k += 2) {
if ($k === -$d || ($k !== $d && $v[$k - 1] < $v[$k + 1])) {
$x = $v[$k + 1];
} else {
$x = $v[$k - 1] + 1;
}
$y = $x - $k;
while ($x < $n && $y < $m && ($this->isEqual)($old[$x], $new[$y])) {
$x++;
$y++;
}
$v[$k] = $x;
if ($x >= $n && $y >= $m) {
return [$trace, $x, $y];
}
}
}
throw new \Exception('Should not happen');
}
/**
* @param array<int, array<int, int>> $trace
* @param T[] $old
* @param T[] $new
* @return DiffElem[]
*/
private function extractDiff(array $trace, int $x, int $y, array $old, array $new): array {
$result = [];
for ($d = \count($trace) - 1; $d >= 0; $d--) {
$v = $trace[$d];
$k = $x - $y;
if ($k === -$d || ($k !== $d && $v[$k - 1] < $v[$k + 1])) {
$prevK = $k + 1;
} else {
$prevK = $k - 1;
}
$prevX = $v[$prevK];
$prevY = $prevX - $prevK;
while ($x > $prevX && $y > $prevY) {
$result[] = new DiffElem(DiffElem::TYPE_KEEP, $old[$x - 1], $new[$y - 1]);
$x--;
$y--;
}
if ($d === 0) {
break;
}
while ($x > $prevX) {
$result[] = new DiffElem(DiffElem::TYPE_REMOVE, $old[$x - 1], null);
$x--;
}
while ($y > $prevY) {
$result[] = new DiffElem(DiffElem::TYPE_ADD, null, $new[$y - 1]);
$y--;
}
}
return array_reverse($result);
}
/**
* Coalesce equal-length sequences of remove+add into a replace operation.
*
* @param DiffElem[] $diff
* @return DiffElem[]
*/
private function coalesceReplacements(array $diff): array {
$newDiff = [];
$c = \count($diff);
for ($i = 0; $i < $c; $i++) {
$diffType = $diff[$i]->type;
if ($diffType !== DiffElem::TYPE_REMOVE) {
$newDiff[] = $diff[$i];
continue;
}
$j = $i;
while ($j < $c && $diff[$j]->type === DiffElem::TYPE_REMOVE) {
$j++;
}
$k = $j;
while ($k < $c && $diff[$k]->type === DiffElem::TYPE_ADD) {
$k++;
}
if ($j - $i === $k - $j) {
$len = $j - $i;
for ($n = 0; $n < $len; $n++) {
$newDiff[] = new DiffElem(
DiffElem::TYPE_REPLACE, $diff[$i + $n]->old, $diff[$j + $n]->new
);
}
} else {
for (; $i < $k; $i++) {
$newDiff[] = $diff[$i];
}
}
$i = $k - 1;
}
return $newDiff;
}
}

View File

@ -0,0 +1,71 @@
<?php declare(strict_types=1);
namespace PhpParser\Internal;
use PhpParser\Node;
use PhpParser\Node\Expr;
/**
* This node is used internally by the format-preserving pretty printer to print anonymous classes.
*
* The normal anonymous class structure violates assumptions about the order of token offsets.
* Namely, the constructor arguments are part of the Expr\New_ node and follow the class node, even
* though they are actually interleaved with them. This special node type is used temporarily to
* restore a sane token offset order.
*
* @internal
*/
class PrintableNewAnonClassNode extends Expr {
/** @var Node\AttributeGroup[] PHP attribute groups */
public array $attrGroups;
/** @var int Modifiers */
public int $flags;
/** @var (Node\Arg|Node\VariadicPlaceholder)[] Arguments */
public array $args;
/** @var null|Node\Name Name of extended class */
public ?Node\Name $extends;
/** @var Node\Name[] Names of implemented interfaces */
public array $implements;
/** @var Node\Stmt[] Statements */
public array $stmts;
/**
* @param Node\AttributeGroup[] $attrGroups PHP attribute groups
* @param (Node\Arg|Node\VariadicPlaceholder)[] $args Arguments
* @param Node\Name|null $extends Name of extended class
* @param Node\Name[] $implements Names of implemented interfaces
* @param Node\Stmt[] $stmts Statements
* @param array<string, mixed> $attributes Attributes
*/
public function __construct(
array $attrGroups, int $flags, array $args, ?Node\Name $extends, array $implements,
array $stmts, array $attributes
) {
parent::__construct($attributes);
$this->attrGroups = $attrGroups;
$this->flags = $flags;
$this->args = $args;
$this->extends = $extends;
$this->implements = $implements;
$this->stmts = $stmts;
}
public static function fromNewNode(Expr\New_ $newNode): self {
$class = $newNode->class;
assert($class instanceof Node\Stmt\Class_);
// We don't assert that $class->name is null here, to allow consumers to assign unique names
// to anonymous classes for their own purposes. We simplify ignore the name here.
