friendica_2019-12_sharedHos.../mod/parse_url.php

<?php
/** 
 * @file mod/parse_url.php
 * 
 * @todo https://developers.google.com/+/plugins/snippet/
 * 
 * @verbatim
 * <meta itemprop="name" content="Toller Titel">
 * <meta itemprop="description" content="Eine tolle Beschreibung">
 * <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
 * 
 * <body itemscope itemtype="http://schema.org/Product">
 *   <h1 itemprop="name">Shiny Trinket</h1>
 *   <img itemprop="image" src="{image-url}" />
 *   <p itemprop="description">Shiny trinkets are shiny.</p>
 * </body>
 * @endverbatim
*/

if(!function_exists('deletenode')) {
	function deletenode(&$doc, $node)
	{
		$xpath = new DomXPath($doc);
		$list = $xpath->query("//".$node);
		foreach ($list as $child)
			$child->parentNode->removeChild($child);
	}
}

function completeurl($url, $scheme) {
	$urlarr = parse_url($url);

	if (isset($urlarr["scheme"]))
		return($url);

	$schemearr = parse_url($scheme);

	$complete = $schemearr["scheme"]."://".$schemearr["host"];

	if (@$schemearr["port"] != "")
		$complete .= ":".$schemearr["port"];

		if(strpos($urlarr['path'],'/') !== 0)
			$complete .= '/';

	$complete .= $urlarr["path"];

	if (@$urlarr["query"] != "")
		$complete .= "?".$urlarr["query"];

	if (@$urlarr["fragment"] != "")
		$complete .= "#".$urlarr["fragment"];

	return($complete);
}

function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) {

	if ($url == "")
		return false;

	$r = q("SELECT * FROM `parsed_url` WHERE `url` = '%s' AND `guessing` = %d AND `oembed` = %d",
		dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed));

	if ($r)
		$data = $r[0]["content"];

	if (!is_null($data)) {
		$data = unserialize($data);
		return $data;
	}

	$data = parseurl_getsiteinfo($url, $no_guessing, $do_oembed);

	q("INSERT INTO `parsed_url` (`url`, `guessing`, `oembed`, `content`, `created`) VALUES ('%s', %d, %d, '%s', '%s')
		 ON DUPLICATE KEY UPDATE `content` = '%s', `created` = '%s'",
		dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed),
		dbesc(serialize($data)), dbesc(datetime_convert()),
		dbesc(serialize($data)), dbesc(datetime_convert()));

	return $data;
}

function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1) {
	require_once("include/network.php");
	require_once("include/Photo.php");

	$a = get_app();

	$siteinfo = array();

	// Check if the URL does contain a scheme
	$scheme = parse_url($url, PHP_URL_SCHEME);

	if ($scheme == "") {
		$url = "http://".trim($url, "/");
	}

	if ($count > 10) {
		logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
		return($siteinfo);
	}

	$url = trim($url, "'");
	$url = trim($url, '"');

	$url = original_url($url);

	$siteinfo["url"] = $url;
	$siteinfo["type"] = "link";

	$check_cert = get_config('system','verifyssl');

	$stamp1 = microtime(true);

	$ch = curl_init();
	curl_setopt($ch, CURLOPT_URL, $url);
	curl_setopt($ch, CURLOPT_HEADER, 1);
	curl_setopt($ch, CURLOPT_NOBODY, 1);
	curl_setopt($ch, CURLOPT_TIMEOUT, 3);
	curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
	curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
	curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
	curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, (($check_cert) ? 2 : false));

	$header = curl_exec($ch);
	$curl_info = @curl_getinfo($ch);
	$http_code = $curl_info['http_code'];
	curl_close($ch);

	$a->save_timestamp($stamp1, "network");

	if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
		AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
		if ($curl_info['redirect_url'] != "")
			$siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
		else
			$siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
		return($siteinfo);
	}

	// if the file is too large then exit
	if ($curl_info["download_content_length"] > 1000000)
		return($siteinfo);

	// if it isn't a HTML file then exit
	if (($curl_info["content_type"] != "") AND !strstr(strtolower($curl_info["content_type"]),"html"))
		return($siteinfo);

	if ($do_oembed) {
		require_once("include/oembed.php");

		$oembed_data = oembed_fetch_url($url);

		if (!in_array($oembed_data->type, array("error", "rich"))) {
			$siteinfo["type"] = $oembed_data->type;
		}

		if (($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
			if (isset($oembed_data->title))
				$siteinfo["title"] = $oembed_data->title;
			if (isset($oembed_data->description))
				$siteinfo["text"] = trim($oembed_data->description);
			if (isset($oembed_data->thumbnail_url))
				$siteinfo["image"] = $oembed_data->thumbnail_url;
		}
	}

	$stamp1 = microtime(true);

	// Now fetch the body as well
	$ch = curl_init();
	curl_setopt($ch, CURLOPT_URL, $url);
	curl_setopt($ch, CURLOPT_HEADER, 1);
	curl_setopt($ch, CURLOPT_NOBODY, 0);
	curl_setopt($ch, CURLOPT_TIMEOUT, 10);
	curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
	curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
	curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
	curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, (($check_cert) ? 2 : false));

	$header = curl_exec($ch);
	$curl_info = @curl_getinfo($ch);
	$http_code = $curl_info['http_code'];
	curl_close($ch);

	$a->save_timestamp($stamp1, "network");

	// Fetch the first mentioned charset. Can be in body or header
	$charset = "";
	if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
		$charset = trim(trim(trim(array_pop($matches)), ';,'));

	if ($charset == "")
		$charset = "utf-8";

