Fixed Broken Shit

This commit is contained in:
2022-10-30 14:32:20 -07:00
parent 7335796263
commit 4dabf5a6bf
635 changed files with 74885 additions and 17688 deletions

View File

@ -0,0 +1,373 @@
<?php
declare(strict_types=1);
namespace MaxMind\Db;
use ArgumentCountError;
use BadMethodCallException;
use Exception;
use InvalidArgumentException;
use MaxMind\Db\Reader\Decoder;
use MaxMind\Db\Reader\InvalidDatabaseException;
use MaxMind\Db\Reader\Metadata;
use MaxMind\Db\Reader\Util;
use UnexpectedValueException;
/**
* Instances of this class provide a reader for the MaxMind DB format. IP
* addresses can be looked up using the get method.
*/
class Reader
{
/**
* @var int
*/
private static $DATA_SECTION_SEPARATOR_SIZE = 16;
/**
* @var string
*/
private static $METADATA_START_MARKER = "\xAB\xCD\xEFMaxMind.com";
/**
* @var int
*/
private static $METADATA_START_MARKER_LENGTH = 14;
/**
* @var int
*/
private static $METADATA_MAX_SIZE = 131072; // 128 * 1024 = 128KiB
/**
* @var Decoder
*/
private $decoder;
/**
* @var resource
*/
private $fileHandle;
/**
* @var int
*/
private $fileSize;
/**
* @var int
*/
private $ipV4Start;
/**
* @var Metadata
*/
private $metadata;
/**
* Constructs a Reader for the MaxMind DB format. The file passed to it must
* be a valid MaxMind DB file such as a GeoIp2 database file.
*
* @param string $database
* the MaxMind DB file to use
*
* @throws InvalidArgumentException for invalid database path or unknown arguments
* @throws InvalidDatabaseException
* if the database is invalid or there is an error reading
* from it
*/
public function __construct(string $database)
{
if (\func_num_args() !== 1) {
throw new ArgumentCountError(
sprintf('%s() expects exactly 1 parameter, %d given', __METHOD__, \func_num_args())
);
}
$fileHandle = @fopen($database, 'rb');
if ($fileHandle === false) {
throw new InvalidArgumentException(
"The file \"$database\" does not exist or is not readable."
);
}
$this->fileHandle = $fileHandle;
$fileSize = @filesize($database);
if ($fileSize === false) {
throw new UnexpectedValueException(
"Error determining the size of \"$database\"."
);
}
$this->fileSize = $fileSize;
$start = $this->findMetadataStart($database);
$metadataDecoder = new Decoder($this->fileHandle, $start);
[$metadataArray] = $metadataDecoder->decode($start);
$this->metadata = new Metadata($metadataArray);
$this->decoder = new Decoder(
$this->fileHandle,
$this->metadata->searchTreeSize + self::$DATA_SECTION_SEPARATOR_SIZE
);
$this->ipV4Start = $this->ipV4StartNode();
}
/**
* Retrieves the record for the IP address.
*
* @param string $ipAddress
* the IP address to look up
*
* @throws BadMethodCallException if this method is called on a closed database
* @throws InvalidArgumentException if something other than a single IP address is passed to the method
* @throws InvalidDatabaseException
* if the database is invalid or there is an error reading
* from it
*
* @return mixed the record for the IP address
*/
public function get(string $ipAddress)
{
if (\func_num_args() !== 1) {
throw new ArgumentCountError(
sprintf('%s() expects exactly 1 parameter, %d given', __METHOD__, \func_num_args())
);
}
[$record] = $this->getWithPrefixLen($ipAddress);
return $record;
}
/**
* Retrieves the record for the IP address and its associated network prefix length.
*
* @param string $ipAddress
* the IP address to look up
*
* @throws BadMethodCallException if this method is called on a closed database
* @throws InvalidArgumentException if something other than a single IP address is passed to the method
* @throws InvalidDatabaseException
* if the database is invalid or there is an error reading
* from it
*
* @return array an array where the first element is the record and the
* second the network prefix length for the record
*/
public function getWithPrefixLen(string $ipAddress): array
{
if (\func_num_args() !== 1) {
throw new ArgumentCountError(
sprintf('%s() expects exactly 1 parameter, %d given', __METHOD__, \func_num_args())
);
}
if (!\is_resource($this->fileHandle)) {
throw new BadMethodCallException(
'Attempt to read from a closed MaxMind DB.'
);
}
[$pointer, $prefixLen] = $this->findAddressInTree($ipAddress);
if ($pointer === 0) {
return [null, $prefixLen];
}
return [$this->resolveDataPointer($pointer), $prefixLen];
}
private function findAddressInTree(string $ipAddress): array
{
$packedAddr = @inet_pton($ipAddress);
if ($packedAddr === false) {
throw new InvalidArgumentException(
"The value \"$ipAddress\" is not a valid IP address."