return new self(
$class->attrGroups, $class->flags, $newNode->args, $class->extends, $class->implements,
$class->stmts, $newNode->getAttributes()
);
}
public function getType(): string {
return 'Expr_PrintableNewAnonClass';
}
public function getSubNodeNames(): array {
return ['attrGroups', 'flags', 'args', 'extends', 'implements', 'stmts'];
}
}

View File

@ -0,0 +1,237 @@
<?php declare(strict_types=1);
namespace PhpParser\Internal;
if (\PHP_VERSION_ID >= 80000) {
class TokenPolyfill extends \PhpToken {
}
return;
}
/**
* This is a polyfill for the PhpToken class introduced in PHP 8.0. We do not actually polyfill
* PhpToken, because composer might end up picking a different polyfill implementation, which does
* not meet our requirements.
*
* @internal
*/
class TokenPolyfill {
/** @var int The ID of the token. Either a T_* constant of a character code < 256. */
public int $id;
/** @var string The textual content of the token. */
public string $text;
/** @var int The 1-based starting line of the token (or -1 if unknown). */
public int $line;
/** @var int The 0-based starting position of the token (or -1 if unknown). */
public int $pos;
/** @var array<int, bool> Tokens ignored by the PHP parser. */
private const IGNORABLE_TOKENS = [
\T_WHITESPACE => true,
\T_COMMENT => true,
\T_DOC_COMMENT => true,
\T_OPEN_TAG => true,
];
/** @var array<int, bool> Tokens that may be part of a T_NAME_* identifier. */
private static array $identifierTokens;
/**
* Create a Token with the given ID and text, as well optional line and position information.
*/
final public function __construct(int $id, string $text, int $line = -1, int $pos = -1) {
$this->id = $id;
$this->text = $text;
$this->line = $line;
$this->pos = $pos;
}
/**
* Get the name of the token. For single-char tokens this will be the token character.
* Otherwise it will be a T_* style name, or null if the token ID is unknown.
*/
public function getTokenName(): ?string {
if ($this->id < 256) {
return \chr($this->id);
}
$name = token_name($this->id);
return $name === 'UNKNOWN' ? null : $name;
}
/**
* Check whether the token is of the given kind. The kind may be either an integer that matches
* the token ID, a string that matches the token text, or an array of integers/strings. In the
* latter case, the function returns true if any of the kinds in the array match.
*
* @param int|string|(int|string)[] $kind
*/
public function is($kind): bool {
if (\is_int($kind)) {
return $this->id === $kind;
}
if (\is_string($kind)) {
return $this->text === $kind;
}
if (\is_array($kind)) {
foreach ($kind as $entry) {
if (\is_int($entry)) {
if ($this->id === $entry) {
return true;
}
} elseif (\is_string($entry)) {
if ($this->text === $entry) {
return true;
}
} else {
throw new \TypeError(
'Argument #1 ($kind) must only have elements of type string|int, ' .
gettype($entry) . ' given');
}
}
return false;
}
throw new \TypeError(
'Argument #1 ($kind) must be of type string|int|array, ' .gettype($kind) . ' given');
}
/**
* Check whether this token would be ignored by the PHP parser. Returns true for T_WHITESPACE,
* T_COMMENT, T_DOC_COMMENT and T_OPEN_TAG, and false for everything else.
*/
public function isIgnorable(): bool {
return isset(self::IGNORABLE_TOKENS[$this->id]);
}
/**
* Return the textual content of the token.
*/
public function __toString(): string {
return $this->text;
}
/**
* Tokenize the given source code and return an array of tokens.
*
* This performs certain canonicalizations to match the PHP 8.0 token format:
* * Bad characters are represented using T_BAD_CHARACTER rather than omitted.
* * T_COMMENT does not include trailing newlines, instead the newline is part of a following
* T_WHITESPACE token.