	$pos = strpos($header, "\r\n\r\n");

	if ($pos)
		$body = trim(substr($header, $pos));
	else
		$body = $header;

	if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
		logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
		//$body = mb_convert_encoding($body, "UTF-8", $charset);
		$body = iconv($charset, "UTF-8//TRANSLIT", $body);
	}

	$body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");

	$doc = new DOMDocument();
	@$doc->loadHTML($body);

	deletenode($doc, 'style');
	deletenode($doc, 'script');
	deletenode($doc, 'option');
	deletenode($doc, 'h1');
	deletenode($doc, 'h2');
	deletenode($doc, 'h3');
	deletenode($doc, 'h4');
	deletenode($doc, 'h5');
	deletenode($doc, 'h6');
	deletenode($doc, 'ol');
	deletenode($doc, 'ul');

	$xpath = new DomXPath($doc);

	$list = $xpath->query("//meta[@content]");
	foreach ($list as $node) {
		$attr = array();
		if ($node->attributes->length)
			foreach ($node->attributes as $attribute)
				$attr[$attribute->name] = $attribute->value;

		if (@$attr["http-equiv"] == 'refresh') {
			$path = $attr["content"];
			$pathinfo = explode(";", $path);
			$content = "";
			foreach ($pathinfo AS $value) {
				if (substr(strtolower($value), 0, 4) == "url=")
					$content = substr($value, 4);
			}
			if ($content != "") {
				$siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
				return($siteinfo);
			}
		}
	}

	$list = $xpath->query("//title");
	if ($list->length > 0)
		$siteinfo["title"] = $list->item(0)->nodeValue;

	//$list = $xpath->query("head/meta[@name]");
	$list = $xpath->query("//meta[@name]");
	foreach ($list as $node) {
		$attr = array();
		if ($node->attributes->length)
			foreach ($node->attributes as $attribute)
				$attr[$attribute->name] = $attribute->value;

		$attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));

		if ($attr["content"] != "")
			switch (strtolower($attr["name"])) {
				case "fulltitle":
					$siteinfo["title"] = $attr["content"];
					break;
				case "description":
					$siteinfo["text"] = $attr["content"];
					break;
				case "thumbnail":
					$siteinfo["image"] = $attr["content"];
					break;
				case "twitter:image":
					$siteinfo["image"] = $attr["content"];
					break;
				case "twitter:image:src":
					$siteinfo["image"] = $attr["content"];
					break;
				case "twitter:card":
					if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
						$siteinfo["type"] = $attr["content"];
					break;
				case "twitter:description":
					$siteinfo["text"] = $attr["content"];
					break;
				case "twitter:title":
					$siteinfo["title"] = $attr["content"];
					break;
				case "dc.title":
					$siteinfo["title"] = $attr["content"];
					break;
				case "dc.description":
					$siteinfo["text"] = $attr["content"];
					break;
				case "keywords":
					$keywords = explode(",", $attr["content"]);
					break;
				case "news_keywords":
					$keywords = explode(",", $attr["content"]);
					break;
			}
		if ($siteinfo["type"] == "summary")
			$siteinfo["type"] = "link";
	}

	if (isset($keywords)) {
		$siteinfo["keywords"] = array();
		foreach ($keywords as $keyword)
			if (!in_array(trim($keyword), $siteinfo["keywords"]))
				$siteinfo["keywords"][] = trim($keyword);
	}

	//$list = $xpath->query("head/meta[@property]");
	$list = $xpath->query("//meta[@property]");
	foreach ($list as $node) {
		$attr = array();
		if ($node->attributes->length)
			foreach ($node->attributes as $attribute)
				$attr[$attribute->name] = $attribute->value;

		$attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));

		if ($attr["content"] != "")
			switch (strtolower($attr["property"])) {
				case "og:image":
					$siteinfo["image"] = $attr["content"];
					break;
				case "og:title":
					$siteinfo["title"] = $attr["content"];
					break;
				case "og:description":
					$siteinfo["text"] = $attr["content"];
					break;
			}
	}

	if ((@$siteinfo["image"] == "") AND !$no_guessing) {
	    $list = $xpath->query("//img[@src]");
	    foreach ($list as $node) {
		$attr = array();
		if ($node->attributes->length)
		    foreach ($node->attributes as $attribute)
			$attr[$attribute->name] = $attribute->value;

			$src = completeurl($attr["src"], $url);
			$photodata = get_photo_info($src);

			if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
				if ($photodata[0] > 300) {
					$photodata[1] = round($photodata[1] * (300 / $photodata[0]));
					$photodata[0] = 300;
				}
				if ($photodata[1] > 300) {
					$photodata[0] = round($photodata[0] * (300 / $photodata[1]));
					$photodata[1] = 300;
				}
				$siteinfo["images"][] = array("src"=>$src,
								"width"=>$photodata[0],
								"height"=>$photodata[1]);
			}

		}
    } elseif ($siteinfo["image"] != "") {
		$src = completeurl($siteinfo["image"], $url);

		unset($siteinfo["image"]);

		$photodata = get_photo_info($src);

		if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
			$siteinfo["images"][] = array("src"=>$src,
							"width"=>$photodata[0],
							"height"=>$photodata[1]);
	}

	if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
		$text = "";

		$list = $xpath->query("//div[@class='article']");
		foreach ($list as $node)
			if (strlen($node->nodeValue) > 40)
				$text .= " ".trim($node->nodeValue);

		if ($text == "") {
			$list = $xpath->query("//div[@class='content']");
			foreach ($list as $node)
				if (strlen($node->nodeValue) > 40)
					$text .= " ".trim($node->nodeValue);
		}

