annotate core/lib/Drupal/Component/Utility/UserAgent.php @ 19:fa3358dc1485 tip

Add ndrum files
author Chris Cannam
date Wed, 28 Aug 2019 13:14:47 +0100
parents 4c8ae668cc8c
children
rev   line source
Chris@0 1 <?php
Chris@0 2
Chris@0 3 namespace Drupal\Component\Utility;
Chris@0 4
Chris@0 5 /**
Chris@0 6 * Provides user agent related utility functions.
Chris@0 7 *
Chris@0 8 * @ingroup utility
Chris@0 9 */
Chris@0 10 class UserAgent {
Chris@0 11
Chris@0 12 /**
Chris@0 13 * Identifies user agent language from the Accept-language HTTP header.
Chris@0 14 *
Chris@0 15 * The algorithm works as follows:
Chris@0 16 * - map user agent language codes to available language codes.
Chris@0 17 * - order all user agent language codes by qvalue from high to low.
Chris@0 18 * - add generic user agent language codes if they aren't already specified
Chris@0 19 * but with a slightly lower qvalue.
Chris@0 20 * - find the most specific available language code with the highest qvalue.
Chris@0 21 * - if 2 or more languages are having the same qvalue, respect the order of
Chris@0 22 * them inside the $languages array.
Chris@0 23 *
Chris@0 24 * We perform user agent accept-language parsing only if page cache is
Chris@0 25 * disabled, otherwise we would cache a user-specific preference.
Chris@0 26 *
Chris@0 27 * @param string $http_accept_language
Chris@0 28 * The value of the "Accept-Language" HTTP header.
Chris@0 29 * @param array $langcodes
Chris@0 30 * An array of available language codes to pick from.
Chris@0 31 * @param array $mappings
Chris@0 32 * (optional) Custom mappings to support user agents that are sending non
Chris@0 33 * standard language codes. No mapping is assumed by default.
Chris@0 34 *
Chris@0 35 * @return string
Chris@0 36 * The selected language code or FALSE if no valid language can be
Chris@0 37 * identified.
Chris@0 38 */
Chris@0 39 public static function getBestMatchingLangcode($http_accept_language, $langcodes, $mappings = []) {
Chris@0 40 // The Accept-Language header contains information about the language
Chris@0 41 // preferences configured in the user's user agent / operating system.
Chris@0 42 // RFC 2616 (section 14.4) defines the Accept-Language header as follows:
Chris@0 43 // Accept-Language = "Accept-Language" ":"
Chris@0 44 // 1#( language-range [ ";" "q" "=" qvalue ] )
Chris@0 45 // language-range = ( ( 1*8ALPHA *( "-" 1*8ALPHA ) ) | "*" )
Chris@0 46 // Samples: "hu, en-us;q=0.66, en;q=0.33", "hu,en-us;q=0.5"
Chris@0 47 $ua_langcodes = [];
Chris@0 48 if (preg_match_all('@(?<=[, ]|^)([a-zA-Z-]+|\*)(?:;q=([0-9.]+))?(?:$|\s*,\s*)@', trim($http_accept_language), $matches, PREG_SET_ORDER)) {
Chris@0 49 foreach ($matches as $match) {
Chris@0 50 if ($mappings) {
Chris@0 51 $langcode = strtolower($match[1]);
Chris@0 52 foreach ($mappings as $ua_langcode => $standard_langcode) {
Chris@0 53 if ($langcode == $ua_langcode) {
Chris@0 54 $match[1] = $standard_langcode;
Chris@0 55 }
Chris@0 56 }
Chris@0 57 }
Chris@0 58 // We can safely use strtolower() here, tags are ASCII.
Chris@0 59 // RFC2616 mandates that the decimal part is no more than three digits,
Chris@0 60 // so we multiply the qvalue by 1000 to avoid floating point
Chris@0 61 // comparisons.
Chris@0 62 $langcode = strtolower($match[1]);
Chris@0 63 $qvalue = isset($match[2]) ? (float) $match[2] : 1;
Chris@0 64 // Take the highest qvalue for this langcode. Although the request
Chris@0 65 // supposedly contains unique langcodes, our mapping possibly resolves
Chris@0 66 // to the same langcode for different qvalues. Keep the highest.
