<?php
+
/**
* @file src/Util/Strings.php
*/
namespace Friendica\Util;
+use Friendica\Content\ContactSelector;
+use Friendica\Core\Logger;
+
/**
* @brief This class handles string functions
*/
*
* @param int $size
* @return string
+ * @throws \Exception
*/
public static function getRandomHex($size = 64)
{
}
/**
- * This is our primary input filter.
+ * Checks, if the given string is a valid hexadecimal code
*
- * The high bit hack only involved some old IE browser, forget which (IE5/Mac?)
- * that had an XSS attack vector due to stripping the high-bit on an 8-bit character
- * after cleansing, and angle chars with the high bit set could get through as markup.
+ * @param string $hexCode
*
- * This is now disabled because it was interfering with some legitimate unicode sequences
- * and hopefully there aren't a lot of those browsers left.
+ * @return bool
+ */
+ public static function isHex($hexCode)
+ {
+ return !empty($hexCode) ? @preg_match("/^[a-f0-9]{2,}$/i", $hexCode) && !(strlen($hexCode) & 1) : false;
+ }
+
+ /**
+ * @brief This is our primary input filter.
*
* Use this on any text input where angle chars are not valid or permitted
* They will be replaced with safer brackets. This may be filtered further
* @param string $string Input string
* @return string Filtered string
*/
- public static function removeTags($string)
+ public static function escapeTags($string)
{
return str_replace(["<", ">"], ['[', ']'], $string);
}
* @brief Use this on "body" or "content" input where angle chars shouldn't be removed,
* and allow them to be safely displayed.
* @param string $string
- *
+ *
* @return string
*/
- public static function escapeTags($string)
+ public static function escapeHtml($string)
{
- return htmlspecialchars($string, ENT_COMPAT, 'UTF-8', false);
+ return htmlentities($string, ENT_QUOTES | ENT_HTML5, "UTF-8", false);
}
/**
* @brief Generate a string that's random, but usually pronounceable. Used to generate initial passwords
- *
+ *
* @param int $len length
- *
+ *
* @return string
*/
public static function getRandomName($len)
$vowels = ['a', 'a', 'ai', 'au', 'e', 'e', 'e', 'ee', 'ea', 'i', 'ie', 'o', 'ou', 'u'];
- if (mt_rand(0, 5) == 4) {
+ if (random_int(0, 5) == 4) {
$vowels[] = 'y';
}
$cons = [
- 'b', 'bl', 'br',
- 'c', 'ch', 'cl', 'cr',
- 'd', 'dr',
- 'f', 'fl', 'fr',
- 'g', 'gh', 'gl', 'gr',
- 'h',
- 'j',
- 'k', 'kh', 'kl', 'kr',
- 'l',
- 'm',
- 'n',
- 'p', 'ph', 'pl', 'pr',
- 'qu',
- 'r', 'rh',
- 's' ,'sc', 'sh', 'sm', 'sp', 'st',
- 't', 'th', 'tr',
- 'v',
- 'w', 'wh',
- 'x',
- 'z', 'zh'
- ];
-
- $midcons = ['ck', 'ct', 'gn', 'ld', 'lf', 'lm', 'lt', 'mb', 'mm', 'mn', 'mp',
- 'nd', 'ng', 'nk', 'nt', 'rn', 'rp', 'rt'];
-
- $noend = ['bl', 'br', 'cl', 'cr', 'dr', 'fl', 'fr', 'gl', 'gr',
- 'kh', 'kl', 'kr', 'mn', 'pl', 'pr', 'rh', 'tr', 'qu', 'wh', 'q'];
-
- $start = mt_rand(0, 2);
+ 'b', 'bl', 'br',
+ 'c', 'ch', 'cl', 'cr',
+ 'd', 'dr',
+ 'f', 'fl', 'fr',
+ 'g', 'gh', 'gl', 'gr',
+ 'h',
+ 'j',
+ 'k', 'kh', 'kl', 'kr',
+ 'l',
+ 'm',
+ 'n',
+ 'p', 'ph', 'pl', 'pr',
+ 'qu',
+ 'r', 'rh',
+ 's', 'sc', 'sh', 'sm', 'sp', 'st',
+ 't', 'th', 'tr',
+ 'v',
+ 'w', 'wh',
+ 'x',
+ 'z', 'zh'
+ ];
+
+ $midcons = [
+ 'ck', 'ct', 'gn', 'ld', 'lf', 'lm', 'lt', 'mb', 'mm', 'mn', 'mp',
+ 'nd', 'ng', 'nk', 'nt', 'rn', 'rp', 'rt'
+ ];
+
+ $noend = [
+ 'bl', 'br', 'cl', 'cr', 'dr', 'fl', 'fr', 'gl', 'gr',
+ 'kh', 'kl', 'kr', 'mn', 'pl', 'pr', 'rh', 'tr', 'qu', 'wh', 'q'
+ ];
+
+ $start = random_int(0, 2);
if ($start == 0) {
$table = $vowels;
} else {
$word = '';
- for ($x = 0; $x < $len; $x ++) {
- $r = mt_rand(0, count($table) - 1);
+ for ($x = 0; $x < $len; $x++) {
+ $r = random_int(0, count($table) - 1);
$word .= $table[$r];
if ($table == $vowels) {
} else {
$table = $vowels;
}
-
}
$word = substr($word, 0, $len);
}
/**
- * @brief translate and format the networkname of a contact