		// If none text was found then take the paragraph content
		if ($text == "") {
			$list = $xpath->query("//p");
			foreach ($list as $node)
				if (strlen($node->nodeValue) > 40)
					$text .= " ".trim($node->nodeValue);
		}

		if ($text != "") {
			$text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));

			while (strpos($text, "  "))
				$text = trim(str_replace("  ", " ", $text));

			$siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
		}
	}

	logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);

	call_hooks('getsiteinfo', $siteinfo);

	return($siteinfo);
}

function arr_add_hashes(&$item,$k) {
	$item = '#' . $item;
}

function parse_url_content(&$a) {

	require_once("include/items.php");

	$text = null;
	$str_tags = '';

	$textmode = false;

	if(local_user() && (! feature_enabled(local_user(),'richtext')))
		$textmode = true;

	//if($textmode)
	$br = (($textmode) ? "\n" : '<br />');

	if(x($_GET,'binurl'))
		$url = trim(hex2bin($_GET['binurl']));
	else
		$url = trim($_GET['url']);

	if($_GET['title'])
		$title = strip_tags(trim($_GET['title']));

	if($_GET['description'])
		$text = strip_tags(trim($_GET['description']));

	if($_GET['tags']) {
		$arr_tags = str_getcsv($_GET['tags']);
		if(count($arr_tags)) {
			array_walk($arr_tags,'arr_add_hashes');
			$str_tags = $br . implode(' ',$arr_tags) . $br;
		}
	}

	// add url scheme if missing
	$arrurl = parse_url($url);
	if (!x($arrurl, 'scheme')) {
		if (x($arrurl, 'host'))
			$url = "http:".$url;
		else
			$url = "http://".$url;
	}

	logger('parse_url: ' . $url);

	if($textmode)
		$template = '[bookmark=%s]%s[/bookmark]%s';
	else
		$template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";

	$arr = array('url' => $url, 'text' => '');

	call_hooks('parse_link', $arr);

	if(strlen($arr['text'])) {
		echo $arr['text'];
		killme();
	}


	if($url && $title && $text) {

		$title = str_replace(array("\r","\n"),array('',''),$title);

		if($textmode)
			$text = '[quote]' . trim($text) . '[/quote]' . $br;
		else {
			$text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
			$title = htmlspecialchars($title);
		}

		$result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;

		logger('parse_url (unparsed): returns: ' . $result);

		echo $result;
		killme();
	}

	$siteinfo = parseurl_getsiteinfo($url);

	unset($siteinfo["keywords"]);

	$info = add_page_info_data($siteinfo);

	if (!$textmode)
		// Replace ' with ’ - not perfect - but the richtext editor has problems otherwise
		$info = str_replace(array("&#039;"), array("&#8217;"), $info);

	echo $info;

	killme();
}
?>
-												ajaxify the wall poster

											
										
										
											2010-07-23 07:41:45 +02:00
+								<?php
-												Revert "Updated modules to allow for partial overrides without errors"

This reverts commit db949bb802448184bfe5164d8d3dd86ddf51b187.

											
										
										
											2016-02-07 15:11:34 +01:00
+								/**
-												reformat todo according to doxygen style
											
										
										
											2015-12-25 23:17:34 +01:00
+								 * @file mod/parse_url.php
-												Revert "Updated modules to allow for partial overrides without errors"

This reverts commit db949bb802448184bfe5164d8d3dd86ddf51b187.

											
										
										
											2016-02-07 15:11:34 +01:00
+								 *
-												reformat todo according to doxygen style
											
										
										
											2015-12-25 23:17:34 +01:00
+								 * @todo https://developers.google.com/+/plugins/snippet/
-												Revert "Updated modules to allow for partial overrides without errors"

This reverts commit db949bb802448184bfe5164d8d3dd86ddf51b187.

											
										
										
											2016-02-07 15:11:34 +01:00
+								 *
-												reformat todo according to doxygen style
											
										
										
											2015-12-25 23:17:34 +01:00
+								 * @verbatim
 								 * <meta itemprop="name" content="Toller Titel">
 								 * <meta itemprop="description" content="Eine tolle Beschreibung">
 								 * <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
-												Revert "Updated modules to allow for partial overrides without errors"

This reverts commit db949bb802448184bfe5164d8d3dd86ddf51b187.

											
										
										
											2016-02-07 15:11:34 +01:00
+								 *
-												reformat todo according to doxygen style
											
										
										
											2015-12-25 23:17:34 +01:00
+								 * <body itemscope itemtype="http://schema.org/Product">
 								 *   <h1 itemprop="name">Shiny Trinket</h1>
 								 *   <img itemprop="image" src="{image-url}" />
 								 *   <p itemprop="description">Shiny trinkets are shiny.</p>
 								 * </body>
 								 * @endverbatim
-												parse_url: Just added a comment for future elements that should be added as well.

											
										
										
											2012-07-18 21:06:38 +02:00
+								*/
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+								if(!function_exists('deletenode')) {
 									function deletenode(&$doc, $node)
 									{
 										$xpath = new DomXPath($doc);
 										$list = $xpath->query("//".$node);
 										foreach ($list as $child)
 											$child->parentNode->removeChild($child);
 									}
 								}
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+								function completeurl($url, $scheme) {
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									$urlarr = parse_url($url);
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									if (isset($urlarr["scheme"]))
 										return($url);
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									$schemearr = parse_url($scheme);
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									$complete = $schemearr["scheme"]."://".$schemearr["host"];
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									if (@$schemearr["port"] != "")
 										$complete .= ":".$schemearr["port"];
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
-												update writable for dfrn contacts when using local delivery

											
										
										
											2012-09-06 01:26:11 +02:00
+										if(strpos($urlarr['path'],'/') !== 0)
 											$complete .= '/';
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									$complete .= $urlarr["path"];
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									if (@$urlarr["query"] != "")
 										$complete .= "?".$urlarr["query"];
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									if (@$urlarr["fragment"] != "")
 										$complete .= "#".$urlarr["fragment"];
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									return($complete);
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+								}
-												Poco: New fields for last contact and last failure of a contact.