* * Namespaced names are represented using T_NAME_* tokens.
*
* @return static[]
*/
public static function tokenize(string $code, int $flags = 0): array {
self::init();
$tokens = [];
$line = 1;
$pos = 0;
$origTokens = \token_get_all($code, $flags);
$numTokens = \count($origTokens);
for ($i = 0; $i < $numTokens; $i++) {
$token = $origTokens[$i];
if (\is_string($token)) {
if (\strlen($token) === 2) {
// b" and B" are tokenized as single-char tokens, even though they aren't.
$tokens[] = new static(\ord('"'), $token, $line, $pos);
$pos += 2;
} else {
$tokens[] = new static(\ord($token), $token, $line, $pos);
$pos++;
}
} else {
$id = $token[0];
$text = $token[1];
// Emulate PHP 8.0 comment format, which does not include trailing whitespace anymore.
if ($id === \T_COMMENT && \substr($text, 0, 2) !== '/*' &&
\preg_match('/(\r\n|\n|\r)$/D', $text, $matches)
) {
$trailingNewline = $matches[0];
$text = \substr($text, 0, -\strlen($trailingNewline));
$tokens[] = new static($id, $text, $line, $pos);
$pos += \strlen($text);
if ($i + 1 < $numTokens && $origTokens[$i + 1][0] === \T_WHITESPACE) {
// Move trailing newline into following T_WHITESPACE token, if it already exists.
$origTokens[$i + 1][1] = $trailingNewline . $origTokens[$i + 1][1];
$origTokens[$i + 1][2]--;
} else {
// Otherwise, we need to create a new T_WHITESPACE token.
$tokens[] = new static(\T_WHITESPACE, $trailingNewline, $line, $pos);
$line++;
$pos += \strlen($trailingNewline);
}
continue;
}
// Emulate PHP 8.0 T_NAME_* tokens, by combining sequences of T_NS_SEPARATOR and
// T_STRING into a single token.
if (($id === \T_NS_SEPARATOR || isset(self::$identifierTokens[$id]))) {
$newText = $text;
$lastWasSeparator = $id === \T_NS_SEPARATOR;
for ($j = $i + 1; $j < $numTokens; $j++) {
if ($lastWasSeparator) {
if (!isset(self::$identifierTokens[$origTokens[$j][0]])) {
break;
}
$lastWasSeparator = false;
} else {
if ($origTokens[$j][0] !== \T_NS_SEPARATOR) {
break;
}
$lastWasSeparator = true;
}
$newText .= $origTokens[$j][1];
}
if ($lastWasSeparator) {
// Trailing separator is not part of the name.
$j--;
$newText = \substr($newText, 0, -1);
}
if ($j > $i + 1) {
if ($id === \T_NS_SEPARATOR) {
$id = \T_NAME_FULLY_QUALIFIED;
} elseif ($id === \T_NAMESPACE) {
$id = \T_NAME_RELATIVE;
} else {
$id = \T_NAME_QUALIFIED;
}
$tokens[] = new static($id, $newText, $line, $pos);
$pos += \strlen($newText);
$i = $j - 1;
continue;
}
}
$tokens[] = new static($id, $text, $line, $pos);
$line += \substr_count($text, "\n");
$pos += \strlen($text);
}
}
return $tokens;
}
/** Initialize private static state needed by tokenize(). */
private static function init(): void {
if (isset(self::$identifierTokens)) {
return;
}
// Based on semi_reserved production.
