From 8b31ff15f51c6936daedf28cc88c753214b569e0 Mon Sep 17 00:00:00 2001 From: Friendika Date: Mon, 1 Aug 2011 16:51:01 -0700 Subject: breaking up boot file (part of zot refactor) --- include/text.php | 933 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 933 insertions(+) create mode 100644 include/text.php (limited to 'include/text.php') diff --git a/include/text.php b/include/text.php new file mode 100644 index 000000000..d65c77872 --- /dev/null +++ b/include/text.php @@ -0,0 +1,933 @@ + replace) +// returns substituted string. +// WARNING: this is pretty basic, and doesn't properly handle search strings that are substrings of each other. +// For instance if 'test' => "foo" and 'testing' => "bar", testing could become either bar or fooing, +// depending on the order in which they were declared in the array. + +require_once("include/template_processor.php"); + +if(! function_exists('replace_macros')) { +function replace_macros($s,$r) { + global $t; + + return $t->replace($s,$r); + +}} + + +// random hash, 64 chars + +if(! function_exists('random_string')) { +function random_string() { + return(hash('sha256',uniqid(rand(),true))); +}} + +/** + * This is our primary input filter. + * + * The high bit hack only involved some old IE browser, forget which (IE5/Mac?) + * that had an XSS attack vector due to stripping the high-bit on an 8-bit character + * after cleansing, and angle chars with the high bit set could get through as markup. + * + * This is now disabled because it was interfering with some legitimate unicode sequences + * and hopefully there aren't a lot of those browsers left. + * + * Use this on any text input where angle chars are not valid or permitted + * They will be replaced with safer brackets. This may be filtered further + * if these are not allowed either. + * + */ + +if(! function_exists('notags')) { +function notags($string) { + + return(str_replace(array("<",">"), array('[',']'), $string)); + +// High-bit filter no longer used +// return(str_replace(array("<",">","\xBA","\xBC","\xBE"), array('[',']','','',''), $string)); +}} + +// use this on "body" or "content" input where angle chars shouldn't be removed, +// and allow them to be safely displayed. + +if(! function_exists('escape_tags')) { +function escape_tags($string) { + + return(htmlspecialchars($string)); +}} + + +// generate a string that's random, but usually pronounceable. +// used to generate initial passwords + +if(! function_exists('autoname')) { +function autoname($len) { + + $vowels = array('a','a','ai','au','e','e','e','ee','ea','i','ie','o','ou','u'); + if(mt_rand(0,5) == 4) + $vowels[] = 'y'; + + $cons = array( + '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 = array('ck','ct','gn','ld','lf','lm','lt','mb','mm', 'mn','mp', + 'nd','ng','nk','nt','rn','rp','rt'); + + $noend = array('bl', 'br', 'cl','cr','dr','fl','fr','gl','gr', + 'kh', 'kl','kr','mn','pl','pr','rh','tr','qu','wh'); + + $start = mt_rand(0,2); + if($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 $noe) { + if((strlen($word) > 2) && (substr($word,-2) == $noe)) { + $word = substr($word,0,-1); + break; + } + } + if(substr($word,-1) == 'q') + $word = substr($word,0,-1); + return $word; +}} + + +// escape text ($str) for XML transport +// returns escaped text. + +if(! function_exists('xmlify')) { +function xmlify($str) { + $buffer = ''; + + for($x = 0; $x < strlen($str); $x ++) { + $char = $str[$x]; + + switch( $char ) { + + case "\r" : + break; + case "&" : + $buffer .= '&'; + break; + case "'" : + $buffer .= '''; + break; + case "\"" : + $buffer .= '"'; + break; + case '<' : + $buffer .= '<'; + break; + case '>' : + $buffer .= '>'; + break; + case "\n" : + $buffer .= "\n"; + break; + default : + $buffer .