<?php

[0] Fix | Delete

[1] Fix | Delete

namespace GuzzleHttp\Psr7;

[2] Fix | Delete

[3] Fix | Delete

use Psr\Http\Message\UriInterface;

[4] Fix | Delete

[5] Fix | Delete

/**

[6] Fix | Delete

* Provides methods to normalize and compare URIs.

[7] Fix | Delete

*

[8] Fix | Delete

* @author Tobias Schultze

[9] Fix | Delete

*

[10] Fix | Delete

* @link https://tools.ietf.org/html/rfc3986#section-6

[11] Fix | Delete

*/

[12] Fix | Delete

final class UriNormalizer

[13] Fix | Delete

{

[14] Fix | Delete

/**

[15] Fix | Delete

* Default normalizations which only include the ones that preserve semantics.

[16] Fix | Delete

*

[17] Fix | Delete

* self::CAPITALIZE_PERCENT_ENCODING | self::DECODE_UNRESERVED_CHARACTERS | self::CONVERT_EMPTY_PATH |

[18] Fix | Delete

* self::REMOVE_DEFAULT_HOST | self::REMOVE_DEFAULT_PORT | self::REMOVE_DOT_SEGMENTS

[19] Fix | Delete

*/

[20] Fix | Delete

const PRESERVING_NORMALIZATIONS = 63;

[21] Fix | Delete

[22] Fix | Delete

/**

[23] Fix | Delete

* All letters within a percent-encoding triplet (e.g., "%3A") are case-insensitive, and should be capitalized.

[24] Fix | Delete

*

[25] Fix | Delete

* Example: http://example.org/a%c2%b1b → http://example.org/a%C2%B1b

[26] Fix | Delete

*/

[27] Fix | Delete

const CAPITALIZE_PERCENT_ENCODING = 1;

[28] Fix | Delete

[29] Fix | Delete

/**

[30] Fix | Delete

* Decodes percent-encoded octets of unreserved characters.

[31] Fix | Delete

*

[32] Fix | Delete

* For consistency, percent-encoded octets in the ranges of ALPHA (%41–%5A and %61–%7A), DIGIT (%30–%39),

[33] Fix | Delete

* hyphen (%2D), period (%2E), underscore (%5F), or tilde (%7E) should not be created by URI producers and,

[34] Fix | Delete

* when found in a URI, should be decoded to their corresponding unreserved characters by URI normalizers.

[35] Fix | Delete

*

[36] Fix | Delete

* Example: http://example.org/%7Eusern%61me/ → http://example.org/~username/

[37] Fix | Delete

*/

[38] Fix | Delete

const DECODE_UNRESERVED_CHARACTERS = 2;

[39] Fix | Delete

[40] Fix | Delete

/**

[41] Fix | Delete

* Converts the empty path to "/" for http and https URIs.

[42] Fix | Delete

*

[43] Fix | Delete

* Example: http://example.org → http://example.org/

[44] Fix | Delete

*/

[45] Fix | Delete

const CONVERT_EMPTY_PATH = 4;

[46] Fix | Delete

[47] Fix | Delete

/**

[48] Fix | Delete

* Removes the default host of the given URI scheme from the URI.

[49] Fix | Delete

*

[50] Fix | Delete

* Only the "file" scheme defines the default host "localhost".

[51] Fix | Delete

* All of `file:/myfile`, `file:///myfile`, and `file://localhost/myfile`

[52] Fix | Delete

* are equivalent according to RFC 3986. The first format is not accepted

[53] Fix | Delete

* by PHPs stream functions and thus already normalized implicitly to the

[54] Fix | Delete

* second format in the Uri class. See `GuzzleHttp\Psr7\Uri::composeComponents`.

[55] Fix | Delete

*

[56] Fix | Delete

* Example: file://localhost/myfile → file:///myfile

[57] Fix | Delete

*/

[58] Fix | Delete

const REMOVE_DEFAULT_HOST = 8;

[59] Fix | Delete

[60] Fix | Delete

/**

[61] Fix | Delete

* Removes the default port of the given URI scheme from the URI.

[62] Fix | Delete

*

[63] Fix | Delete

* Example: http://example.org:80/ → http://example.org/

[64] Fix | Delete

*/

[65] Fix | Delete

const REMOVE_DEFAULT_PORT = 16;

[66] Fix | Delete

[67] Fix | Delete

/**

[68] Fix | Delete

* Removes unnecessary dot-segments.

[69] Fix | Delete

*

[70] Fix | Delete

* Dot-segments in relative-path references are not removed as it would

[71] Fix | Delete

* change the semantics of the URI reference.

