Merge pull request #5715 from MrPetovan/bug/fix-photo-attachment

Fix 'photo' attachment
This commit is contained in:
Michael Vogel 2018-09-03 22:00:53 +00:00 committed by GitHub
commit 62c5e5d861
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 207 additions and 203 deletions

View file

@ -8,128 +8,127 @@
* information and does format this information to BBCode
*
* @see ParseUrl::getSiteinfo() for more information about scraping embeddable content
*/
*/
use Friendica\App;
use Friendica\Core\Addon;
use Friendica\Util\Network;
use Friendica\Util\ParseUrl;
require_once("include/items.php");
function parse_url_content(App $a) {
require_once 'include/items.php';
function parse_url_content(App $a)
{
$text = null;
$str_tags = "";
$str_tags = '';
$br = "\n";
if (!empty($_GET["binurl"])) {
$url = trim(hex2bin($_GET["binurl"]));
if (!empty($_GET['binurl'])) {
$url = trim(hex2bin($_GET['binurl']));
} else {
$url = trim($_GET["url"]);
$url = trim($_GET['url']);
}
if (!empty($_GET["title"])) {
$title = strip_tags(trim($_GET["title"]));
if (!empty($_GET['title'])) {
$title = strip_tags(trim($_GET['title']));
}
if (!empty($_GET["description"])) {
$text = strip_tags(trim($_GET["description"]));
if (!empty($_GET['description'])) {
$text = strip_tags(trim($_GET['description']));
}
if (!empty($_GET["tags"])) {
$arr_tags = ParseUrl::convertTagsToArray($_GET["tags"]);
if (!empty($_GET['tags'])) {
$arr_tags = ParseUrl::convertTagsToArray($_GET['tags']);
if (count($arr_tags)) {
$str_tags = $br . implode(" ", $arr_tags) . $br;
$str_tags = $br . implode(' ', $arr_tags) . $br;
}
}
// Add url scheme if it is missing
$arrurl = parse_url($url);
if (!x($arrurl, "scheme")) {
if (x($arrurl, "host")) {
$url = "http:".$url;
if (!x($arrurl, 'scheme')) {
if (x($arrurl, 'host')) {
$url = 'http:' . $url;
} else {
$url = "http://".$url;
$url = 'http://' . $url;
}
}
logger("prse_url: " . $url);
logger($url);
// Check if the URL is an image, video or audio file. If so format
// the URL with the corresponding BBCode media tag
$redirects = 0;
// Fetch the header of the URL
$result = Network::curl($url, false, $redirects, ["novalidate" => true, "nobody" => true]);
if($result["success"]) {
$result = Network::curl($url, false, $redirects, ['novalidate' => true, 'nobody' => true]);
if ($result['success']) {
// Convert the header fields into an array
$hdrs = [];
$h = explode("\n", $result["header"]);
$h = explode("\n", $result['header']);
foreach ($h as $l) {
$header = array_map("trim", explode(":", trim($l), 2));
$header = array_map('trim', explode(':', trim($l), 2));
if (count($header) == 2) {
list($k,$v) = $header;
list($k, $v) = $header;
$hdrs[$k] = $v;
}
}
if (array_key_exists("Content-Type", $hdrs)) {
$type = $hdrs["Content-Type"];
if (array_key_exists('Content-Type', $hdrs)) {
$type = $hdrs['Content-Type'];
}
if ($type) {
if(stripos($type, "image/") !== false) {
echo $br . "[img]" . $url . "[/img]" . $br;
killme();
if (stripos($type, 'image/') !== false) {
echo $br . '[img]' . $url . '[/img]' . $br;
exit();
}
if (stripos($type, "video/") !== false) {
echo $br . "[video]" . $url . "[/video]" . $br;
killme();
if (stripos($type, 'video/') !== false) {
echo $br . '[video]' . $url . '[/video]' . $br;
exit();
}
if (stripos($type, "audio/") !== false) {
echo $br . "[audio]" . $url . "[/audio]" . $br;
killme();
if (stripos($type, 'audio/') !== false) {
echo $br . '[audio]' . $url . '[/audio]' . $br;
exit();
}
}
}
$template = "[bookmark=%s]%s[/bookmark]%s";
$template = '[bookmark=%s]%s[/bookmark]%s';
$arr = ["url" => $url, "text" => ""];
$arr = ['url' => $url, 'text' => ''];
Addon::callHooks("parse_link", $arr);
Addon::callHooks('parse_link', $arr);
if (strlen($arr["text"])) {
echo $arr["text"];
killme();
if (strlen($arr['text'])) {
echo $arr['text'];
exit();
}
// If there is already some content information submitted we don't
// need to parse the url for content.
if (!empty($url) && !empty($title) && !empty($text)) {
$title = str_replace(["\r", "\n"], ['', ''], $title);
$title = str_replace(["\r","\n"],["",""],$title);
$text = "[quote]" . trim($text) . "[/quote]" . $br;
$text = '[quote]' . trim($text) . '[/quote]' . $br;
$result = sprintf($template, $url, ($title) ? $title : $url, $text) . $str_tags;
logger("parse_url (unparsed): returns: " . $result);
logger('(unparsed): returns: ' . $result);
echo $result;
killme();
exit();
}
// Fetch the information directly from the webpage
$siteinfo = ParseUrl::getSiteinfo($url);
unset($siteinfo["keywords"]);
unset($siteinfo['keywords']);
// Format it as BBCode attachment
$info = add_page_info_data($siteinfo);
echo $info;
killme();
exit();
}
/**
@ -151,7 +150,8 @@ function parse_url_content(App $a) {
* @todo Remove this function after all Addons has been changed to use
* ParseUrl::getSiteinfoCached
*/
function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) {
function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true)
{
$siteinfo = ParseUrl::getSiteinfoCached($url, $no_guessing, $do_oembed);
return $siteinfo;
}