);
}
$rawAddress = unpack('C*', $packedAddr);
$bitCount = \count($rawAddress) * 8;
// The first node of the tree is always node 0, at the beginning of the
// value
$node = 0;
$metadata = $this->metadata;
// Check if we are looking up an IPv4 address in an IPv6 tree. If this
// is the case, we can skip over the first 96 nodes.
if ($metadata->ipVersion === 6) {
if ($bitCount === 32) {
$node = $this->ipV4Start;
}
} elseif ($metadata->ipVersion === 4 && $bitCount === 128) {
throw new InvalidArgumentException(
"Error looking up $ipAddress. You attempted to look up an"
. ' IPv6 address in an IPv4-only database.'
);
}
$nodeCount = $metadata->nodeCount;
for ($i = 0; $i < $bitCount && $node < $nodeCount; ++$i) {
$tempBit = 0xFF & $rawAddress[($i >> 3) + 1];
$bit = 1 & ($tempBit >> 7 - ($i % 8));
$node = $this->readNode($node, $bit);
}
if ($node === $nodeCount) {
// Record is empty
return [0, $i];
}
if ($node > $nodeCount) {
// Record is a data pointer
return [$node, $i];
}
throw new InvalidDatabaseException(
'Invalid or corrupt database. Maximum search depth reached without finding a leaf node'
);
}
private function ipV4StartNode(): int
{
// If we have an IPv4 database, the start node is the first node
if ($this->metadata->ipVersion === 4) {
return 0;
}
$node = 0;
for ($i = 0; $i < 96 && $node < $this->metadata->nodeCount; ++$i) {
$node = $this->readNode($node, 0);
}
return $node;
}
private function readNode(int $nodeNumber, int $index): int
{
$baseOffset = $nodeNumber * $this->metadata->nodeByteSize;
switch ($this->metadata->recordSize) {
case 24:
$bytes = Util::read($this->fileHandle, $baseOffset + $index * 3, 3);
[, $node] = unpack('N', "\x00" . $bytes);
return $node;
case 28:
$bytes = Util::read($this->fileHandle, $baseOffset + 3 * $index, 4);
if ($index === 0) {
$middle = (0xF0 & \ord($bytes[3])) >> 4;
} else {
$middle = 0x0F & \ord($bytes[0]);
}
[, $node] = unpack('N', \chr($middle) . substr($bytes, $index, 3));
return $node;
case 32:
$bytes = Util::read($this->fileHandle, $baseOffset + $index * 4, 4);
[, $node] = unpack('N', $bytes);
return $node;
default:
throw new InvalidDatabaseException(
'Unknown record size: '
. $this->metadata->recordSize
);
}
}
/**
* @return mixed
*/
private function resolveDataPointer(int $pointer)
{
$resolved = $pointer - $this->metadata->nodeCount
+ $this->metadata->searchTreeSize;
if ($resolved >= $this->fileSize) {
throw new InvalidDatabaseException(
"The MaxMind DB file's search tree is corrupt"
);
}
[$data] = $this->decoder->decode($resolved);
return $data;
}
/*
* This is an extremely naive but reasonably readable implementation. There
* are much faster algorithms (e.g., Boyer-Moore) for this if speed is ever
* an issue, but I suspect it won't be.
*/
private function findMetadataStart(string $filename): int
{
$handle = $this->fileHandle;
$fstat = fstat($handle);
$fileSize = $fstat['size'];
$marker = self::$METADATA_START_MARKER;
$markerLength = self::$METADATA_START_MARKER_LENGTH;
$minStart = $fileSize - min(self::$METADATA_MAX_SIZE, $fileSize);
for ($offset = $fileSize - $markerLength; $offset >= $minStart; --$offset) {
if (fseek($handle, $offset) !== 0) {
break;
}
$value = fread($handle, $markerLength);
if ($value === $marker) {
return $offset + $markerLength;
}
}
throw new InvalidDatabaseException(
"Error opening database file ($filename). " .
'Is this a valid MaxMind DB file?'
);
}
/**
* @throws InvalidArgumentException if arguments are passed to the method
* @throws BadMethodCallException if the database has been closed
*
* @return Metadata object for the database
*/
public function metadata(): Metadata
{
if (\func_num_args()) {
throw new ArgumentCountError(
sprintf('%s() expects exactly 0 parameters, %d given', __METHOD__, \func_num_args())
);
}
// Not technically required, but this makes it consistent with
// C extension and it allows us to change our implementation later.
if (!\is_resource($this->fileHandle)) {
throw new BadMethodCallException(
'Attempt to read from a closed MaxMind DB.'
);
}
return clone $this->metadata;
}
/**
* Closes the MaxMind DB and returns resources to the system.
*
* @throws Exception
* if an I/O error occurs
*/
public function close(): void
{
if (\func_num_args()) {
throw new ArgumentCountError(
sprintf('%s() expects exactly 0 parameters, %d given', __METHOD__, \func_num_args())
);
}
if (!\is_resource($this->fileHandle)) {
throw new BadMethodCallException(
'Attempt to close a closed MaxMind DB.'
);
}
fclose($this->fileHandle);
}
}

