Decoder::decodeFastString() - Code Metrics - Inspection of "Streamlined Decoder::readDigits() for common paths" - s9e/Bencode - Measure and Improve Code Quality continuously with Scrutinizer

Passed

Push — master ( f536e5...5fa974 )

by Josh

created 2023-05-28 22:36 UTC

Decoder::decodeFastString() A

↳ Parent: Decoder

Complexity

Conditions	2
Paths	2

Size

Total Lines	10
Code Lines	4

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	5
CRAP Score	2

Importance

Changes

Metric	Value
cc	2
eloc	4
nc	2
nop	3
dl	0
loc	10
ccs	5
cts	5
cp	1
crap	2
rs	10
c	0
b	0
f	0

<?php declare(strict_types=1);

/**
* @package   s9e\Bencode
* @copyright Copyright (c) The s9e authors
* @license   http://www.opensource.org/licenses/mit-license.php The MIT License
*/
namespace s9e\Bencode;

use ArrayObject;
use Throwable;
use TypeError;
use const PHP_INT_MAX, PHP_INT_MIN, false;
use function is_float, str_contains, strcmp, strlen, strspn, substr, substr_compare;
use s9e\Bencode\Exceptions\ComplianceError;
use s9e\Bencode\Exceptions\DecodingException;

class Decoder
{
	/**
	* @var int Length of the bencoded string
	*/
	protected int $len;

	/**
	* @var int Safe rightmost boundary
	*/
	protected int $max;

	/**
	* @var int Position of the cursor while decoding
	*/
	protected int $offset = 0;

	/**
	* @param string $bencoded Bencoded string to decode
	*/
	public static function decode(string $bencoded): ArrayObject|array|int|string
	{
		$decoder = new static($bencoded);
		try
		{
			$value = $decoder->decodeAnything();
		}
		catch (TypeError $e)
		{
			throw static::convertTypeError($e, $decoder->offset);
		}

		$decoder->checkCursorPosition();

		return $value;
	}

	/**
	* @param string $bencoded Bencoded string being decoded
	*/
	protected function __construct(protected readonly string $bencoded)
	{
		$this->len = strlen($bencoded);
		$this->max = $this->getSafeBoundary();

		$this->checkBoundary();
	}

	/**
	* Cast given string as an integer and check for clamping
	*/
	protected function castInteger(string $string, int $clamp): int
	{
		$value = (int) $string;
		if ($value === $clamp && is_float(+$string))
		{
			throw new DecodingException('Integer overflow', $this->offset - 1 - strlen($string));
		}

		return $value;
	}

	protected function checkBoundary(): void
	{
		if ($this->max < 1)
		{
			throw match (substr($this->bencoded, 0, 1))
			{
				''       => new DecodingException('Premature end of data', 0),
				'-'      => new DecodingException('Illegal character',     0),
				'e'      => new DecodingException('Illegal character',     0),
				default  => new DecodingException('Premature end of data', $this->len - 1)
			};
		}
	}

	/**
	* Check the cursor's position after decoding is done
	*/
	protected function checkCursorPosition(): void
	{
		if ($this->offset === $this->len)
		{
			return;
		}
		if ($this->offset > $this->len)
		{
			throw new DecodingException('Premature end of data', $this->len - 1);
		}

		$this->complianceError('Superfluous content', $this->offset);
	}

	protected function complianceError(string $message, int $offset): void
	{
		throw new ComplianceError($message, $offset);
	}

	protected static function convertTypeError(TypeError $e, int $offset): Throwable
	{
		// A type error can occur in decodeString() if the string length exceeds an int
		$frame  = $e->getTrace()[0];
		$caller = $frame['class'] . $frame['type'] . $frame['function'];
		if ($caller === __CLASS__ . '->decodeString')
		{
			return new DecodingException('String length overflow', $offset - 1);
		}

		// Return any other error as-is
		return $e;
	}

	protected function decodeAnything(): ArrayObject|array|int|string
	{
		return match ($this->bencoded[$this->offset])
		{
			'i'     => $this->decodeInteger(),
			'd'     => $this->decodeDictionary(),
			'l'     => $this->decodeList(),
			default => $this->decodeString()
		};
	}

	protected function decodeDictionary(): ArrayObject
	{
		$values  = [];
		$lastKey = null;