											
										
										
											2015-07-17 01:08:28 +02:00
+								function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) {
-												OEmbed and parse_url are now cached in dedicated tables

											
										
										
											2016-01-14 23:59:51 +01:00
+									if ($url == "")
 										return false;
 									$r = q("SELECT * FROM `parsed_url` WHERE `url` = '%s' AND `guessing` = %d AND `oembed` = %d",
 										dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed));
 									if ($r)
 										$data = $r[0]["content"];
-												Poco: New fields for last contact and last failure of a contact.

											
										
										
											2015-07-17 01:08:28 +02:00
+									if (!is_null($data)) {
 										$data = unserialize($data);
 										return $data;
 									}
 									$data = parseurl_getsiteinfo($url, $no_guessing, $do_oembed);
-												Avoid SQL errors because of existing entries

											
										
										
											2016-05-03 19:20:58 +02:00
+									q("INSERT INTO `parsed_url` (`url`, `guessing`, `oembed`, `content`, `created`) VALUES ('%s', %d, %d, '%s', '%s')
 										 ON DUPLICATE KEY UPDATE `content` = '%s', `created` = '%s'",
 										dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed),
 										dbesc(serialize($data)), dbesc(datetime_convert()),
 										dbesc(serialize($data)), dbesc(datetime_convert()));
-												Poco: New fields for last contact and last failure of a contact.

											
										
										
											2015-07-17 01:08:28 +02:00
 									return $data;
 								}
-												Preventing an endless loop while fetching site infos

											
										
										
											2014-07-14 08:37:40 +02:00
+								function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1) {
-												Created hashtags from keywords now are links. URL are cleaned from tracking data.

											
										
										
											2014-10-03 13:02:15 +02:00
+									require_once("include/network.php");
-												When a server isn't reachable there were timeout problems with fetching image data

											
										
										
											2016-01-18 15:38:38 +01:00
+									require_once("include/Photo.php");
-												Building a central functionality to export postings to all networks that only support plaintext postings. In near future this will replace the similar code in the several addons.

											
										
										
											2014-05-03 12:04:54 +02:00
-												The useragent string doesn't contains of the words "Mozilla" anymore - but the name "Friendica" and the base url

											
										
										
											2014-09-07 10:47:37 +02:00
+									$a = get_app();
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+									$siteinfo = array();
-												Support for twitter cards (in both directions)

											
										
										
											2014-02-22 15:46:19 +01:00
-												Improved handling with wrong formatted urls and fetchinfo

											
										
										
											2016-11-12 21:23:00 +01:00
+									// Check if the URL does contain a scheme
 									$scheme = parse_url($url, PHP_URL_SCHEME);
 									if ($scheme == "") {
 										$url = "http://".trim($url, "/");
 									}
-												Preventing an endless loop while fetching site infos

											
										
										
											2014-07-14 08:37:40 +02:00
+									if ($count > 10) {
 										logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
 										return($siteinfo);
 									}
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+									$url = trim($url, "'");
 									$url = trim($url, '"');
-												Created hashtags from keywords now are links. URL are cleaned from tracking data.

											
										
										
											2014-10-03 13:02:15 +02:00
 									$url = original_url($url);
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+									$siteinfo["url"] = $url;
-												Support for twitter cards (in both directions)

											
										
										
											2014-02-22 15:46:19 +01:00
+									$siteinfo["type"] = "link";
-												Improved handling with wrong formatted urls and fetchinfo

											
										
										
											2016-11-12 21:23:00 +01:00
+									$check_cert = get_config('system','verifyssl');
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									$stamp1 = microtime(true);
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+									$ch = curl_init();
 									curl_setopt($ch, CURLOPT_URL, $url);
 									curl_setopt($ch, CURLOPT_HEADER, 1);
-												Check for filesize before trying to fetch site infos.

											
										
										
											2015-01-04 20:04:59 +01:00
+									curl_setopt($ch, CURLOPT_NOBODY, 1);
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+									curl_setopt($ch, CURLOPT_TIMEOUT, 3);
 									curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
-												The useragent string doesn't contains of the words "Mozilla" anymore - but the name "Friendica" and the base url

											
										
										
											2014-09-07 10:47:37 +02:00
+									curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
-												Improved handling with wrong formatted urls and fetchinfo

											
										
										
											2016-11-12 21:23:00 +01:00
+									curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
 									curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, (($check_cert) ? 2 : false));
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
 									$header = curl_exec($ch);
-												Suppoorting Open Graph and Dublin Core when displaying single item
Following link shorteners in parse_url

											
										
										
											2013-03-02 14:46:06 +01:00
+									$curl_info = @curl_getinfo($ch);
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									$http_code = $curl_info['http_code'];
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+									curl_close($ch);
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									$a->save_timestamp($stamp1, "network");
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+									if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
-												Suppoorting Open Graph and Dublin Core when displaying single item
Following link shorteners in parse_url

											
										
										
											2013-03-02 14:46:06 +01:00
+										AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
 										if ($curl_info['redirect_url'] != "")
-												Preventing an endless loop while fetching site infos

											
										
										
											2014-07-14 08:37:40 +02:00
+											$siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
-												Suppoorting Open Graph and Dublin Core when displaying single item
Following link shorteners in parse_url

											
										
										
											2013-03-02 14:46:06 +01:00
+										else
-												Preventing an endless loop while fetching site infos

											
										
										
											2014-07-14 08:37:40 +02:00
+											$siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
-												Suppoorting Open Graph and Dublin Core when displaying single item
Following link shorteners in parse_url

											
										
										
											2013-03-02 14:46:06 +01:00
+										return($siteinfo);
 									}
-												Check for filesize before trying to fetch site infos.