View File

@ -0,0 +1,374 @@
<?php
declare(strict_types=1);
namespace MaxMind\Db\Reader;
// @codingStandardsIgnoreLine
use RuntimeException;
class Decoder
{
/**
* @var resource
*/
private $fileStream;
/**
* @var int
*/
private $pointerBase;
/**
* @var float
*/
private $pointerBaseByteSize;
/**
* This is only used for unit testing.
*
* @var bool
*/
private $pointerTestHack;
/**
* @var bool
*/
private $switchByteOrder;
private const _EXTENDED = 0;
private const _POINTER = 1;
private const _UTF8_STRING = 2;
private const _DOUBLE = 3;
private const _BYTES = 4;
private const _UINT16 = 5;
private const _UINT32 = 6;
private const _MAP = 7;
private const _INT32 = 8;
private const _UINT64 = 9;
private const _UINT128 = 10;
private const _ARRAY = 11;
private const _CONTAINER = 12;
private const _END_MARKER = 13;
private const _BOOLEAN = 14;
private const _FLOAT = 15;
/**
* @param resource $fileStream
*/
public function __construct(
$fileStream,
int $pointerBase = 0,
bool $pointerTestHack = false
) {
$this->fileStream = $fileStream;
$this->pointerBase = $pointerBase;
$this->pointerBaseByteSize = $pointerBase > 0 ? log($pointerBase, 2) / 8 : 0;
$this->pointerTestHack = $pointerTestHack;
$this->switchByteOrder = $this->isPlatformLittleEndian();
}
public function decode(int $offset): array
{
$ctrlByte = \ord(Util::read($this->fileStream, $offset, 1));
++$offset;
$type = $ctrlByte >> 5;
// Pointers are a special case, we don't read the next $size bytes, we
// use the size to determine the length of the pointer and then follow
// it.
if ($type === self::_POINTER) {
[$pointer, $offset] = $this->decodePointer($ctrlByte, $offset);
// for unit testing
if ($this->pointerTestHack) {
return [$pointer];
}
[$result] = $this->decode($pointer);
return [$result, $offset];
}
if ($type === self::_EXTENDED) {
$nextByte = \ord(Util::read($this->fileStream, $offset, 1));
$type = $nextByte + 7;
if ($type < 8) {
throw new InvalidDatabaseException(
'Something went horribly wrong in the decoder. An extended type '
. 'resolved to a type number < 8 ('
. $type
. ')'
);
}
++$offset;
}
[$size, $offset] = $this->sizeFromCtrlByte($ctrlByte, $offset);
return $this->decodeByType($type, $offset, $size);
}
private function decodeByType(int $type, int $offset, int $size): array
{
switch ($type) {
case self::_MAP:
return $this->decodeMap($size, $offset);
case self::_ARRAY:
return $this->decodeArray($size, $offset);
case self::_BOOLEAN:
return [$this->decodeBoolean($size), $offset];
}
$newOffset = $offset + $size;
$bytes = Util::read($this->fileStream, $offset, $size);
switch ($type) {
case self::_BYTES:
case self::_UTF8_STRING:
return [$bytes, $newOffset];
case self::_DOUBLE:
$this->verifySize(8, $size);
return [$this->decodeDouble($bytes), $newOffset];
case self::_FLOAT:
$this->verifySize(4, $size);
return [$this->decodeFloat($bytes), $newOffset];
case self::_INT32:
return [$this->decodeInt32($bytes, $size), $newOffset];
case self::_UINT16:
case self::_UINT32:
case self::_UINT64:
case self::_UINT128:
return [$this->decodeUint($bytes, $size), $newOffset];
default:
throw new InvalidDatabaseException(
'Unknown or unexpected type: ' . $type
);
}
}
private function verifySize(int $expected, int $actual): void
{
if ($expected !== $actual) {
throw new InvalidDatabaseException(
"The MaxMind DB file's data section contains bad data (unknown data type or corrupt data)"
);
}
}
private function decodeArray(int $size, int $offset): array
{
$array = [];
for ($i = 0; $i < $size; ++$i) {
[$value, $offset] = $this->decode($offset);
$array[] = $value;
}
return [$array, $offset];
}
private function decodeBoolean(int $size): bool
{
return $size !== 0;
}
private function decodeDouble(string $bytes): float
{
// This assumes IEEE 754 doubles, but most (all?) modern platforms
// use them.
[, $double] = unpack('E', $bytes);