		++$this->offset;
		while ($this->offset <= $this->max)
		{
			$c = $this->bencoded[$this->offset];
			if ($c === 'e')
			{
				++$this->offset;

				return new ArrayObject($values, ArrayObject::ARRAY_AS_PROPS);
			}

			// Quickly match the most common keys found in dictionaries
			$key = match ($c)
			{
				'4'     => $this->decodeFastString('4:path',   6, 'path'),
				'6'     => $this->decodeFastString('6:length', 8, 'length'),
				default => $this->decodeString()
			};
			if (isset($lastKey) && strcmp($key, $lastKey) <= 0)
			{
				$this->dictionaryComplianceError($key, $lastKey);
			}
			if ($this->offset > $this->max)
			{
				break;
			}
			$values[$key] = $this->decodeAnything();
			$lastKey      = $key;
		}

		throw new DecodingException('Premature end of data', $this->len - 1);
	}

	/**
	* @param string $match Bencoded string to match
	* @param int    $len   Length of the bencoded string
	* @param string $value String value to return if the string matches
	*/
	protected function decodeFastString(string $match, int $len, string $value): string
	{
		if (substr_compare($this->bencoded, $match, $this->offset, $len, false) === 0)
		{
			$this->offset += $len;

			return $value;
		}

		return $this->decodeString();
	}

	protected function decodeInteger(): int
	{
		if ($this->bencoded[++$this->offset] === '-')
		{
			if ($this->bencoded[++$this->offset] === '0')
			{
				$this->complianceError('Illegal character', $this->offset);
			}

			$clamp  = PHP_INT_MIN;
			$string = '-' . $this->readDigits('e');
		}
		else
		{
			$clamp  = PHP_INT_MAX;
			$string = $this->readDigits('e');
		}

		return $this->castInteger($string, $clamp);
	}

	protected function decodeList(): array
	{
		++$this->offset;

		$list = [];
		while ($this->offset <= $this->max)
		{
			if ($this->bencoded[$this->offset] === 'e')
			{
				++$this->offset;

				return $list;
			}

			$list[] = $this->decodeAnything();
		}

		throw new DecodingException('Premature end of data', $this->len - 1);
	}

	protected function decodeString(): string
	{
		$len = (int) $this->readDigits(':');
		$string = substr($this->bencoded, $this->offset, $len);
		$this->offset += $len;

		return $string;
	}

	protected function dictionaryComplianceError(string $key, string $lastKey): void
	{
		// Compute the offset of the start of the string used as key
		$offset = $this->offset - strlen(strlen($key) . ':') - strlen($key);

		$msg = ($key === $lastKey) ? 'Duplicate' : 'Out of order';
		$this->complianceError($msg . " dictionary entry '" . $key . "'", $offset);
	}

	protected function digitException(): DecodingException
	{
		return (str_contains('0123456789', $this->bencoded[$this->offset]))
		     ? new ComplianceError('Illegal character', $this->offset)
		     : new DecodingException('Illegal character', $this->offset);
	}

	/**
	* Return the rightmost boundary to the last safe character that can start a value
	*
	* Will rewind the boundary to skip the rightmost digits, optionally preceded by "i" or "i-"
	*/
	protected function getSafeBoundary(): int
	{
		$boundary = $this->len - 1;
		do
		{
			$c = substr($this->bencoded, $boundary, 1);
		}
		while (str_contains('0123456789', $c) && --$boundary >= 0);

		return match ($c)
		{
			'-'     => $boundary - 2,
			'i'     => $boundary - 1,
			default => $boundary
		};
	}

	protected function readDigits(string $terminator): string
	{
		if ($this->bencoded[$this->offset] === '0')
		{
			++$this->offset;
			$string = '0';
		}
		else
		{
			// Digits sorted by decreasing frequency as observed on a random sample of torrent files
			$spn = strspn($this->bencoded, '1463720859', $this->offset);
			if ($spn === 0)
			{
				throw new DecodingException('Illegal character', $this->offset);
			}
			$string = substr($this->bencoded, $this->offset, $spn);
			$this->offset += $spn;
		}

		if ($this->bencoded[$this->offset] !== $terminator)
		{
			throw $this->digitException();
		}
		++$this->offset;