											
										
										
											2015-01-04 20:04:59 +01:00
+									// if the file is too large then exit
 									if ($curl_info["download_content_length"] > 1000000)
 										return($siteinfo);
 									// if it isn't a HTML file then exit
 									if (($curl_info["content_type"] != "") AND !strstr(strtolower($curl_info["content_type"]),"html"))
 										return($siteinfo);
-												The order for the page parsing is changed.

											
										
										
											2015-04-05 20:40:31 +02:00
+									if ($do_oembed) {
 										require_once("include/oembed.php");
 										$oembed_data = oembed_fetch_url($url);
-												Improved handling with wrong formatted urls and fetchinfo

											
										
										
											2016-11-12 21:23:00 +01:00
+										if (!in_array($oembed_data->type, array("error", "rich"))) {
-												The order for the page parsing is changed.

											
										
										
											2015-04-05 20:40:31 +02:00
+											$siteinfo["type"] = $oembed_data->type;
-												Improved handling with wrong formatted urls and fetchinfo

											
										
										
											2016-11-12 21:23:00 +01:00
+										}
-												The order for the page parsing is changed.

											
										
										
											2015-04-05 20:40:31 +02:00
 										if (($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
 											if (isset($oembed_data->title))
 												$siteinfo["title"] = $oembed_data->title;
 											if (isset($oembed_data->description))
 												$siteinfo["text"] = trim($oembed_data->description);
 											if (isset($oembed_data->thumbnail_url))
 												$siteinfo["image"] = $oembed_data->thumbnail_url;
 										}
 									}
-												The timestamp was at the wrong place

											
										
										
											2015-04-06 13:49:57 +02:00
+									$stamp1 = microtime(true);
-												Check for filesize before trying to fetch site infos.

											
										
										
											2015-01-04 20:04:59 +01:00
+									// Now fetch the body as well
 									$ch = curl_init();
 									curl_setopt($ch, CURLOPT_URL, $url);
 									curl_setopt($ch, CURLOPT_HEADER, 1);
 									curl_setopt($ch, CURLOPT_NOBODY, 0);
 									curl_setopt($ch, CURLOPT_TIMEOUT, 10);
 									curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 									curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
-												Improved handling with wrong formatted urls and fetchinfo

											
										
										
											2016-11-12 21:23:00 +01:00
+									curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
 									curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, (($check_cert) ? 2 : false));
-												Check for filesize before trying to fetch site infos.

											
										
										
											2015-01-04 20:04:59 +01:00
 									$header = curl_exec($ch);
 									$curl_info = @curl_getinfo($ch);
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									$http_code = $curl_info['http_code'];
-												Check for filesize before trying to fetch site infos.

											
										
										
											2015-01-04 20:04:59 +01:00
+									curl_close($ch);
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									$a->save_timestamp($stamp1, "network");
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+									// Fetch the first mentioned charset. Can be in body or header
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+									$charset = "";
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+									if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
-												parse_url: Removing of warnings with invalid charsets

											
										
										
											2014-07-24 22:50:56 +02:00
+										$charset = trim(trim(trim(array_pop($matches)), ';,'));
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
 									if ($charset == "")
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+										$charset = "utf-8";
 									$pos = strpos($header, "\r\n\r\n");
 									if ($pos)
 										$body = trim(substr($header, $pos));
 									else
 										$body = $header;
-												parse_url: Removing of warnings with invalid charsets

											
										
										
											2014-07-24 22:50:56 +02:00
+									if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
 										logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
 										//$body = mb_convert_encoding($body, "UTF-8", $charset);
 										$body = iconv($charset, "UTF-8//TRANSLIT", $body);
 									}
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+									$body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 									$doc = new DOMDocument();
 									@$doc->loadHTML($body);
 									deletenode($doc, 'style');
 									deletenode($doc, 'script');
 									deletenode($doc, 'option');
 									deletenode($doc, 'h1');
 									deletenode($doc, 'h2');
 									deletenode($doc, 'h3');
 									deletenode($doc, 'h4');
 									deletenode($doc, 'h5');
 									deletenode($doc, 'h6');
 									deletenode($doc, 'ol');
 									deletenode($doc, 'ul');
 									$xpath = new DomXPath($doc);
-												community: Only show top postings, no comments
parse_url: Follow location in "meta"-header

											
										
										
											2013-02-24 12:54:53 +01:00
+									$list = $xpath->query("//meta[@content]");
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									foreach ($list as $node) {
 										$attr = array();
 										if ($node->attributes->length)
 											foreach ($node->attributes as $attribute)
 												$attr[$attribute->name] = $attribute->value;
 										if (@$attr["http-equiv"] == 'refresh') {
 											$path = $attr["content"];
 											$pathinfo = explode(";", $path);
 											$content = "";
 											foreach ($pathinfo AS $value) {
 												if (substr(strtolower($value), 0, 4) == "url=")
 													$content = substr($value, 4);
 											}
 											if ($content != "") {
 												$siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
 												return($siteinfo);
 											}
 										}
-												community: Only show top postings, no comments
parse_url: Follow location in "meta"-header