self::$identifierTokens = \array_fill_keys([
\T_STRING,
\T_STATIC, \T_ABSTRACT, \T_FINAL, \T_PRIVATE, \T_PROTECTED, \T_PUBLIC, \T_READONLY,
\T_INCLUDE, \T_INCLUDE_ONCE, \T_EVAL, \T_REQUIRE, \T_REQUIRE_ONCE, \T_LOGICAL_OR, \T_LOGICAL_XOR, \T_LOGICAL_AND,
\T_INSTANCEOF, \T_NEW, \T_CLONE, \T_EXIT, \T_IF, \T_ELSEIF, \T_ELSE, \T_ENDIF, \T_ECHO, \T_DO, \T_WHILE,
\T_ENDWHILE, \T_FOR, \T_ENDFOR, \T_FOREACH, \T_ENDFOREACH, \T_DECLARE, \T_ENDDECLARE, \T_AS, \T_TRY, \T_CATCH,
\T_FINALLY, \T_THROW, \T_USE, \T_INSTEADOF, \T_GLOBAL, \T_VAR, \T_UNSET, \T_ISSET, \T_EMPTY, \T_CONTINUE, \T_GOTO,
\T_FUNCTION, \T_CONST, \T_RETURN, \T_PRINT, \T_YIELD, \T_LIST, \T_SWITCH, \T_ENDSWITCH, \T_CASE, \T_DEFAULT,
\T_BREAK, \T_ARRAY, \T_CALLABLE, \T_EXTENDS, \T_IMPLEMENTS, \T_NAMESPACE, \T_TRAIT, \T_INTERFACE, \T_CLASS,
\T_CLASS_C, \T_TRAIT_C, \T_FUNC_C, \T_METHOD_C, \T_LINE, \T_FILE, \T_DIR, \T_NS_C, \T_HALT_COMPILER, \T_FN,
\T_MATCH,
], true);
}
}

View File

@ -0,0 +1,275 @@
<?php declare(strict_types=1);
namespace PhpParser\Internal;
use PhpParser\Token;
/**
* Provides operations on token streams, for use by pretty printer.
*
* @internal
*/
class TokenStream {
/** @var Token[] Tokens (in PhpToken::tokenize() format) */
private array $tokens;
/** @var int[] Map from position to indentation */
private array $indentMap;
/**
* Create token stream instance.
*
* @param Token[] $tokens Tokens in PhpToken::tokenize() format
*/
public function __construct(array $tokens) {
$this->tokens = $tokens;
$this->indentMap = $this->calcIndentMap();
}
/**
* Whether the given position is immediately surrounded by parenthesis.
*
* @param int $startPos Start position
* @param int $endPos End position
*/
public function haveParens(int $startPos, int $endPos): bool {
return $this->haveTokenImmediatelyBefore($startPos, '(')
&& $this->haveTokenImmediatelyAfter($endPos, ')');
}
/**
* Whether the given position is immediately surrounded by braces.
*
* @param int $startPos Start position
* @param int $endPos End position
*/
public function haveBraces(int $startPos, int $endPos): bool {
return ($this->haveTokenImmediatelyBefore($startPos, '{')
|| $this->haveTokenImmediatelyBefore($startPos, T_CURLY_OPEN))
&& $this->haveTokenImmediatelyAfter($endPos, '}');
}
/**
* Check whether the position is directly preceded by a certain token type.
*
* During this check whitespace and comments are skipped.
*
* @param int $pos Position before which the token should occur
* @param int|string $expectedTokenType Token to check for
*
* @return bool Whether the expected token was found
*/
public function haveTokenImmediatelyBefore(int $pos, $expectedTokenType): bool {
$tokens = $this->tokens;
$pos--;
for (; $pos >= 0; $pos--) {
$token = $tokens[$pos];
if ($token->is($expectedTokenType)) {
return true;
}
if (!$token->isIgnorable()) {
break;
}
}
return false;
}
/**
* Check whether the position is directly followed by a certain token type.
*
* During this check whitespace and comments are skipped.
*
* @param int $pos Position after which the token should occur
* @param int|string $expectedTokenType Token to check for
*
* @return bool Whether the expected token was found
*/
public function haveTokenImmediatelyAfter(int $pos, $expectedTokenType): bool {
$tokens = $this->tokens;
$pos++;
for ($c = \count($tokens); $pos < $c; $pos++) {
$token = $tokens[$pos];
if ($token->is($expectedTokenType)) {
return true;
}
if (!$token->isIgnorable()) {
break;
}
}
return false;
}
/** @param int|string|(int|string)[] $skipTokenType */
public function skipLeft(int $pos, $skipTokenType): int {
$tokens = $this->tokens;
$pos = $this->skipLeftWhitespace($pos);
if ($skipTokenType === \T_WHITESPACE) {
return $pos;
}
if (!$tokens[$pos]->is($skipTokenType)) {
// Shouldn't happen. The skip token MUST be there
throw new \Exception('Encountered unexpected token');
}
$pos--;
return $this->skipLeftWhitespace($pos);
}
/** @param int|string|(int|string)[] $skipTokenType */
public function skipRight(int $pos, $skipTokenType): int {
$tokens = $this->tokens;
$pos = $this->skipRightWhitespace($pos);
if ($skipTokenType === \T_WHITESPACE) {
return $pos;
}
if (!$tokens[$pos]->is($skipTokenType)) {
// Shouldn't happen. The skip token MUST be there
throw new \Exception('Encountered unexpected token');
}
$pos++;
return $this->skipRightWhitespace($pos);
}
/**
* Return first non-whitespace token position smaller or equal to passed position.