Chris@0 67 $ua_langcodes[$langcode] = max(
Chris@0 68 (int) ($qvalue * 1000),
Chris@0 69 (isset($ua_langcodes[$langcode]) ? $ua_langcodes[$langcode] : 0)
Chris@0 70 );
Chris@0 71 }
Chris@0 72 }
Chris@0 73
Chris@0 74 // We should take pristine values from the HTTP headers, but Internet
Chris@0 75 // Explorer from version 7 sends only specific language tags (eg. fr-CA)
Chris@0 76 // without the corresponding generic tag (fr) unless explicitly configured.
Chris@0 77 // In that case, we assume that the lowest value of the specific tags is the
Chris@0 78 // value of the generic language to be as close to the HTTP 1.1 spec as
Chris@0 79 // possible.
Chris@0 80 // See http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.4 and
Chris@0 81 // http://blogs.msdn.com/b/ie/archive/2006/10/17/accept-language-header-for-internet-explorer-7.aspx
Chris@0 82 asort($ua_langcodes);
Chris@0 83 foreach ($ua_langcodes as $langcode => $qvalue) {
Chris@0 84 // For Chinese languages the generic tag is either zh-hans or zh-hant, so
Chris@0 85 // we need to handle this separately, we can not split $langcode on the
Chris@0 86 // first occurrence of '-' otherwise we get a non-existing language zh.
Chris@0 87 // All other languages use a langcode without a '-', so we can safely
Chris@0 88 // split on the first occurrence of it.
Chris@0 89 if (strlen($langcode) > 7 && (substr($langcode, 0, 7) == 'zh-hant' || substr($langcode, 0, 7) == 'zh-hans')) {
Chris@0 90 $generic_tag = substr($langcode, 0, 7);
Chris@0 91 }
Chris@0 92 else {
Chris@0 93 $generic_tag = strtok($langcode, '-');
Chris@0 94 }
Chris@0 95 if (!empty($generic_tag) && !isset($ua_langcodes[$generic_tag])) {
Chris@0 96 // Add the generic langcode, but make sure it has a lower qvalue as the
Chris@0 97 // more specific one, so the more specific one gets selected if it's
Chris@0 98 // defined by both the user agent and us.
Chris@0 99 $ua_langcodes[$generic_tag] = $qvalue - 0.1;
Chris@0 100 }
Chris@0 101 }
Chris@0 102
Chris@0 103 // Find the added language with the greatest qvalue, following the rules
Chris@0 104 // of RFC 2616 (section 14.4). If several languages have the same qvalue,
Chris@0 105 // prefer the one with the greatest weight.
Chris@0 106 $best_match_langcode = FALSE;
Chris@0 107 $max_qvalue = 0;
Chris@0 108 foreach ($langcodes as $langcode_case_sensitive) {
Chris@0 109 // Language tags are case insensitive (RFC2616, sec 3.10).
Chris@0 110 $langcode = strtolower($langcode_case_sensitive);
Chris@0 111
Chris@0 112 // If nothing matches below, the default qvalue is the one of the wildcard
Chris@0 113 // language, if set, or is 0 (which will never match).
Chris@0 114 $qvalue = isset($ua_langcodes['*']) ? $ua_langcodes['*'] : 0;
Chris@0 115
Chris@0 116 // Find the longest possible prefix of the user agent supplied language
Chris@0 117 // ('the language-range') that matches this site language ('the language
Chris@0 118 // tag').
Chris@0 119 $prefix = $langcode;
Chris@0 120 do {
Chris@0 121 if (isset($ua_langcodes[$prefix])) {
Chris@0 122 $qvalue = $ua_langcodes[$prefix];
Chris@0 123 break;
Chris@0 124 }
Chris@0 125 } while ($prefix = substr($prefix, 0, strrpos($prefix, '-')));
Chris@0 126
Chris@0 127 // Find the best match.
Chris@0 128 if ($qvalue > $max_qvalue) {
Chris@0 129 $best_match_langcode = $langcode_case_sensitive;
Chris@0 130 $max_qvalue = $qvalue;
Chris@0 131 }
Chris@0 132 }
Chris@0 133
Chris@0 134 return $best_match_langcode;
Chris@0 135 }
Chris@0 136
Chris@0 137 }