+ * Translate and format the network name of a contact
+ *
+ * @param string $network Network name of the contact (e.g. dfrn, rss and so on)
+ * @param string $url The contact url
*
- * @param string $network Networkname of the contact (e.g. dfrn, rss and so on)
- * @param string $url The contact url
- *
- * @return string Formatted network name
+ * @return string Formatted network name
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function formatNetworkName($network, $url = 0)
+ public static function formatNetworkName($network, $url = '')
{
- if ($network != "") {
- if ($url != "") {
- $network_name = '<a href="'.$url.'">'.ContactSelector::networkToName($network, $url)."</a>";
+ if ($network != '') {
+ if ($url != '') {
+ $network_name = '<a href="' . $url . '">' . ContactSelector::networkToName($network, $url) . '</a>';
} else {
$network_name = ContactSelector::networkToName($network);
}
}
/**
- * @brief Remove intentation from a text
- *
+ * @brief Remove indentation from a text
+ *
* @param string $text String to be transformed.
* @param string $chr Optional. Indentation tag. Default tab (\t).
* @param int $count Optional. Default null.
- *
+ *
* @return string Transformed string.
*/
public static function deindent($text, $chr = "[\t ]", $count = NULL)
/**
* @brief Get byte size returned in a Data Measurement (KB, MB, GB)
- *
+ *
* @param int $bytes The number of bytes to be measured
* @param int $precision Optional. Default 2.
- *
+ *
* @return string Size with measured units.
*/
public static function formatBytes($bytes, $precision = 2)
/**
* @brief Protect percent characters in sprintf calls
- *
+ *
* @param string $s String to transform.
- *
+ *
* @return string Transformed string.
*/
public static function protectSprintf($s)
/**
* @brief Base64 Encode URL and translate +/ to -_ Optionally strip padding.
- *
+ *
* @param string $s URL to encode
* @param boolean $strip_padding Optional. Default false
- *
+ *
* @return string Encoded URL
*/
public static function base64UrlEncode($s, $strip_padding = false)
/**
* @brief Decode Base64 Encoded URL and translate -_ to +/
* @param string $s URL to decode
- *
+ *
* @return string Decoded URL
+ * @throws \Exception
*/
public static function base64UrlDecode($s)
{
return base64_decode(strtr($s, '-_', '+/'));
}
- /**
- * @brief Pull out all #hashtags and @person tags from $string.
- *
- * We also get @person@domain.com - which would make
- * the regex quite complicated as tags can also
- * end a sentence. So we'll run through our results
- * and strip the period from any tags which end with one.
- * Returns array of tags found, or empty array.
- *
- * @param string $string Post content
- *
- * @return array List of tag and person names
- */
- public static function getTags($string)
- {
- $ret = [];
-
- // Convert hashtag links to hashtags
- $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2', $string);
-
- // ignore anything in a code block
- $string = preg_replace('/\[code\](.*?)\[\/code\]/sm', '', $string);
-
- // Force line feeds at bbtags
- $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
-
- // ignore anything in a bbtag
- $string = preg_replace('/\[(.*?)\]/sm', '', $string);
-
- // Match full names against @tags including the space between first and last
- // We will look these up afterward to see if they are full names or not recognisable.
-
- if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
- foreach ($matches[1] as $match) {
- if (strstr($match, ']')) {
- // we might be inside a bbcode color tag - leave it alone
- continue;
- }
-
- if (substr($match, -1, 1) === '.') {
- $ret[] = substr($match, 0, -1);
- } else {
- $ret[] = $match;
- }
- }
- }
-
- // Otherwise pull out single word tags. These can be @nickname, @first_last
- // and #hash tags.