											
										
										
											2013-02-24 12:54:53 +01:00
+									}
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+									$list = $xpath->query("//title");
-												Always parse the first title element

											
										
										
											2016-06-22 21:00:58 +02:00
+									if ($list->length > 0)
 										$siteinfo["title"] = $list->item(0)->nodeValue;
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+									//$list = $xpath->query("head/meta[@name]");
 									$list = $xpath->query("//meta[@name]");
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+									foreach ($list as $node) {
 										$attr = array();
 										if ($node->attributes->length)
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+											foreach ($node->attributes as $attribute)
 												$attr[$attribute->name] = $attribute->value;
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
-												parse_url: Trim the content of the description field.

											
										
										
											2014-03-02 01:00:36 +01:00
+										$attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+										if ($attr["content"] != "")
 											switch (strtolower($attr["name"])) {
 												case "fulltitle":
 													$siteinfo["title"] = $attr["content"];
 													break;
 												case "description":
 													$siteinfo["text"] = $attr["content"];
 													break;
-												In "parse_url" now links with [ or ] are replaced with shortened versions to avoid parsing problems.

											
										
										
											2014-12-01 22:54:01 +01:00
+												case "thumbnail":
 													$siteinfo["image"] = $attr["content"];
 													break;
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+												case "twitter:image":
 													$siteinfo["image"] = $attr["content"];
 													break;
-												Security issue: Encoding of GUID in itemcache to avoid directory bypassing with a malificious formatted GUID.

											
										
										
											2014-09-27 12:49:00 +02:00
+												case "twitter:image:src":
 													$siteinfo["image"] = $attr["content"];
 													break;
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+												case "twitter:card":
-												Building a central functionality to export postings to all networks that only support plaintext postings. In near future this will replace the similar code in the several addons.

											
										
										
											2014-05-03 12:04:54 +02:00
+													if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+														$siteinfo["type"] = $attr["content"];
 													break;
 												case "twitter:description":
 													$siteinfo["text"] = $attr["content"];
 													break;
 												case "twitter:title":
 													$siteinfo["title"] = $attr["content"];
 													break;
 												case "dc.title":
 													$siteinfo["title"] = $attr["content"];
 													break;
 												case "dc.description":
 													$siteinfo["text"] = $attr["content"];
 													break;
-												Creating hashtags from keywords of a feed, when additional data should be fetched.

											
										
										
											2014-10-03 12:18:33 +02:00
+												case "keywords":
 													$keywords = explode(",", $attr["content"]);
 													break;
 												case "news_keywords":
 													$keywords = explode(",", $attr["content"]);
 													break;
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+											}
-												parse url: Improvements for detecting the page type

											
										
										
											2014-03-16 17:48:28 +01:00
+										if ($siteinfo["type"] == "summary")
 											$siteinfo["type"] = "link";
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+									}
-												Creating hashtags from keywords of a feed, when additional data should be fetched.

											
										
										
											2014-10-03 12:18:33 +02:00
+									if (isset($keywords)) {
 										$siteinfo["keywords"] = array();
 										foreach ($keywords as $keyword)
-												Dupilcated tags shouldn't happen anymore when parsing a webpage.

											
										
										
											2015-11-28 13:15:00 +01:00
+											if (!in_array(trim($keyword), $siteinfo["keywords"]))
 												$siteinfo["keywords"][] = trim($keyword);
-												Creating hashtags from keywords of a feed, when additional data should be fetched.

											
										
										
											2014-10-03 12:18:33 +02:00
+									}
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+									//$list = $xpath->query("head/meta[@property]");
 									$list = $xpath->query("//meta[@property]");
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+									foreach ($list as $node) {
 										$attr = array();
 										if ($node->attributes->length)
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+											foreach ($node->attributes as $attribute)
 												$attr[$attribute->name] = $attribute->value;
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
-												parse_url: Trim the content of the description field.

											
										
										
											2014-03-02 01:00:36 +01:00
+										$attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+										if ($attr["content"] != "")
 											switch (strtolower($attr["property"])) {
 												case "og:image":
 													$siteinfo["image"] = $attr["content"];
 													break;
 												case "og:title":
 													$siteinfo["title"] = $attr["content"];
 													break;
 												case "og:description":
 													$siteinfo["text"] = $attr["content"];
 													break;
 											}
 									}
-												Optional parameter for parseurl to only fetch data from meta information

											
										
										
											2013-12-02 20:24:41 +01:00
+									if ((@$siteinfo["image"] == "") AND !$no_guessing) {
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+									    $list = $xpath->query("//img[@src]");
 									    foreach ($list as $node) {
 										$attr = array();
 										if ($node->attributes->length)
 										    foreach ($node->attributes as $attribute)
 											$attr[$attribute->name] = $attribute->value;
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+											$src = completeurl($attr["src"], $url);
-												When a server isn't reachable there were timeout problems with fetching image data

											
										
										
											2016-01-18 15:38:38 +01:00
+											$photodata = get_photo_info($src);
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
-												allow comments on other folks wall-to-wall postings

											
										
										
											2012-09-06 03:24:06 +02:00
+											if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+												if ($photodata[0] > 300) {
-												parse_url: Problem when resizing images

											
										
										
											2012-07-14 13:59:42 +02:00
+													$photodata[1] = round($photodata[1] * (300 / $photodata[0]));
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+													$photodata[0] = 300;
 												}
 												if ($photodata[1] > 300) {
-												parse_url: Problem when resizing images

											
										
										
											2012-07-14 13:59:42 +02:00
+													$photodata[0] = round($photodata[0] * (300 / $photodata[1]));
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+													$photodata[1] = 300;
 												}
 												$siteinfo["images"][] = array("src"=>$src,
 																"width"=>$photodata[0],
 																"height"=>$photodata[1]);
-												parse_url: Fetch multiple pictures so that the user can decide what to take.