[72] Fix | Delete

*

[73] Fix | Delete

* Example: http://example.org/../a/b/../c/./d.html → http://example.org/a/c/d.html

[74] Fix | Delete

*/

[75] Fix | Delete

const REMOVE_DOT_SEGMENTS = 32;

[76] Fix | Delete

[77] Fix | Delete

/**

[78] Fix | Delete

* Paths which include two or more adjacent slashes are converted to one.

[79] Fix | Delete

*

[80] Fix | Delete

* Webservers usually ignore duplicate slashes and treat those URIs equivalent.

[81] Fix | Delete

* But in theory those URIs do not need to be equivalent. So this normalization

[82] Fix | Delete

* may change the semantics. Encoded slashes (%2F) are not removed.

[83] Fix | Delete

*

[84] Fix | Delete

* Example: http://example.org//foo///bar.html → http://example.org/foo/bar.html

[85] Fix | Delete

*/

[86] Fix | Delete

const REMOVE_DUPLICATE_SLASHES = 64;

[87] Fix | Delete

[88] Fix | Delete

/**

[89] Fix | Delete

* Sort query parameters with their values in alphabetical order.

[90] Fix | Delete

*

[91] Fix | Delete

* However, the order of parameters in a URI may be significant (this is not defined by the standard).

[92] Fix | Delete

* So this normalization is not safe and may change the semantics of the URI.

[93] Fix | Delete

*

[94] Fix | Delete

* Example: ?lang=en&article=fred → ?article=fred&lang=en

[95] Fix | Delete

*

[96] Fix | Delete

* Note: The sorting is neither locale nor Unicode aware (the URI query does not get decoded at all) as the

[97] Fix | Delete

* purpose is to be able to compare URIs in a reproducible way, not to have the params sorted perfectly.

[98] Fix | Delete

*/

[99] Fix | Delete

const SORT_QUERY_PARAMETERS = 128;

[100] Fix | Delete

[101] Fix | Delete

/**

[102] Fix | Delete

* Returns a normalized URI.

[103] Fix | Delete

*

[104] Fix | Delete

* The scheme and host component are already normalized to lowercase per PSR-7 UriInterface.

[105] Fix | Delete

* This methods adds additional normalizations that can be configured with the $flags parameter.

[106] Fix | Delete

*

[107] Fix | Delete

* PSR-7 UriInterface cannot distinguish between an empty component and a missing component as

[108] Fix | Delete

* getQuery(), getFragment() etc. always return a string. This means the URIs "/?#" and "/" are

[109] Fix | Delete

* treated equivalent which is not necessarily true according to RFC 3986. But that difference

[110] Fix | Delete

* is highly uncommon in reality. So this potential normalization is implied in PSR-7 as well.

[111] Fix | Delete

*

[112] Fix | Delete

* @param UriInterface $uri The URI to normalize

[113] Fix | Delete

* @param int $flags A bitmask of normalizations to apply, see constants

[114] Fix | Delete

*

[115] Fix | Delete

* @return UriInterface The normalized URI

[116] Fix | Delete

*

[117] Fix | Delete

* @link https://tools.ietf.org/html/rfc3986#section-6.2

[118] Fix | Delete

*/

[119] Fix | Delete

public static function normalize(UriInterface $uri, $flags = self::PRESERVING_NORMALIZATIONS)

[120] Fix | Delete

{

[121] Fix | Delete

if ($flags & self::CAPITALIZE_PERCENT_ENCODING) {

[122] Fix | Delete

$uri = self::capitalizePercentEncoding($uri);

[123] Fix | Delete

}

[124] Fix | Delete

[125] Fix | Delete

if ($flags & self::DECODE_UNRESERVED_CHARACTERS) {

[126] Fix | Delete

$uri = self::decodeUnreservedCharacters($uri);

[127] Fix | Delete

}

[128] Fix | Delete

[129] Fix | Delete

if ($flags & self::CONVERT_EMPTY_PATH && $uri->getPath() === '' &&

[130] Fix | Delete

($uri->getScheme() === 'http' || $uri->getScheme() === 'https')

[131] Fix | Delete

) {

[132] Fix | Delete

$uri = $uri->withPath('/');

[133] Fix | Delete

}

[134] Fix | Delete

[135] Fix | Delete

if ($flags & self::REMOVE_DEFAULT_HOST && $uri->getScheme() === 'file' && $uri->getHost() === 'localhost') {

[136] Fix | Delete

$uri = $uri->withHost('');

[137] Fix | Delete

}

[138] Fix | Delete

[139] Fix | Delete

if ($flags & self::REMOVE_DEFAULT_PORT && $uri->getPort() !== null && Uri::isDefaultPort($uri)) {