View file

@ -578,9 +578,7 @@ class BBCode extends BaseObject
$return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], self::proxyUrl($data["preview"], $simplehtml), $data["title"]);
}
if (($data["type"] == "photo") && !empty($data["url"]) && !empty($data["image"])) {
$return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]);
} else {
if (!empty($data['title']) && !empty($data['url'])) {
$return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
}
@ -589,7 +587,8 @@ class BBCode extends BaseObject
$bbcode = HTML::toBBCode($data["description"]);
$return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
}
if ($data["type"] == "link") {
if (!empty($data['url'])) {
$return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
}

View file

@ -118,13 +118,13 @@ class ParseUrl
// Check if the URL does contain a scheme
$scheme = parse_url($url, PHP_URL_SCHEME);
if ($scheme == "") {
$url = "http://".trim($url, "/");
if ($scheme == '') {
$url = 'http://' . trim($url, '/');
}
if ($count > 10) {
logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
return($siteinfo);
logger('Endless loop detected for ' . $url, LOGGER_DEBUG);
return $siteinfo;
}
$url = trim($url, "'");
@ -132,220 +132,223 @@ class ParseUrl
$url = Network::stripTrackingQueryParams($url);
$siteinfo["url"] = $url;
$siteinfo["type"] = "link";
$siteinfo['url'] = $url;
$siteinfo['type'] = 'link';
$data = Network::curl($url);
if (!$data['success']) {
return($siteinfo);
return $siteinfo;
}
// If the file is too large then exit
if ($data["info"]["download_content_length"] > 1000000) {
return($siteinfo);
if ($data['info']['download_content_length'] > 1000000) {
return $siteinfo;
}
// If it isn't a HTML file then exit
if (($data["info"]["content_type"] != "") && !strstr(strtolower($data["info"]["content_type"]), "html")) {
return($siteinfo);
if (($data['info']['content_type'] != '') && !strstr(strtolower($data['info']['content_type']), 'html')) {
return $siteinfo;
}
$header = $data["header"];
$body = $data["body"];
$header = $data['header'];
$body = $data['body'];
if ($do_oembed) {
$oembed_data = OEmbed::fetchURL($url);
if (!empty($oembed_data->type)) {
if (!in_array($oembed_data->type, ["error", "rich", ""])) {
$siteinfo["type"] = $oembed_data->type;
if (!in_array($oembed_data->type, ['error', 'rich', ''])) {
$siteinfo['type'] = $oembed_data->type;
}
if (($oembed_data->type == "link") && ($siteinfo["type"] != "photo")) {
if (($oembed_data->type == 'link') && ($siteinfo['type'] != 'photo')) {
if (isset($oembed_data->title)) {
$siteinfo["title"] = trim($oembed_data->title);
$siteinfo['title'] = trim($oembed_data->title);
}
if (isset($oembed_data->description)) {
$siteinfo["text"] = trim($oembed_data->description);
$siteinfo['text'] = trim($oembed_data->description);
}
if (isset($oembed_data->thumbnail_url)) {
$siteinfo["image"] = $oembed_data->thumbnail_url;
$siteinfo['image'] = $oembed_data->thumbnail_url;
}
}
}
}
// Fetch the first mentioned charset. Can be in body or header
$charset = "";
if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches)) {
$charset = '';
if (preg_match('/charset=(.*?)[\'"\s\n]/', $header, $matches)) {
$charset = trim(trim(trim(array_pop($matches)), ';,'));
}
if ($charset == "") {
$charset = "utf-8";
if ($charset == '') {
$charset = 'utf-8';
}
if (($charset != "") && (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", $body);
if (($charset != '') && (strtoupper($charset) != 'UTF-8')) {
logger('detected charset ' . $charset, LOGGER_DEBUG);
$body = iconv($charset, 'UTF-8//TRANSLIT', $body);
}
$body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
$body = mb_convert_encoding($body, 'HTML-ENTITIES', 'UTF-8');