											
										
										
											2012-07-12 08:20:27 +02:00
+											}
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
-												Added some more timestamp measuring

											
										
										
											2015-03-07 23:14:26 +01:00
+										}
-												When a server isn't reachable there were timeout problems with fetching image data

											
										
										
											2016-01-18 15:38:38 +01:00
+								    } elseif ($siteinfo["image"] != "") {
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+										$src = completeurl($siteinfo["image"], $url);
-												parse_url: Fixed a problem in the detection if there was an image or not. Additionally removed some test content.

											
										
										
											2012-08-02 10:25:08 +02:00
 										unset($siteinfo["image"]);
-												When a server isn't reachable there were timeout problems with fetching image data

											
										
										
											2016-01-18 15:38:38 +01:00
+										$photodata = get_photo_info($src);
-												parse_url: Fetch multiple pictures so that the user can decide what to take.

											
										
										
											2012-07-12 08:20:27 +02:00
-												allow comments on other folks wall-to-wall postings

											
										
										
											2012-09-06 03:24:06 +02:00
+										if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+											$siteinfo["images"][] = array("src"=>$src,
 															"width"=>$photodata[0],
 															"height"=>$photodata[1]);
-												parse_url: Fetch multiple pictures so that the user can decide what to take.

											
										
										
											2012-07-12 08:20:27 +02:00
+									}
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
-												Optional parameter for parseurl to only fetch data from meta information

											
										
										
											2013-12-02 20:24:41 +01:00
+									if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+										$text = "";
 										$list = $xpath->query("//div[@class='article']");
 										foreach ($list as $node)
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+											if (strlen($node->nodeValue) > 40)
 												$text .= " ".trim($node->nodeValue);
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
 										if ($text == "") {
 											$list = $xpath->query("//div[@class='content']");
 											foreach ($list as $node)
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+												if (strlen($node->nodeValue) > 40)
 													$text .= " ".trim($node->nodeValue);
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+										}
 										// If none text was found then take the paragraph content
 										if ($text == "") {
 											$list = $xpath->query("//p");
 											foreach ($list as $node)
-												parse_url: Further improvements of the new method to fetch page data

											
										
										
											2012-07-12 23:41:04 +02:00
+												if (strlen($node->nodeValue) > 40)
 													$text .= " ".trim($node->nodeValue);
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+										}
 										if ($text != "") {
 											$text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 											while (strpos($text, "  "))
 												$text = trim(str_replace("  ", " ", $text));
-												parse_url: Trim the content of the description field.

											
										
										
											2014-03-02 01:00:36 +01:00
+											$siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+										}
 									}
-												parse url: Emtpy values in meta headers are now ignored. Oembed data has priority. Videos are embedded with some further information.

											
										
										
											2014-04-04 10:58:31 +02:00
+									logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
-												New hook "getsiteinfo", separating fetching site info from generating bbcode

											
										
										
											2014-10-20 08:03:47 +02:00
+									call_hooks('getsiteinfo', $siteinfo);
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+									return($siteinfo);
 								}
-												begin plugin api

											
										
										
											2010-12-21 04:38:34 +01:00
-												social bookmark API, oexchange enhancements

											
										
										
											2011-09-21 01:31:45 +02:00
+								function arr_add_hashes(&$item,$k) {
 									$item = '#' . $item;
 								}
-												ajaxify the wall poster

											
										
										
											2010-07-23 07:41:45 +02:00
+								function parse_url_content(&$a) {
-												enhance url parse

											
										
										
											2010-07-23 08:17:41 +02:00
-												Use the "attachment" element instead of a mixture of several elements

											
										
										
											2016-04-17 20:43:41 +02:00
+									require_once("include/items.php");
-												allow setting all the bookmarked URL fields via script/API

											
										
										
											2011-09-20 07:21:55 +02:00
+									$text = null;
-												social bookmark API, oexchange enhancements

											
										
										
											2011-09-21 01:31:45 +02:00
+									$str_tags = '';
-												allow setting all the bookmarked URL fields via script/API

											
										
										
											2011-09-20 07:21:55 +02:00
-												more fixes for textmode

											
										
										
											2012-02-07 09:13:16 +01:00
+									$textmode = false;
-												fix "insert web link" for those without richtext feature enabled

											
										
										
											2012-12-04 01:31:28 +01:00
+									if(local_user() && (! feature_enabled(local_user(),'richtext')))
 										$textmode = true;
-												parse_url: Fetch multiple pictures so that the user can decide what to take.

