X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=include%2Ftext.php;h=ee83345c09954bc3cb351441ac6a3ff03e561149;hb=9dfc970bd63086c85569f144d1d2a507d51935de;hp=db8ebcc4eb17adac7acfe218dc1fc7e071c84355;hpb=d50a1edef468b9e3de456cb7c4570da5272b0cd8;p=friendica.git

diff --git a/include/text.php b/include/text.php
index db8ebcc4eb..ee83345c09 100644
--- a/include/text.php
+++ b/include/text.php
@@ -26,144 +26,12 @@ use Friendica\Util\Proxy as ProxyUtils;
 use Friendica\Core\Logger;
 use Friendica\Core\Renderer;
 use Friendica\Model\FileTag;
+use Friendica\Util\Strings;
 use Friendica\Util\XML;
 use Friendica\Content\Text\HTML;
 
 require_once "include/conversation.php";
 
-/**
- * @brief Generates a pseudo-random string of hexadecimal characters
- *
- * @param int $size
- * @return string
- */
-function random_string($size = 64)
-{
-	$byte_size = ceil($size / 2);
-
-	$bytes = random_bytes($byte_size);
-
-	$return = substr(bin2hex($bytes), 0, $size);
-
-	return $return;
-}
-
-/**
- * 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.
- *
- * 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
- * if these are not allowed either.
- *
- * @param string $string Input string
- * @return string Filtered string
- */
-function notags($string) {
-	return str_replace(["<", ">"], ['[', ']'], $string);
-
-//  High-bit filter no longer used
-//	return str_replace(array("<",">","\xBA","\xBC","\xBE"), array('[',']','','',''), $string);
-}
-
-
-/**
- * use this on "body" or "content" input where angle chars shouldn't be removed,
- * and allow them to be safely displayed.
- * @param string $string
- * @return string
- */
-function escape_tags($string) {
-	return htmlspecialchars($string, ENT_COMPAT, 'UTF-8', false);
-}
-
-
-/**
- * generate a string that's random, but usually pronounceable.
- * used to generate initial passwords
- * @param int $len
- * @return string
- */
-function autoname($len) {
-
-	if ($len <= 0) {
-		return '';
-	}
-
-	$vowels = ['a','a','ai','au','e','e','e','ee','ea','i','ie','o','ou','u'];
-	if (mt_rand(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'];
-
-	$start = mt_rand(0,2);
-	if ($start == 0) {
-		$table = $vowels;
-	} else {
-		$table = $cons;
-	}
-
-	$word = '';
-
-	for ($x = 0; $x < $len; $x ++) {
-		$r = mt_rand(0,count($table) - 1);
-		$word .= $table[$r];
-
-		if ($table == $vowels) {
-			$table = array_merge($cons,$midcons);
-		} else {
-			$table = $vowels;
-		}
-
-	}
-
-	$word = substr($word,0,$len);
-
-	foreach ($noend as $noe) {
-		$noelen = strlen($noe);
-		if ((strlen($word) > $noelen) && (substr($word, -$noelen) == $noe)) {
-			$word = autoname($len);
-			break;
-		}
-	}
-
-	return $word;
-}
-
 /**
  * Turn user/group ACLs stored as angle bracketed text into arrays
  *
@@ -194,7 +62,7 @@ function expand_acl($s) {
  */
 function sanitise_acl(&$item) {
 	if (intval($item)) {
-		$item = '<' . intval(notags(trim($item))) . '>';
+		$item = '<' . intval(Strings::escapeTags(trim($item))) . '>';
 	} else {
 		unset($item);
 	}
@@ -255,78 +123,6 @@ function activity_match($haystack,$needle) {
 	return (($haystack === $needle) || ((basename($needle) === $haystack) && strstr($needle, NAMESPACE_ACTIVITY_SCHEMA)));
 }
 
-
-/**
- * @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 $string Post content
- * @return array List of tag and person names
- */
-function get_tags($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 $ret;
-}
-
-
 /**
  * quick and dirty quoted_printable encoding
  *
@@ -337,45 +133,6 @@ function qp($s) {
 	return str_replace("%", "=", rawurlencode($s));
 }
 
-/**
- * @brief Check for a valid email string
- *
- * @param string $email_address
- * @return boolean
- */
-function valid_email($email_address)
-{
-	return preg_match('/^[_a-zA-Z0-9\-\+]+(\.[_a-zA-Z0-9\-\+]+)*@[a-zA-Z0-9-]+(\.[a-zA-Z0-9-]+)+$/', $email_address);
-}
-
-/**
- * Normalize url
- *
- * @param string $url
- * @return string
- */
-function normalise_link($url) {
-	$ret = str_replace(['https:', '//www.'], ['http:', '//'], $url);
-	return rtrim($ret,'/');
-}
-
-
-/**
- * Compare two URLs to see if they are the same, but ignore
- * slight but hopefully insignificant differences such as if one
- * is https and the other isn't, or if one is www.something and
- * the other isn't - and also ignore case differences.
- *
- * @param string $a first url
- * @param string $b second url
- * @return boolean True if the URLs match, otherwise False
- *
- */
-function link_compare($a, $b) {
-	return (strcasecmp(normalise_link($a), normalise_link($b)) === 0);
-}
-
-
 /**
  * @brief Find any non-embedded images in private items and add redir links to them
  *
@@ -493,42 +250,6 @@ function get_cats_and_terms($item)
 	return [$categories, $folders];
 }
 