		return $string;
	}
}

1		<?php declare(strict_types=1);
2
3		/**
4		* @package s9e\Bencode
5		* @copyright Copyright (c) The s9e authors
6		* @license http://www.opensource.org/licenses/mit-license.php The MIT License
7		*/
8		namespace s9e\Bencode;
9
10		use ArrayObject;
11		use Throwable;
12		use TypeError;
13		use const PHP_INT_MAX, PHP_INT_MIN, false;
14		use function is_float, str_contains, strcmp, strlen, strspn, substr, substr_compare;
15		use s9e\Bencode\Exceptions\ComplianceError;
16		use s9e\Bencode\Exceptions\DecodingException;
17
18		class Decoder
19		{
20		/**
21		* @var int Length of the bencoded string
22		*/
23		protected int $len;
24
25		/**
26		* @var int Safe rightmost boundary
27		*/
28		protected int $max;
29
30		/**
31		* @var int Position of the cursor while decoding
32		*/
33		protected int $offset = 0;
34
35		/**
36		* @param string $bencoded Bencoded string to decode
37		*/
38	92	public static function decode(string $bencoded): ArrayObject\|array\|int\|string
39		{
40	92	$decoder = new static($bencoded);
41		try
42		{
43	79	$value = $decoder->decodeAnything();
44		}
45	36	catch (TypeError $e)
46		{
47	3	throw static::convertTypeError($e, $decoder->offset);
48		}
49
50	43	$decoder->checkCursorPosition();
51
52	35	return $value;
53		}
54
55		/**
56		* @param string $bencoded Bencoded string being decoded
57		*/
58	92	protected function __construct(protected readonly string $bencoded)
59		{
60	92	$this->len = strlen($bencoded);
61	92	$this->max = $this->getSafeBoundary();
62
63	92	$this->checkBoundary();
64		}
65
66		/**
67		* Cast given string as an integer and check for clamping
68		*/
69	27	protected function castInteger(string $string, int $clamp): int
70		{
71	27	$value = (int) $string;
72	27	if ($value === $clamp && is_float(+$string))
73		{
74	2	throw new DecodingException('Integer overflow', $this->offset - 1 - strlen($string));
75		}
76
77	25	return $value;
78		}
79
80	92	protected function checkBoundary(): void
81		{
82	92	if ($this->max < 1)
83		{
84	13	throw match (substr($this->bencoded, 0, 1))
85		{
86	1	'' => new DecodingException('Premature end of data', 0),
87	1	'-' => new DecodingException('Illegal character', 0),
88	1	'e' => new DecodingException('Illegal character', 0),
89	10	default => new DecodingException('Premature end of data', $this->len - 1)
90		};
91		}
92		}
93
94		/**
95		* Check the cursor's position after decoding is done
96		*/
97	43	protected function checkCursorPosition(): void
98		{
99	43	if ($this->offset === $this->len)
100		{
101	32	return;
102		}
103	11	if ($this->offset > $this->len)
104		{
105	3	throw new DecodingException('Premature end of data', $this->len - 1);
106		}
107
108	8	$this->complianceError('Superfluous content', $this->offset);
109		}
110
111	11	protected function complianceError(string $message, int $offset): void
112		{
113	11	throw new ComplianceError($message, $offset);
114		}
115
116	3	protected static function convertTypeError(TypeError $e, int $offset): Throwable
117		{
118		// A type error can occur in decodeString() if the string length exceeds an int
119	3	$frame = $e->getTrace()[0];
120	3	$caller = $frame['class'] . $frame['type'] . $frame['function'];
121	3	if ($caller === __CLASS__ . '->decodeString')
122		{
123	1	return new DecodingException('String length overflow', $offset - 1);
124		}
125
126		// Return any other error as-is
127	2	return $e;
128		}
129
130	79	protected function decodeAnything(): ArrayObject\|array\|int\|string
131		{
132	79	return match ($this->bencoded[$this->offset])
133	79	{
134	79	'i' => $this->decodeInteger(),
135	79	'd' => $this->decodeDictionary(),
136	79	'l' => $this->decodeList(),
137	79	default => $this->decodeString()
138	79	};
139		}
140
141	24	protected function decodeDictionary(): ArrayObject
142		{
143	24	$values = [];
144	24	$lastKey = null;
145
146	24	++$this->offset;
147	23	while ($this->offset <= $this->max)
148		{
149	23	$c = $this->bencoded[$this->offset];
150	23	if ($c === 'e')
151		{
152	13	++$this->offset;
153
154	13	return new ArrayObject($values, ArrayObject::ARRAY_AS_PROPS);
155		}
156
157		// Quickly match the most common keys found in dictionaries
158	22	$key = match ($c)
159	22	{
160	22	'4' => $this->decodeFastString('4:path', 6, 'path'),