[140] Fix | Delete

$uri = $uri->withPort(null);

[141] Fix | Delete

}

[142] Fix | Delete

[143] Fix | Delete

if ($flags & self::REMOVE_DOT_SEGMENTS && !Uri::isRelativePathReference($uri)) {

[144] Fix | Delete

$uri = $uri->withPath(UriResolver::removeDotSegments($uri->getPath()));

[145] Fix | Delete

}

[146] Fix | Delete

[147] Fix | Delete

if ($flags & self::REMOVE_DUPLICATE_SLASHES) {

[148] Fix | Delete

$uri = $uri->withPath(preg_replace('#//++#', '/', $uri->getPath()));

[149] Fix | Delete

}

[150] Fix | Delete

[151] Fix | Delete

if ($flags & self::SORT_QUERY_PARAMETERS && $uri->getQuery() !== '') {

[152] Fix | Delete

$queryKeyValues = explode('&', $uri->getQuery());

[153] Fix | Delete

sort($queryKeyValues);

[154] Fix | Delete

$uri = $uri->withQuery(implode('&', $queryKeyValues));

[155] Fix | Delete

}

[156] Fix | Delete

[157] Fix | Delete

return $uri;

[158] Fix | Delete

}

[159] Fix | Delete

[160] Fix | Delete

/**

[161] Fix | Delete

* Whether two URIs can be considered equivalent.

[162] Fix | Delete

*

[163] Fix | Delete

* Both URIs are normalized automatically before comparison with the given $normalizations bitmask. The method also

[164] Fix | Delete

* accepts relative URI references and returns true when they are equivalent. This of course assumes they will be

[165] Fix | Delete

* resolved against the same base URI. If this is not the case, determination of equivalence or difference of

[166] Fix | Delete

* relative references does not mean anything.

[167] Fix | Delete

*

[168] Fix | Delete

* @param UriInterface $uri1 An URI to compare

[169] Fix | Delete

* @param UriInterface $uri2 An URI to compare

[170] Fix | Delete

* @param int $normalizations A bitmask of normalizations to apply, see constants

[171] Fix | Delete

*

[172] Fix | Delete

* @return bool

[173] Fix | Delete

*

[174] Fix | Delete

* @link https://tools.ietf.org/html/rfc3986#section-6.1

[175] Fix | Delete

*/

[176] Fix | Delete

public static function isEquivalent(UriInterface $uri1, UriInterface $uri2, $normalizations = self::PRESERVING_NORMALIZATIONS)

[177] Fix | Delete

{

[178] Fix | Delete

return (string) self::normalize($uri1, $normalizations) === (string) self::normalize($uri2, $normalizations);

[179] Fix | Delete

}

[180] Fix | Delete

[181] Fix | Delete

private static function capitalizePercentEncoding(UriInterface $uri)

[182] Fix | Delete

{

[183] Fix | Delete

$regex = '/(?:%[A-Fa-f0-9]{2})++/';

[184] Fix | Delete

[185] Fix | Delete

$callback = function (array $match) {

[186] Fix | Delete

return strtoupper($match[0]);

[187] Fix | Delete

};

[188] Fix | Delete

[189] Fix | Delete

return

[190] Fix | Delete

$uri->withPath(

[191] Fix | Delete

preg_replace_callback($regex, $callback, $uri->getPath())

[192] Fix | Delete

)->withQuery(

[193] Fix | Delete

preg_replace_callback($regex, $callback, $uri->getQuery())

[194] Fix | Delete

);

[195] Fix | Delete

}

[196] Fix | Delete

[197] Fix | Delete

private static function decodeUnreservedCharacters(UriInterface $uri)

[198] Fix | Delete

{

[199] Fix | Delete

$regex = '/%(?:2D|2E|5F|7E|3[0-9]|[46][1-9A-F]|[57][0-9A])/i';

[200] Fix | Delete

[201] Fix | Delete

$callback = function (array $match) {

[202] Fix | Delete

return rawurldecode($match[0]);

[203] Fix | Delete

};

[204] Fix | Delete

[205] Fix | Delete

return

[206] Fix | Delete

$uri->withPath(

[207] Fix | Delete

preg_replace_callback($regex, $callback, $uri->getPath())

[208] Fix | Delete

)->withQuery(

[209] Fix | Delete

preg_replace_callback($regex, $callback, $uri->getQuery())

[210] Fix | Delete

);

[211] Fix | Delete

}

[212] Fix | Delete

[213] Fix | Delete

private function __construct()

[214] Fix | Delete

{

[215] Fix | Delete

// cannot be instantiated

[216] Fix | Delete

}

[217] Fix | Delete

}

[218] Fix | Delete

[219] Fix | Delete