-
-/**
- * get private link for item
- * @param array $item
- * @return boolean|array False if item has not plink, otherwise array('href'=>plink url, 'title'=>translated title)
- */
-function get_plink($item) {
-	$a = get_app();
-
-	if ($a->user['nickname'] != "") {
-		$ret = [
-				//'href' => "display/" . $a->user['nickname'] . "/" . $item['id'],
-				'href' => "display/" . $item['guid'],
-				'orig' => "display/" . $item['guid'],
-				'title' => L10n::t('View on separate page'),
-				'orig_title' => L10n::t('view on separate page'),
-			];
-
-		if (x($item, 'plink')) {
-			$ret["href"] = $a->removeBaseURL($item['plink']);
-			$ret["title"] = L10n::t('link to source');
-		}
-
-	} elseif (x($item, 'plink') && ($item['private'] != 1)) {
-		$ret = [
-				'href' => $item['plink'],
-				'orig' => $item['plink'],
-				'title' => L10n::t('link to source'),
-			];
-	} else {
-		$ret = [];
-	}
-
-	return $ret;
-}
-
 /**
  * return number of bytes in size (K, M, G)
  * @param string $size_str
@@ -543,53 +264,6 @@ function return_bytes($size_str) {
 	}
 }
 
-/**
- * @param string $s
- * @param boolean $strip_padding
- * @return string
- */
-function base64url_encode($s, $strip_padding = false) {
-
-	$s = strtr(base64_encode($s), '+/', '-_');
-
-	if ($strip_padding) {
-		$s = str_replace('=','',$s);
-	}
-
-	return $s;
-}
-
-/**
- * @param string $s
- * @return string
- */
-function base64url_decode($s) {
-
-	if (is_array($s)) {
-		Logger::log('base64url_decode: illegal input: ' . print_r(debug_backtrace(), true));
-		return $s;
-	}
-
-/*
- *  // Placeholder for new rev of salmon which strips base64 padding.
- *  // PHP base64_decode handles the un-padded input without requiring this step
- *  // Uncomment if you find you need it.
- *
- *	$l = strlen($s);
- *	if (!strpos($s,'=')) {
- *		$m = $l % 4;
- *		if ($m == 2)
- *			$s .= '==';
- *		if ($m == 3)
- *			$s .= '=';
- *	}
- *
- */
-
-	return base64_decode(strtr($s,'-_','+/'));
-}
-
-
 function bb_translate_video($s) {
 
 	$matches = null;
@@ -606,11 +280,6 @@ function bb_translate_video($s) {
 	return $s;
 }
 
-function normalise_openid($s) {
-	return trim(str_replace(['http://', 'https://'], ['', ''], $s), '/');
-}
-
-
 function undo_post_tagging($s) {
 	$matches = null;
 	$cnt = preg_match_all('/([!#@])\[url=(.*?)\](.*?)\[\/url\]/ism', $s, $matches, PREG_SET_ORDER);
@@ -626,10 +295,6 @@ function undo_post_tagging($s) {
 	return $s;
 }
 
-function protect_sprintf($s) {
-	return str_replace('%', '%%', $s);
-}
-
 /// @TODO Rewrite this
 function is_a_date_arg($s) {
 	$i = intval($s);
@@ -648,59 +313,3 @@ function is_a_date_arg($s) {
 
 	return false;
 }
-
-/**
- * remove intentation from a text
- */
-function deindent($text, $chr = "[\t ]", $count = NULL) {
-	$lines = explode("\n", $text);
-
-	if (is_null($count)) {
-		$m = [];
-		$k = 0;
-		while ($k < count($lines) && strlen($lines[$k]) == 0) {
-			$k++;
-		}
-		preg_match("|^" . $chr . "*|", $lines[$k], $m);
-		$count = strlen($m[0]);
-	}
-
-	for ($k = 0; $k < count($lines); $k++) {
-		$lines[$k] = preg_replace("|^" . $chr . "{" . $count . "}|", "", $lines[$k]);
-	}
-
-	return implode("\n", $lines);
-}
-
-function formatBytes($bytes, $precision = 2) {
-	$units = ['B', 'KB', 'MB', 'GB', 'TB'];
-
-	$bytes = max($bytes, 0);
-	$pow = floor(($bytes ? log($bytes) : 0) / log(1024));
-	$pow = min($pow, count($units) - 1);
-
-	$bytes /= pow(1024, $pow);
-
-	return round($bytes, $precision) . ' ' . $units[$pow];
-}
-
-/**
- * @brief translate and format the networkname of a contact
- *
- * @param string $network
- *	Networkname of the contact (e.g. dfrn, rss and so on)
- * @param sting $url
- *	The contact url
- * @return string
- */
-function format_network_name($network, $url = 0) {
-	if ($network != "") {
-		if ($url != "") {
-			$network_name = '<a href="'.$url.'">'.ContactSelector::networkToName($network, $url)."</a>";
-		} else {
-			$network_name = ContactSelector::networkToName($network);
-		}
-
-		return $network_name;
-	}
-}