$doc = new DOMDocument();
@$doc->loadHTML($body);
XML::deleteNode($doc, "style");
XML::deleteNode($doc, "script");
XML::deleteNode($doc, "option");
XML::deleteNode($doc, "h1");
XML::deleteNode($doc, "h2");
XML::deleteNode($doc, "h3");
XML::deleteNode($doc, "h4");
XML::deleteNode($doc, "h5");
XML::deleteNode($doc, "h6");
XML::deleteNode($doc, "ol");
XML::deleteNode($doc, "ul");
XML::deleteNode($doc, 'style');
XML::deleteNode($doc, 'script');
XML::deleteNode($doc, 'option');
XML::deleteNode($doc, 'h1');
XML::deleteNode($doc, 'h2');
XML::deleteNode($doc, 'h3');
XML::deleteNode($doc, 'h4');
XML::deleteNode($doc, 'h5');
XML::deleteNode($doc, 'h6');
XML::deleteNode($doc, 'ol');
XML::deleteNode($doc, 'ul');
$xpath = new DOMXPath($doc);
$list = $xpath->query("//meta[@content]");
$list = $xpath->query('//meta[@content]');
foreach ($list as $node) {
$attr = [];
$meta_tag = [];
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
$attr[$attribute->name] = $attribute->value;
$meta_tag[$attribute->name] = $attribute->value;
}
}
if (@$attr["http-equiv"] == "refresh") {
$path = $attr["content"];
$pathinfo = explode(";", $path);
$content = "";
if (@$meta_tag['http-equiv'] == 'refresh') {
$path = $meta_tag['content'];
$pathinfo = explode(';', $path);
$content = '';
foreach ($pathinfo as $value) {
if (substr(strtolower($value), 0, 4) == "url=") {
if (substr(strtolower($value), 0, 4) == 'url=') {
$content = substr($value, 4);
}
}
if ($content != "") {
if ($content != '') {
$siteinfo = self::getSiteinfo($content, $no_guessing, $do_oembed, ++$count);
return($siteinfo);
return $siteinfo;
}
}
}
$list = $xpath->query("//title");
$list = $xpath->query('//title');
if ($list->length > 0) {
$siteinfo["title"] = trim($list->item(0)->nodeValue);
$siteinfo['title'] = trim($list->item(0)->nodeValue);
}
//$list = $xpath->query("head/meta[@name]");
$list = $xpath->query("//meta[@name]");
$list = $xpath->query('//meta[@name]');
foreach ($list as $node) {
$attr = [];
$meta_tag = [];
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
$attr[$attribute->name] = $attribute->value;
$meta_tag[$attribute->name] = $attribute->value;
}
}
if (!empty($attr["content"])) {
$attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
if (empty($meta_tag['content'])) {
continue;
}
switch (strtolower($attr["name"])) {
case "fulltitle":
$siteinfo["title"] = trim($attr["content"]);
break;
case "description":
$siteinfo["text"] = trim($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"] == "") || ($attr["content"] == "photo")) {
$siteinfo["type"] = $attr["content"];
}
break;
case "twitter:description":
$siteinfo["text"] = trim($attr["content"]);
break;
case "twitter:title":
$siteinfo["title"] = trim($attr["content"]);
break;
case "dc.title":
$siteinfo["title"] = trim($attr["content"]);
break;
case "dc.description":
$siteinfo["text"] = trim($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";
$meta_tag['content'] = trim(html_entity_decode($meta_tag['content'], ENT_QUOTES, 'UTF-8'));
switch (strtolower($meta_tag['name'])) {
case 'fulltitle':
$siteinfo['title'] = trim($meta_tag['content']);
break;
case 'description':
$siteinfo['text'] = trim($meta_tag['content']);
break;
case 'thumbnail':
$siteinfo['image'] = $meta_tag['content'];
break;
case 'twitter:image':
$siteinfo['image'] = $meta_tag['content'];
break;
case 'twitter:image:src':
$siteinfo['image'] = $meta_tag['content'];
break;
case 'twitter:card':
// Obsolete card type
if ($meta_tag['content'] == 'photo') {
$siteinfo['type'] = 'summary_large_image';
} else {
$siteinfo['type'] = $meta_tag['content'];
}
break;
case 'twitter:description':
$siteinfo['text'] = trim($meta_tag['content']);
break;