return $double;
}
private function decodeFloat(string $bytes): float
{
// This assumes IEEE 754 floats, but most (all?) modern platforms
// use them.
[, $float] = unpack('G', $bytes);
return $float;
}
private function decodeInt32(string $bytes, int $size): int
{
switch ($size) {
case 0:
return 0;
case 1:
case 2:
case 3:
$bytes = str_pad($bytes, 4, "\x00", \STR_PAD_LEFT);
break;
case 4:
break;
default:
throw new InvalidDatabaseException(
"The MaxMind DB file's data section contains bad data (unknown data type or corrupt data)"
);
}
[, $int] = unpack('l', $this->maybeSwitchByteOrder($bytes));
return $int;
}
private function decodeMap(int $size, int $offset): array
{
$map = [];
for ($i = 0; $i < $size; ++$i) {
[$key, $offset] = $this->decode($offset);
[$value, $offset] = $this->decode($offset);
$map[$key] = $value;
}
return [$map, $offset];
}
private function decodePointer(int $ctrlByte, int $offset): array
{
$pointerSize = (($ctrlByte >> 3) & 0x3) + 1;
$buffer = Util::read($this->fileStream, $offset, $pointerSize);
$offset = $offset + $pointerSize;
switch ($pointerSize) {
case 1:
$packed = \chr($ctrlByte & 0x7) . $buffer;
[, $pointer] = unpack('n', $packed);
$pointer += $this->pointerBase;
break;
case 2:
$packed = "\x00" . \chr($ctrlByte & 0x7) . $buffer;
[, $pointer] = unpack('N', $packed);
$pointer += $this->pointerBase + 2048;
break;
case 3:
$packed = \chr($ctrlByte & 0x7) . $buffer;
// It is safe to use 'N' here, even on 32 bit machines as the
// first bit is 0.
[, $pointer] = unpack('N', $packed);
$pointer += $this->pointerBase + 526336;
break;
case 4:
// We cannot use unpack here as we might overflow on 32 bit
// machines
$pointerOffset = $this->decodeUint($buffer, $pointerSize);
$pointerBase = $this->pointerBase;
if (\PHP_INT_MAX - $pointerBase >= $pointerOffset) {
$pointer = $pointerOffset + $pointerBase;
} else {
throw new RuntimeException(
'The database offset is too large to be represented on your platform.'
);
}
break;
default:
throw new InvalidDatabaseException(
'Unexpected pointer size ' . $pointerSize
);
}
return [$pointer, $offset];
}
// @phpstan-ignore-next-line
private function decodeUint(string $bytes, int $byteLength)
{
if ($byteLength === 0) {
return 0;
}
$integer = 0;
// PHP integers are signed. PHP_INT_SIZE - 1 is the number of
// complete bytes that can be converted to an integer. However,
// we can convert another byte if the leading bit is zero.
$useRealInts = $byteLength <= \PHP_INT_SIZE - 1
|| ($byteLength === \PHP_INT_SIZE && (\ord($bytes[0]) & 0x80) === 0);
for ($i = 0; $i < $byteLength; ++$i) {
$part = \ord($bytes[$i]);
// We only use gmp or bcmath if the final value is too big
if ($useRealInts) {
$integer = ($integer << 8) + $part;
} elseif (\extension_loaded('gmp')) {
$integer = gmp_strval(gmp_add(gmp_mul((string) $integer, '256'), $part));
} elseif (\extension_loaded('bcmath')) {
$integer = bcadd(bcmul((string) $integer, '256'), (string) $part);
} else {
throw new RuntimeException(
'The gmp or bcmath extension must be installed to read this database.'
);
}
}
return $integer;
}
private function sizeFromCtrlByte(int $ctrlByte, int $offset): array
{
$size = $ctrlByte & 0x1F;
if ($size < 29) {
return [$size, $offset];
}
$bytesToRead = $size - 28;
$bytes = Util::read($this->fileStream, $offset, $bytesToRead);
if ($size === 29) {
$size = 29 + \ord($bytes);
} elseif ($size === 30) {
[, $adjust] = unpack('n', $bytes);
$size = 285 + $adjust;
} else {
[, $adjust] = unpack('N', "\x00" . $bytes);
$size = $adjust + 65821;
}
return [$size, $offset + $bytesToRead];
}
private function maybeSwitchByteOrder(string $bytes): string
{
return $this->switchByteOrder ? strrev($bytes) : $bytes;
}
private function isPlatformLittleEndian(): bool
{
$testint = 0x00FF;
$packed = pack('S', $testint);
return $testint === current(unpack('v', $packed));
}
}