= $char; + break; + } + } + $buffer = trim($buffer); + return($buffer); +}} + +// undo an xmlify +// pass xml escaped text ($s), returns unescaped text + +if(! function_exists('unxmlify')) { +function unxmlify($s) { + $ret = str_replace('&','&', $s); + $ret = str_replace(array('<','>','"','''),array('<','>','"',"'"),$ret); + return $ret; +}} + +// convenience wrapper, reverse the operation "bin2hex" + +if(! function_exists('hex2bin')) { +function hex2bin($s) { + if(! ctype_xdigit($s)) { + logger('hex2bin: illegal input: ' . print_r(debug_backtrace(), true)); + return($s); + } + + return(pack("H*",$s)); +}} + +// Automatic pagination. +// To use, get the count of total items. +// Then call $a->set_pager_total($number_items); +// Optionally call $a->set_pager_itemspage($n) to the number of items to display on each page +// Then call paginate($a) after the end of the display loop to insert the pager block on the page +// (assuming there are enough items to paginate). +// When using with SQL, the setting LIMIT %d, %d => $a->pager['start'],$a->pager['itemspage'] +// will limit the results to the correct items for the current page. +// The actual page handling is then accomplished at the application layer. + +if(! function_exists('paginate')) { +function paginate(&$a) { + $o = ''; + $stripped = preg_replace('/(&page=[0-9]*)/','',$a->query_string); + $stripped = str_replace('q=','',$stripped); + $stripped = trim($stripped,'/'); + $pagenum = $a->pager['page']; + $url = $a->get_baseurl() . '/' . $stripped; + + + if($a->pager['total'] > $a->pager['itemspage']) { + $o .= '
'; + if($a->pager['page'] != 1) + $o .= ''."pager['page'] - 1).'">' . t('prev') . ' '; + + $o .= "" . t('first') . " "; + + $numpages = $a->pager['total'] / $a->pager['itemspage']; + + $numstart = 1; + $numstop = $numpages; + + if($numpages > 14) { + $numstart = (($pagenum > 7) ? ($pagenum - 7) : 1); + $numstop = (($pagenum > ($numpages - 7)) ? $numpages : ($numstart + 14)); + } + + for($i = $numstart; $i <= $numstop; $i++){ + if($i == $a->pager['page']) + $o .= ''.(($i < 10) ? ' '.$i : $i); + else + $o .= "".(($i < 10) ? ' '.$i : $i).""; + $o .= ' '; + } + + if(($a->pager['total'] % $a->pager['itemspage']) != 0) { + if($i == $a->pager['page']) + $o .= ''.(($i < 10) ? ' '.$i : $i); + else + $o .= "".(($i < 10) ? ' '.$i : $i).""; + $o .= ' '; + } + + $lastpage = (($numpages > intval($numpages)) ? intval($numpages)+1 : $numpages); + $o .= "" . t('last') . " "; + + if(($a->pager['total'] - ($a->pager['itemspage'] * $a->pager['page'])) > 0) + $o .= ''."pager['page'] + 1).'">' . t('next') . ''; + $o .= '
'."\r\n"; + } + return $o; +}} + +// Turn user/group ACLs stored as angle bracketed text into arrays + +if(! function_exists('expand_acl')) { +function expand_acl($s) { + // turn string array of angle-bracketed elements into numeric array + // e.g. "<1><2><3>" => array(1,2,3); + $ret = array(); + + if(strlen($s)) { + $t = str_replace('<','',$s); + $a = explode('>',$t); + foreach($a as $aa) { + if(intval($aa)) + $ret[] = intval($aa); + } + } + return $ret; +}} + +// Used to wrap ACL elements in angle brackets for storage + +if(! function_exists('sanitise_acl')) { +function sanitise_acl(&$item) { + if(intval($item)) + $item = '<' . intval(notags(trim($item))) . '>'; + else + unset($item); +}} + + +// Convert an ACL array to a storable string + +if(! function_exists('perms2str')) { +function perms2str($p) { + $ret = ''; + $tmp = $p; + if(is_array($tmp)) { + array_walk($tmp,'sanitise_acl'); + $ret = implode('',$tmp); + } + return $ret; +}} + +// generate a guaranteed unique (for this domain) item ID for ATOM +// safe from birthday paradox + +if(! function_exists('item_new_uri')) { +function item_new_uri($hostname,$uid) { + + do { + $dups = false; + $hash = random_string(); + + $uri = "urn:X-dfrn:" . $hostname . ':' . $uid . ':' . $hash; + + $r = q("SELECT `id` FROM `item` WHERE `uri` = '%s' LIMIT 1", + dbesc($uri)); + if(count($r)) + $dups = true; + } while($dups == true); + return $uri; +}} + +// Generate a guaranteed unique photo ID. +// safe from birthday paradox + +if(! function_exists('photo_new_resource')) { +function photo_new_resource() { + + do { + $found = false; + $resource = hash('md5',uniqid(mt_rand(),true)); + $r = q("SELECT `id` FROM `photo` WHERE `resource-id` = '%s' LIMIT 1", + dbesc($resource) + ); + if(count($r)) + $found = true; + } while($found == true); + return $resource; +}} + + +// wrapper to load a view template, checking for alternate +// languages before falling back to the default + +// obsolete, deprecated. + +if(! function_exists('load_view_file')) { +function load_view_file($s) { + global $lang, $a; + if(! isset($lang)) + $lang = 'en'; + $b = basename($s); + $d = dirname($s); + if(file_exists("$d/$lang/$b")) + return file_get_contents("$d/$lang/$b"); + + $theme = current_theme(); + + if(file_exists("$d/theme/$theme/$b")) + return file_get_contents("$d/theme/$theme/$b"); + + return file_get_contents($s); +}} + +if(! function_exists('get_intltext_template')) { +function get_intltext_template($s) { + global $lang; + + if(! isset($lang)) + $lang = 'en'; + + if(file_exists("view/$lang/$s")) + return file_get_contents("view/$lang/$s"); + elseif(file_exists("view/en/$s")) + return file_get_contents("view/en/$s"); + else + return file_get_contents("view/$s"); +}} + +if(! function_exists('get_markup_template')) { +function get_markup_template($s) { + + $theme = current_theme(); + + if(file_exists("view/theme/$theme/$s")) + return file_get_contents("view/theme/$theme/$s"); + else + return file_get_contents("view/$s"); + +}} + + + + + +// for html,xml parsing - let's say you've got +// an attribute foobar="class1 class2 class3" +// and you want to find out if it contains 'class3'. +// you can't use a normal sub string search because you +// might match 'notclass3' and a regex to do the job is +// possible but a bit complicated. +// pass the attribute string as $attr and the attribute you +// are looking for as $s - returns true if found, otherwise false + +if(! function_exists('attribute_contains')) { +function attribute_contains($attr,$s) { + $a = explode(' ', $attr); + if(count($a) && in_array($s,$a)) + return true; + return false; +}} + +if(! function_exists('logger')) { +function logger($msg,$level = 0) { + $debugging = get_config('system','debugging'); + $loglevel = intval(get_config('system','loglevel')); + $logfile = get_config('system','logfile'); + + if((! $debugging) || (! $logfile) || ($level > $loglevel)) + return; + + @file_put_contents($logfile, datetime_convert() . ':' . session_id() . ' ' . $msg . "\n", FILE_APPEND); + return; +}} + + +if(! function_exists('activity_match')) { +function activity_match($haystack,$needle) { + if(($haystack === $needle) || ((basename($needle) === $haystack) && strstr($needle,NAMESPACE_ACTIVITY_SCHEMA))) + return true; + return false; +}} + + +// Pull out all #hashtags and @person tags from $s; +// We also get @person@domain.com - which would make +// the regex quite complicated as tags can also +// end a sentence. So we'll run through our results +// and strip the period from any tags which end with one. +// Returns array of tags found, or empty array. + + +if(! function_exists('get_tags')) { +function get_tags($s) { + $ret = array(); + + // ignore anything in a code block + + $s = preg_replace('/\[code\](.*?)\[\/code\]/sm','',$s); + + // Match full names against @tags including the space between first and last + // We will look these up afterward to see if they are full names or not recognisable. + + if(preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A,:?]+)([ \x0D\x0A,:?]|$)/',$s,$match)) { + foreach($match[1] as $mtch) { + if(strstr($mtch,"]")) { + // we might be inside a bbcode color tag - leave it alone + continue; + } + if(substr($mtch,-1,1) === '.') + $ret[] = substr($mtch,0,-1); + else + $ret[] = $mtch; + } + } + + // Otherwise pull out single word tags. These can be @nickname, @first_last + // and #hash tags. + + if(preg_match_all('/([@#][^ \x0D\x0A,:?]+)([ \x0D\x0A,:?]|$)/',$s,$match)) { + foreach($match[1] as $mtch) { + if(strstr($mtch,"]")) { + // we might be inside a bbcode color tag - leave it alone + continue; + } + // ignore strictly numeric tags like #1 + if((strpos($mtch,'#') === 0) && ctype_digit(substr($mtch,1))) + continue; + if(substr($mtch,-1,1) === '.') + $ret[] = substr($mtch,0,-1); + else + $ret[] = $mtch; + } + } + return $ret; +}} + + +// quick and dirty quoted_printable encoding + +if(! function_exists('qp')) { +function qp($s) { +return str_replace ("%","=",rawurlencode($s)); +}} + + + +if(! function_exists('get_mentions')) { +function get_mentions($item) { + $o = ''; + if(! strlen($item['tag'])) + return $o; + + $arr = explode(',',$item['tag']); + foreach($arr as $x) { + $matches = null; + if(preg_match('/@\[url=([^\]]*)\]/',$x,$matches)) { + $o .