-
- if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) {
- foreach ($matches[1] as $match) {
- if (strstr($match, ']')) {
- // we might be inside a bbcode color tag - leave it alone
- continue;
- }
- if (substr($match, -1, 1) === '.') {
- $match = substr($match,0,-1);
- }
- // ignore strictly numeric tags like #1
- if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
- continue;
- }
- // try not to catch url fragments
- if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
- continue;
- }
- $ret[] = $match;
- }
- }
-
- return $ret;
- }
-
- /**
- * @brief Check for a valid email string
- *
- * @param string $email_address Email address to be evaluated.
- *
- * @return boolean Value indicating whether or not the string is a valid email address.
- */
- public static function isValidEmail($email_address)
- {
- return preg_match('/^[_a-zA-Z0-9\-\+]+(\.[_a-zA-Z0-9\-\+]+)*@[a-zA-Z0-9-]+(\.[a-zA-Z0-9-]+)+$/', $email_address);
- }
-
/**
* @brief Normalize url
*
* @param string $url URL to be normalized.
- *
+ *
* @return string Normalized URL.
*/
- public static function normaliseLink($url) // normalise_link()
+ public static function normaliseLink($url)
{
$ret = str_replace(['https:', '//www.'], ['http:', '//'], $url);
return rtrim($ret, '/');
/**
* @brief Normalize OpenID identity
- *
+ *
* @param string $s OpenID Identity
- *
+ *
* @return string normalized OpenId Identity
*/
- function normaliseOpenID($s)
+ public static function normaliseOpenID($s)
{
return trim(str_replace(['http://', 'https://'], ['', ''], $s), '/');
}
{
return (strcasecmp(self::normaliseLink($a), self::normaliseLink($b)) === 0);
}
+
+
+ /**
+ * Ensures the provided URI has its query string punctuation in order.
+ *
+ * @param string $uri
+ * @return string
+ */
+ public static function ensureQueryParameter($uri)
+ {
+ if (strpos($uri, '?') === false && ($pos = strpos($uri, '&')) !== false) {
+ $uri = substr($uri, 0, $pos) . '?' . substr($uri, $pos + 1);
+ }
+
+ return $uri;
+ }
+
+
+ /**
+ * Check if the trimmed provided string is starting with one of the provided characters
+ *
+ * @param string $string
+ * @param array $chars
+ * @return bool
+ */
+ public static function startsWith($string, array $chars)
+ {
+ $return = in_array(substr(trim($string), 0, 1), $chars);
+
+ return $return;
+ }
+
+ /**
+ * Returns the regular expression string to match URLs in a given text
+ *
+ * @return string
+ * @see https://daringfireball.net/2010/07/improved_regex_for_matching_urls
+ */
+ public static function autoLinkRegEx()
+ {
+ return '@
+(?<![=\'\]"/]) # Not preceded by [, =, \', ], ", /
+\b
+( # Capture 1: entire matched URL
+ https?:// # http or https protocol
+ (?:
+ [^/\s\xA0`!()\[\]{};:\'",<>?«»“”‘’.] # Domain can\'t start with a .
+ [^/\s\xA0`!()\[\]{};:\'",<>?«»“”‘’]+ # Domain can\'t end with a .
+ \.
+ [^/\s\xA0`!()\[\]{};:\'".,<>?«»“”‘’]+/? # Followed by a slash
+ )
+ (?: # One or more:
+ [^\s\xA0()<>]+ # Run of non-space, non-()<>
+ | # or
+ \(([^\s\xA0()<>]+|(\([^\s()<>]+\)))*\) # balanced parens, up to 2 levels
+ | # or
+ [^\s\xA0`!()\[\]{};:\'".,<>?«»“”‘’] # not a space or one of these punct chars
+ )*
+)@xiu';
+ }
+
+ /**
+ * Ensures a single path item doesn't contain any path-traversing characters
+ *
+ * @see https://stackoverflow.com/a/46097713
+ * @param string $pathItem
+ * @return string
+ */
+ public static function sanitizeFilePathItem($pathItem)
+ {
+ $pathItem = str_replace('/', '_', $pathItem);
+ $pathItem = str_replace('\\', '_', $pathItem);
+ $pathItem = str_replace(DIRECTORY_SEPARATOR, '_', $pathItem); // In case it does not equal the standard values
+
+ return $pathItem;
+ }
}