*
* @param int $pos Token position
* @return int Non-whitespace token position
*/
public function skipLeftWhitespace(int $pos): int {
$tokens = $this->tokens;
for (; $pos >= 0; $pos--) {
if (!$tokens[$pos]->isIgnorable()) {
break;
}
}
return $pos;
}
/**
* Return first non-whitespace position greater or equal to passed position.
*
* @param int $pos Token position
* @return int Non-whitespace token position
*/
public function skipRightWhitespace(int $pos): int {
$tokens = $this->tokens;
for ($count = \count($tokens); $pos < $count; $pos++) {
if (!$tokens[$pos]->isIgnorable()) {
break;
}
}
return $pos;
}
/** @param int|string|(int|string)[] $findTokenType */
public function findRight(int $pos, $findTokenType): int {
$tokens = $this->tokens;
for ($count = \count($tokens); $pos < $count; $pos++) {
if ($tokens[$pos]->is($findTokenType)) {
return $pos;
}
}
return -1;
}
/**
* Whether the given position range contains a certain token type.
*
* @param int $startPos Starting position (inclusive)
* @param int $endPos Ending position (exclusive)
* @param int|string $tokenType Token type to look for
* @return bool Whether the token occurs in the given range
*/
public function haveTokenInRange(int $startPos, int $endPos, $tokenType): bool {
$tokens = $this->tokens;
for ($pos = $startPos; $pos < $endPos; $pos++) {
if ($tokens[$pos]->is($tokenType)) {
return true;
}
}
return false;
}
public function haveTagInRange(int $startPos, int $endPos): bool {
return $this->haveTokenInRange($startPos, $endPos, \T_OPEN_TAG)
|| $this->haveTokenInRange($startPos, $endPos, \T_CLOSE_TAG);
}
/**
* Get indentation before token position.
*
* @param int $pos Token position
*
* @return int Indentation depth (in spaces)
*/
public function getIndentationBefore(int $pos): int {
return $this->indentMap[$pos];
}
/**
* Get the code corresponding to a token offset range, optionally adjusted for indentation.
*
* @param int $from Token start position (inclusive)
* @param int $to Token end position (exclusive)
* @param int $indent By how much the code should be indented (can be negative as well)
*
* @return string Code corresponding to token range, adjusted for indentation
*/
public function getTokenCode(int $from, int $to, int $indent): string {
$tokens = $this->tokens;
$result = '';
for ($pos = $from; $pos < $to; $pos++) {
$token = $tokens[$pos];
$id = $token->id;
$text = $token->text;
if ($id === \T_CONSTANT_ENCAPSED_STRING || $id === \T_ENCAPSED_AND_WHITESPACE) {
$result .= $text;
} else {
// TODO Handle non-space indentation
if ($indent < 0) {
$result .= str_replace("\n" . str_repeat(" ", -$indent), "\n", $text);
} elseif ($indent > 0) {
$result .= str_replace("\n", "\n" . str_repeat(" ", $indent), $text);
} else {
$result .= $text;
}
}
}
return $result;
}
/**
* Precalculate the indentation at every token position.
*
* @return int[] Token position to indentation map
*/
private function calcIndentMap(): array {
$indentMap = [];
$indent = 0;
foreach ($this->tokens as $i => $token) {
$indentMap[] = $indent;
if ($token->id === \T_WHITESPACE) {
$content = $token->text;
$newlinePos = \strrpos($content, "\n");
if (false !== $newlinePos) {
$indent = \strlen($content) - $newlinePos - 1;
} elseif ($i === 1 && $this->tokens[0]->id === \T_OPEN_TAG &&
$this->tokens[0]->text[\strlen($this->tokens[0]->text) - 1] === "\n") {
// Special case: Newline at the end of opening tag followed by whitespace.
$indent = \strlen($content);
}
}
}
// Add a sentinel for one past end of the file
$indentMap[] = $indent;
return $indentMap;
}
}