											
										
										
											2012-07-12 08:20:27 +02:00
+									//if($textmode)
 									$br = (($textmode) ? "\n" : '<br />');
-												more fixes for textmode

											
										
										
											2012-02-07 09:13:16 +01:00
-												allow setting all the bookmarked URL fields via script/API

											
										
										
											2011-09-20 07:21:55 +02:00
+									if(x($_GET,'binurl'))
 										$url = trim(hex2bin($_GET['binurl']));
 									else
 										$url = trim($_GET['url']);
 									if($_GET['title'])
 										$title = strip_tags(trim($_GET['title']));
-												bug #37, ampersand + everything following in url is stripped when using link icon to share link

											
										
										
											2011-04-10 12:36:12 +02:00
-												social bookmark API, oexchange enhancements

											
										
										
											2011-09-21 01:31:45 +02:00
+									if($_GET['description'])
 										$text = strip_tags(trim($_GET['description']));
 									if($_GET['tags']) {
 										$arr_tags = str_getcsv($_GET['tags']);
 										if(count($arr_tags)) {
 											array_walk($arr_tags,'arr_add_hashes');
-												more fixes for textmode

											
										
										
											2012-02-07 09:13:16 +01:00
+											$str_tags = $br . implode(' ',$arr_tags) . $br;
-												social bookmark API, oexchange enhancements

											
										
										
											2011-09-21 01:31:45 +02:00
+										}
 									}
-												bug #37, ampersand + everything following in url is stripped when using link icon to share link

											
										
										
											2011-04-10 12:36:12 +02:00
-												Add scheme if missing to url inserted by user in 'insert link' jot
button

											
										
										
											2015-03-22 10:53:24 +01:00
+									// add url scheme if missing
 									$arrurl = parse_url($url);
 									if (!x($arrurl, 'scheme')) {
 										if (x($arrurl, 'host'))
 											$url = "http:".$url;
 										else
 											$url = "http://".$url;
 									}
-												bug #37, ampersand + everything following in url is stripped when using link icon to share link

											
										
										
											2011-04-10 12:36:12 +02:00
+									logger('parse_url: ' . $url);
-												ajaxify the wall poster

											
										
										
											2010-07-23 07:41:45 +02:00
-												more fixes for textmode

											
										
										
											2012-02-07 09:13:16 +01:00
+									if($textmode)
-												Support for twitter cards (in both directions)

											
										
										
											2014-02-22 15:46:19 +01:00
+										$template = '[bookmark=%s]%s[/bookmark]%s';
-												more fixes for textmode

											
										
										
											2012-02-07 09:13:16 +01:00
+									else
-												Support for twitter cards (in both directions)

											
										
										
											2014-02-22 15:46:19 +01:00
+										$template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
-												ajaxify the wall poster

											
										
										
											2010-07-23 07:41:45 +02:00
-												more plugin hooks

											
										
										
											2010-12-26 00:01:02 +01:00
+									$arr = array('url' => $url, 'text' => '');
 									call_hooks('parse_link', $arr);
 									if(strlen($arr['text'])) {
 										echo $arr['text'];
 										killme();
 									}
-												better handling of remote urls - especially FB opengraph

											
										
										
											2011-10-28 11:50:00 +02:00
-												allow setting all the bookmarked URL fields via script/API

											
										
										
											2011-09-20 07:21:55 +02:00
+									if($url && $title && $text) {
-												parse url: Characters like < and > has to be escaped when showing the parsed output.

											
										
										
											2014-02-16 17:35:01 +01:00
+										$title = str_replace(array("\r","\n"),array('',''),$title);
-												more fixes for textmode

											
										
										
											2012-02-07 09:13:16 +01:00
+										if($textmode)
-												Merge remote-tracking branch 'upstream/master'

Conflicts:
	mod/crepair.php

											
										
										
											2014-01-05 16:22:42 +01:00
+											$text = '[quote]' . trim($text) . '[/quote]' . $br;
-												parse url: Characters like < and > has to be escaped when showing the parsed output.

											
										
										
											2014-02-16 17:35:01 +01:00
+										else {
 											$text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
 											$title = htmlspecialchars($title);
 										}
-												allow setting all the bookmarked URL fields via script/API

											
										
										
											2011-09-20 07:21:55 +02:00
-												social bookmark API, oexchange enhancements

											
										
										
											2011-09-21 01:31:45 +02:00
+										$result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
-												allow setting all the bookmarked URL fields via script/API

											
										
										
											2011-09-20 07:21:55 +02:00
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+										logger('parse_url (unparsed): returns: ' . $result);
-												allow setting all the bookmarked URL fields via script/API

											
										
										
											2011-09-20 07:21:55 +02:00
 										echo $result;
 										killme();
 									}
-												parse_url: complete new code for fetching website information

											
										
										
											2012-07-12 01:17:33 +02:00
+									$siteinfo = parseurl_getsiteinfo($url);
-												allow setting all the bookmarked URL fields via script/API

											
										
										
											2011-09-20 07:21:55 +02:00
-												Use the "attachment" element instead of a mixture of several elements

											
										
										
											2016-04-17 20:43:41 +02:00
+									unset($siteinfo["keywords"]);
-												Merge remote-tracking branch 'upstream/master'

Conflicts:
	mod/crepair.php

											
										
										
											2014-01-05 16:22:42 +01:00
-												Bugfix: ' in titles were a problem.

											
										
										
											2016-04-18 20:57:01 +02:00
+									$info = add_page_info_data($siteinfo);
 									if (!$textmode)
 										// Replace ' with ’ - not perfect - but the richtext editor has problems otherwise
 										$info = str_replace(array("&#039;"), array("&#8217;"), $info);
 									echo $info;
-												ajaxify the wall poster

											
										
										
											2010-07-23 07:41:45 +02:00
+									killme();
-												Add oembed user option for use oembed instead of standard yotube embedding.
Remove global option and parse_url use of oembed.

											
										
										
											2011-05-23 10:37:09 +02:00
+								}
-												Building a central functionality to export postings to all networks that only support plaintext postings. In near future this will replace the similar code in the several addons.

											
										
										
											2014-05-03 12:04:54 +02:00
+								?>