View File

@ -0,0 +1,14 @@
<?php
declare(strict_types=1);
namespace MaxMind\Db\Reader;
use Exception;
/**
* This class should be thrown when unexpected data is found in the database.
*/
class InvalidDatabaseException extends Exception
{
}

View File

@ -0,0 +1,112 @@
<?php
declare(strict_types=1);
namespace MaxMind\Db\Reader;
use ArgumentCountError;
/**
* This class provides the metadata for the MaxMind DB file.
*/
class Metadata
{
/**
* This is an unsigned 16-bit integer indicating the major version number
* for the database's binary format.
*
* @var int
*/
public $binaryFormatMajorVersion;
/**
* This is an unsigned 16-bit integer indicating the minor version number
* for the database's binary format.
*
* @var int
*/
public $binaryFormatMinorVersion;
/**
* This is an unsigned 64-bit integer that contains the database build
* timestamp as a Unix epoch value.
*
* @var int
*/
public $buildEpoch;
/**
* This is a string that indicates the structure of each data record
* associated with an IP address. The actual definition of these
* structures is left up to the database creator.
*
* @var string
*/
public $databaseType;
/**
* This key will always point to a map (associative array). The keys of
* that map will be language codes, and the values will be a description
* in that language as a UTF-8 string. May be undefined for some
* databases.
*
* @var array
*/
public $description;
/**
* This is an unsigned 16-bit integer which is always 4 or 6. It indicates
* whether the database contains IPv4 or IPv6 address data.
*
* @var int
*/
public $ipVersion;
/**
* An array of strings, each of which is a language code. A given record
* may contain data items that have been localized to some or all of
* these languages. This may be undefined.
*
* @var array
*/
public $languages;
/**
* @var int
*/
public $nodeByteSize;
/**
* This is an unsigned 32-bit integer indicating the number of nodes in
* the search tree.
*
* @var int
*/
public $nodeCount;
/**
* This is an unsigned 16-bit integer. It indicates the number of bits in a
* record in the search tree. Note that each node consists of two records.
*
* @var int
*/
public $recordSize;
/**
* @var int
*/
public $searchTreeSize;
public function __construct(array $metadata)
{
if (\func_num_args() !== 1) {
throw new ArgumentCountError(
sprintf('%s() expects exactly 1 parameter, %d given', __METHOD__, \func_num_args())
);
}
$this->binaryFormatMajorVersion =
$metadata['binary_format_major_version'];
$this->binaryFormatMinorVersion =
$metadata['binary_format_minor_version'];
$this->buildEpoch = $metadata['build_epoch'];
$this->databaseType = $metadata['database_type'];
$this->languages = $metadata['languages'];
$this->description = $metadata['description'];
$this->ipVersion = $metadata['ip_version'];
$this->nodeCount = $metadata['node_count'];
$this->recordSize = $metadata['record_size'];
$this->nodeByteSize = $this->recordSize / 4;
$this->searchTreeSize = $this->nodeCount * $this->nodeByteSize;
}
}

View File

@ -0,0 +1,32 @@
<?php
declare(strict_types=1);
namespace MaxMind\Db\Reader;
class Util
{
/**
* @param resource $stream
*/
public static function read($stream, int $offset, int $numberOfBytes): string
{
if ($numberOfBytes === 0) {
return '';
}
if (fseek($stream, $offset) === 0) {
$value = fread($stream, $numberOfBytes);
// We check that the number of bytes read is equal to the number
// asked for. We use ftell as getting the length of $value is
// much slower.
if ($value !== false && ftell($stream) - $offset === $numberOfBytes) {
return $value;
}
}
throw new InvalidDatabaseException(
'The MaxMind DB file contains bad data'
);
}
}