X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FUtil%2FStrings.php;h=d6583b9c61c0eb29993ee350d2eb2683e3592c3a;hb=0c738c4f8bd820519a4b46b27f419dbc921721f5;hp=3febe60bd8b4139a29927a5564a49ca2b4b53df2;hpb=5276c28a78e188eb4ec8dcbf87a4dd1ac193c8d7;p=friendica.git
diff --git a/src/Util/Strings.php b/src/Util/Strings.php
index 3febe60bd8..d6583b9c61 100644
--- a/src/Util/Strings.php
+++ b/src/Util/Strings.php
@@ -5,17 +5,21 @@
namespace Friendica\Util;
+use Friendica\Content\ContactSelector;
+use Friendica\Core\Logger;
+
/**
* @brief This class handles string functions
*/
class Strings
{
- /**
- * @brief Generates a pseudo-random string of hexadecimal characters
- *
- * @param int $size
- * @return string
- */
+ /**
+ * @brief Generates a pseudo-random string of hexadecimal characters
+ *
+ * @param int $size
+ * @return string
+ * @throws \Exception
+ */
public static function getRandomHex($size = 64)
{
$byte_size = ceil($size / 2);
@@ -28,14 +32,7 @@ class Strings
}
/**
- * This is our primary input filter.
- *
- * 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.
- *
- * This is now disabled because it was interfering with some legitimate unicode sequences
- * and hopefully there aren't a lot of those browsers left.
+ * @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
@@ -44,7 +41,7 @@ class Strings
* @param string $string Input string
* @return string Filtered string
*/
- public static function removeTags($string)
+ public static function escapeTags($string)
{
return str_replace(["<", ">"], ['[', ']'], $string);
}
@@ -56,7 +53,7 @@ class Strings
*
* @return string
*/
- public static function escapeTags($string)
+ public static function escapeHtml($string)
{
return htmlspecialchars($string, ENT_COMPAT, 'UTF-8', false);
}
@@ -143,19 +140,20 @@ class Strings
return $word;
}
- /**
- * @brief translate and format the networkname of a contact
- *
- * @param string $network Networkname of the contact (e.g. dfrn, rss and so on)
- * @param string $url The contact url
- *
- * @return string Formatted network name
- */
- public static function formatNetworkName($network, $url = 0)
+ /**
+ * 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 $url The contact url
+ *
+ * @return string Formatted network name
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
+ */
+ public static function formatNetworkName($network, $url = '')
{
- if ($network != "") {
- if ($url != "") {
- $network_name = ''.ContactSelector::networkToName($network, $url)."";
+ if ($network != '') {
+ if ($url != '') {
+ $network_name = '' . ContactSelector::networkToName($network, $url) . '';
} else {
$network_name = ContactSelector::networkToName($network);
}
@@ -165,7 +163,7 @@ class Strings
}
/**
- * @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).
@@ -244,12 +242,13 @@ class Strings
return $s;
}
- /**
- * @brief Decode Base64 Encoded URL and translate -_ to +/
- * @param string $s URL to decode
- *
- * @return string Decoded URL
- */
+ /**
+ * @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)
{
if (is_array($s)) {
@@ -276,92 +275,6 @@ class Strings
return base64_decode(strtr($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-]+)+$/', $email_address);
- }
-
/**
* @brief Normalize url
*
@@ -369,7 +282,7 @@ class Strings
*
* @return string Normalized URL.
*/
- public static function normaliseLink($url) // normalise_link()
+ public static function normaliseLink($url)
{
$ret = str_replace(['https:', '//www.'], ['http:', '//'], $url);
return rtrim($ret, '/');
@@ -382,7 +295,7 @@ class Strings
*
* @return string normalized OpenId Identity
*/
- function normaliseOpenID($s)
+ public static function normaliseOpenID($s)
{
return trim(str_replace(['http://', 'https://'], ['', ''], $s), '/');
}
@@ -402,4 +315,64 @@ class Strings
{
return (strcasecmp(self::normaliseLink($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 '@(?xi)
+(??«»ââââ.] # Domain can\'t start with a .
+ [^/\s`!()\[\]{};:\'",<>?«»ââââ]+ # Domain can\'t end with a .
+ \.
+ [^/\s`!()\[\]{};:\'".,<>?«»ââââ]+/? # Followed by a slash
+ )
+ (?: # One or more:
+ [^\s()<>]+ # Run of non-space, non-()<>
+ | # or
+ \(([^\s()<>]+|(\([^\s()<>]+\)))*\) # balanced parens, up to 2 levels
+ | # or
+ [^\s`!()\[\]{};:\'".,<>?«»ââââ] # not a space or one of these punct chars
+ )*
+)@';
+ }
}