= "\t\t" . '' . "\r\n"; + $o .= "\t\t" . '' . "\r\n"; + } + } + return $o; +}} + +if(! function_exists('contact_block')) { +function contact_block() { + $o = ''; + $a = get_app(); + + $shown = get_pconfig($a->profile['uid'],'system','display_friend_count'); + if(! $shown) + $shown = 24; + + if((! is_array($a->profile)) || ($a->profile['hide-friends'])) + return $o; + $r = q("SELECT COUNT(*) AS `total` FROM `contact` WHERE `uid` = %d AND `self` = 0 AND `blocked` = 0 and `pending` = 0", + intval($a->profile['uid']) + ); + if(count($r)) { + $total = intval($r[0]['total']); + } + if(! $total) { + $o .= '

' . t('No contacts') . '

'; + return $o; + } + $r = q("SELECT * FROM `contact` WHERE `uid` = %d AND `self` = 0 AND `blocked` = 0 and `pending` = 0 ORDER BY RAND() LIMIT %d", + intval($a->profile['uid']), + intval($shown) + ); + if(count($r)) { + $o .= '

' . sprintf( tt('%d Contact','%d Contacts', $total),$total) . '

'; + foreach($r as $rr) { + $o .= micropro($rr,true,'mpfriend'); + } + $o .= '
'; + $o .= '
' . t('View Contacts') . '
'; + + } + + $arr = array('contacts' => $r, 'output' => $o); + + call_hooks('contact_block_end', $arr); + return $o; + +}} + +if(! function_exists('micropro')) { +function micropro($contact, $redirect = false, $class = '', $textmode = false) { + + if($class) + $class = ' ' . $class; + + $url = $contact['url']; + $sparkle = ''; + + if($redirect) { + $a = get_app(); + $redirect_url = $a->get_baseurl() . '/redir/' . $contact['id']; + if(local_user() && ($contact['uid'] == local_user()) && ($contact['network'] === 'dfrn')) { + $url = $redirect_url; + $sparkle = ' sparkle'; + } + } + $click = ((x($contact,'click')) ? ' onclick="' . $contact['click'] . '" ' : ''); + if($click) + $url = ''; + if($textmode) { + return '
'. $contact['name'] . '
' . "\r\n"; + } + else { + return '
' . $contact['name'] + . '
' . "\r\n"; + } +}} + + + +if(! function_exists('search')) { +function search($s,$id='search-box',$url='/search') { + $a = get_app(); + $o = '
'; + $o .= '
'; + $o .= ''; + $o .= ''; + $o .= '
'; + return $o; +}} + +if(! function_exists('valid_email')) { +function valid_email($x){ + if(preg_match('/^[_a-zA-Z0-9-]+(\.[_a-zA-Z0-9-]+)*@[a-zA-Z0-9-]+(\.[a-zA-Z0-9-]+)+$/',$x)) + return true; + return false; +}} + + +if(! function_exists('aes_decrypt')) { +function aes_decrypt($val,$ky) +{ + $key="\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"; + for($a=0;$a=0 and ord(substr($dec, strlen($dec)-1,1))<=16)? chr(ord( substr($dec,strlen($dec)-1,1))):null)); +}} + + +if(! function_exists('aes_encrypt')) { +function aes_encrypt($val,$ky) +{ + $key="\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"; + for($a=0;$a$1', $s); + return($s); +}} + + +/** + * + * Function: smilies + * + * Description: + * Replaces text emoticons with graphical images + * + * @Parameter: string $s + * + * Returns string + */ + +if(! function_exists('smilies')) { +function smilies($s) { + $a = get_app(); + + return str_replace( + array( '<3', '</3', '<\\3', ':-)', ':)', ';-)', ':-(', ':(', ':-P', ':P', ':-"', ':-x', ':-X', ':-D', '8-|', '8-O'), + array( + '<3', + '</3', + '<\\3', + ':-)', + ':)', + ';-)', + ':-(', + ':(', + ':-P', + ':P', + ':-\', + ':-x', + ':-X', + ':-D', + '8-|', + '8-O' + ), $s); +}} + + + +if(! function_exists('day_translate')) { +function day_translate($s) { + $ret = str_replace(array('Monday','Tuesday','Wednesday','Thursday','Friday','Saturday','Sunday'), + array( t('Monday'), t('Tuesday'), t('Wednesday'), t('Thursday'), t('Friday'), t('Saturday'), t('Sunday')), + $s); + + $ret = str_replace(array('January','February','March','April','May','June','July','August','September','October','November','December'), + array( t('January'), t('February'), t('March'), t('April'), t('May'), t('June'), t('July'), t('August'), t('September'), t('October'), t('November'), t('December')), + $ret); + + return $ret; +}} + + +if(! function_exists('normalise_link')) { +function normalise_link($url) { + $ret = str_replace(array('https:','//www.'), array('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. + * + * Return true if the URLs match, otherwise false. + * + */ + +if(! function_exists('link_compare')) { +function link_compare($a,$b) { + if(strcasecmp(normalise_link($a),normalise_link($b)) === 0) + return true; + return false; +}} + + +if(! function_exists('prepare_body')) { +function prepare_body($item,$attach = false) { + + $s = prepare_text($item['body']); + if(! $attach) + return $s; + + $arr = explode(',',$item['attach']); + if(count($arr)) { + $s .= '
'; + foreach($arr as $r) { + $matches = false; + $icon = ''; + $cnt = preg_match('|\[attach\]href=\"(.*?)\" size=\"(.*?)\" type=\"(.*?)\" title=\"(.*?)\"\[\/attach\]|',$r,$matches); + if($cnt) { + $icontype = strtolower(substr($matches[3],0,strpos($matches[3],'/'))); + switch($icontype) { + case 'video': + case 'audio': + case 'image': + case 'text': + $icon = '
'; + break; + default: + $icon = '
'; + break; + } + $title = ((strlen(trim($matches[4]))) ? escape_tags(trim($matches[4])) : escape_tags($matches[1])); + $title .= ' ' . $matches[2] . ' ' . t('bytes'); + + $s .= '' . $icon . ''; + } + } + $s .= '
'; + } + return $s; +}} + +if(! function_exists('prepare_text')) { +function prepare_text($text) { + + require_once('include/bbcode.php'); + + $s = smilies(bbcode($text)); + + return $s; +}} + + +/** + * return atom link elements for all of our hubs + */ + +if(! function_exists('feed_hublinks')) { +function feed_hublinks() { + + $hub = get_config('system','huburl'); + + $hubxml = ''; + if(strlen($hub)) { + $hubs = explode(',', $hub); + if(count($hubs)) { + foreach($hubs as $h) { + $h = trim($h); + if(! strlen($h)) + continue; + $hubxml .= '' . "\n" ; + } + } + } + return $hubxml; +}} + +/* return atom link elements for salmon endpoints */ + +if(! function_exists('feed_salmonlinks')) { +function feed_salmonlinks($nick) { + + $a = get_app(); + + $salmon = '' . "\n" ; + + // old style links that status.net still needed as of 12/2010 + + $salmon .= ' ' . "\n" ; + $salmon .= ' ' . "\n" ; + return $salmon; +}} + +if(! function_exists('get_plink')) { +function get_plink($item) { + $a = get_app(); + $plink = (((x($item,'plink')) && (! $item['private'])) ? '
' : ''); + return $plink; +}} + +if(! function_exists('unamp')) { +function unamp($s) { + return str_replace('&', '&', $s); +}} + + + + +if(! function_exists('lang_selector')) { +function lang_selector() { + global $lang; + $o = '
'; + $o .= ''; + return $o; +}} + + +if(! function_exists('return_bytes')) { +function return_bytes ($size_str) { + switch (substr ($size_str, -1)) + { + case 'M': case 'm': return (int)$size_str * 1048576; + case 'K': case 'k': return (int)$size_str * 1024; + case 'G': case 'g': return (int)$size_str * 1073741824; + default: return $size_str; + } +}} + +function generate_guid() { + $found = true; + do { + $guid = substr(random_string(),0,16); + $x = q("SELECT `uid` FROM `user` WHERE `guid` = '%s' LIMIT 1", + dbesc($guid) + ); + if(! count($x)) + $found = false; + } while ($found == true ); + return $guid; +} + + +function pkcs5_pad ($text, $blocksize) +{ + $pad = $blocksize - (strlen($text) % $blocksize); + return $text . str_repeat(chr($pad), $pad); +} + +function pkcs5_unpad($text) +{ + $pad = ord($text{strlen($text)-1}); + if ($pad > strlen($text)) return false; + if (strspn($text, chr($pad), strlen($text) - $pad) != $pad) return false; + return substr($text, 0, -1 * $pad); +} + + +function base64url_encode($s, $strip_padding = false) { + + $s = strtr(base64_encode($s),'+/','-_'); + + if($strip_padding) + $s = str_replace('=','',$s); + + return $s; +} + +function base64url_decode($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,'-_','+/')); +} -- cgit v1.2.3