161	22	'6' => $this->decodeFastString('6:length', 8, 'length'),
162	22	default => $this->decodeString()
163	22	};
164	21	if (isset($lastKey) && strcmp($key, $lastKey) <= 0)
165		{
166	8	$this->dictionaryComplianceError($key, $lastKey);
167		}
168	21	if ($this->offset > $this->max)
169		{
170	2	break;
171		}
172	19	$values[$key] = $this->decodeAnything();
173	16	$lastKey = $key;
174		}
175
176	2	throw new DecodingException('Premature end of data', $this->len - 1);
177		}
178
179		/**
180		* @param string $match Bencoded string to match
181		* @param int $len Length of the bencoded string
182		* @param string $value String value to return if the string matches
183		*/
184	4	protected function decodeFastString(string $match, int $len, string $value): string
185		{
186	4	if (substr_compare($this->bencoded, $match, $this->offset, $len, false) === 0)
187		{
188	2	$this->offset += $len;
189
190	2	return $value;
191		}
192
193	3	return $this->decodeString();
194		}
195
196	36	protected function decodeInteger(): int
197		{
198	36	if ($this->bencoded[++$this->offset] === '-')
199		{
200	8	if ($this->bencoded[++$this->offset] === '0')
201		{
202	4	$this->complianceError('Illegal character', $this->offset);
203		}
204
205	6	$clamp = PHP_INT_MIN;
206	6	$string = '-' . $this->readDigits('e');
207		}
208		else
209		{
210	28	$clamp = PHP_INT_MAX;
211	28	$string = $this->readDigits('e');
212		}
213
214	27	return $this->castInteger($string, $clamp);
215		}
216
217	16	protected function decodeList(): array
218		{
219	16	++$this->offset;
220
221	16	$list = [];
222	16	while ($this->offset <= $this->max)
223		{
224	15	if ($this->bencoded[$this->offset] === 'e')
225		{
226	8	++$this->offset;
227
228	8	return $list;
229		}
230
231	12	$list[] = $this->decodeAnything();
232		}
233
234	5	throw new DecodingException('Premature end of data', $this->len - 1);
235		}
236
237	47	protected function decodeString(): string
238		{
239	47	$len = (int) $this->readDigits(':');
240	38	$string = substr($this->bencoded, $this->offset, $len);
241	38	$this->offset += $len;
242
243	37	return $string;
244		}
245
246	4	protected function dictionaryComplianceError(string $key, string $lastKey): void
247		{
248		// Compute the offset of the start of the string used as key
249	4	$offset = $this->offset - strlen(strlen($key) . ':') - strlen($key);
250
251	4	$msg = ($key === $lastKey) ? 'Duplicate' : 'Out of order';
252	4	$this->complianceError($msg . " dictionary entry '" . $key . "'", $offset);
253		}
254
255	11	protected function digitException(): DecodingException
256		{
257	11	return (str_contains('0123456789', $this->bencoded[$this->offset]))
258	3	? new ComplianceError('Illegal character', $this->offset)
259	11	: new DecodingException('Illegal character', $this->offset);
260		}
261
262		/**
263		* Return the rightmost boundary to the last safe character that can start a value
264		*
265		* Will rewind the boundary to skip the rightmost digits, optionally preceded by "i" or "i-"
266		*/
267	92	protected function getSafeBoundary(): int
268		{
269	92	$boundary = $this->len - 1;
270		do
271		{
272	92	$c = substr($this->bencoded, $boundary, 1);
273		}
274	92	while (str_contains('0123456789', $c) && --$boundary >= 0);
275
276	92	return match ($c)
277	92	{
278	92	'-' => $boundary - 2,
279	92	'i' => $boundary - 1,
280	92	default => $boundary
281	92	};
282		}
283
284	67	protected function readDigits(string $terminator): string
285		{
286	67	if ($this->bencoded[$this->offset] === '0')
287		{
288	18	++$this->offset;
289	18	$string = '0';
290		}
291		else
292		{
293		// Digits sorted by decreasing frequency as observed on a random sample of torrent files
294	59	$spn = strspn($this->bencoded, '1463720859', $this->offset);
295	59	if ($spn === 0)
296		{
297	7	throw new DecodingException('Illegal character', $this->offset);
298		}
299	54	$string = substr($this->bencoded, $this->offset, $spn);
300	54	$this->offset += $spn;
301		}
302
303	62	if ($this->bencoded[$this->offset] !== $terminator)
304		{
305	11	throw $this->digitException();
306		}
307	51	++$this->offset;
308
309	51	return $string;
310		}
311		}

s9e / Bencode

Push — master ( f536e5...5fa974 )

Decoder::decodeFastString() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like