case 'twitter:title':
$siteinfo['title'] = trim($meta_tag['content']);
break;
case 'dc.title':
$siteinfo['title'] = trim($meta_tag['content']);
break;
case 'dc.description':
$siteinfo['text'] = trim($meta_tag['content']);
break;
case 'keywords':
$keywords = explode(',', $meta_tag['content']);
break;
case 'news_keywords':
$keywords = explode(',', $meta_tag['content']);
break;
}
}
if ($siteinfo['type'] == 'summary' || $siteinfo['type'] == 'summary_large_image') {
$siteinfo['type'] = 'link';
}
if (isset($keywords)) {
$siteinfo["keywords"] = [];
$siteinfo['keywords'] = [];
foreach ($keywords as $keyword) {
if (!in_array(trim($keyword), $siteinfo["keywords"])) {
$siteinfo["keywords"][] = trim($keyword);
if (!in_array(trim($keyword), $siteinfo['keywords'])) {
$siteinfo['keywords'][] = trim($keyword);
}
}
}
//$list = $xpath->query("head/meta[@property]");
$list = $xpath->query("//meta[@property]");
$list = $xpath->query('//meta[@property]');
foreach ($list as $node) {
$attr = [];
$meta_tag = [];
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
$attr[$attribute->name] = $attribute->value;
$meta_tag[$attribute->name] = $attribute->value;
}
}
if (!empty($attr["content"])) {
$attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
if (!empty($meta_tag['content'])) {
$meta_tag['content'] = trim(html_entity_decode($meta_tag['content'], ENT_QUOTES, 'UTF-8'));
switch (strtolower($attr["property"])) {
case "og:image":
$siteinfo["image"] = $attr["content"];
switch (strtolower($meta_tag['property'])) {
case 'og:image':
$siteinfo['image'] = $meta_tag['content'];
break;
case "og:title":
$siteinfo["title"] = trim($attr["content"]);
case 'og:title':
$siteinfo['title'] = trim($meta_tag['content']);
break;
case "og:description":
$siteinfo["text"] = trim($attr["content"]);
case 'og:description':
$siteinfo['text'] = trim($meta_tag['content']);
break;
}
}
}
if ((@$siteinfo["image"] == "") && !$no_guessing) {
$list = $xpath->query("//img[@src]");
if ((@$siteinfo['image'] == '') && !$no_guessing) {
$list = $xpath->query('//img[@src]');
foreach ($list as $node) {
$attr = [];
$img_tag = [];
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
$attr[$attribute->name] = $attribute->value;
$img_tag[$attribute->name] = $attribute->value;
}
}
$src = self::completeUrl($attr["src"], $url);
$src = self::completeUrl($img_tag['src'], $url);
$photodata = Image::getInfoFromURL($src);
if (($photodata) && ($photodata[0] > 150) && ($photodata[1] > 150)) {
@ -357,70 +360,72 @@ class ParseUrl
$photodata[0] = round($photodata[0] * (300 / $photodata[1]));
$photodata[1] = 300;
}
$siteinfo["images"][] = ["src" => $src,
"width" => $photodata[0],
"height" => $photodata[1]];
$siteinfo['images'][] = [
'src' => $src,
'width' => $photodata[0],
'height' => $photodata[1]
];
}
}
} elseif (!empty($siteinfo["image"])) {
$src = self::completeUrl($siteinfo["image"], $url);
} elseif (!empty($siteinfo['image'])) {
$src = self::completeUrl($siteinfo['image'], $url);
unset($siteinfo["image"]);
unset($siteinfo['image']);
$photodata = Image::getInfoFromURL($src);
if (($photodata) && ($photodata[0] > 10) && ($photodata[1] > 10)) {
$siteinfo["images"][] = ["src" => $src,
"width" => $photodata[0],
"height" => $photodata[1]];
$siteinfo['images'][] = ['src' => $src,
'width' => $photodata[0],
'height' => $photodata[1]];
}
}
if ((@$siteinfo["text"] == "") && (@$siteinfo["title"] != "") && !$no_guessing) {
$text = "";
if ((@$siteinfo['text'] == '') && (@$siteinfo['title'] != '') && !$no_guessing) {
$text = '';
$list = $xpath->query("//div[@class='article']");
$list = $xpath->query('//div[@class="article"]');
foreach ($list as $node) {
if (strlen($node->nodeValue) > 40) {
$text .= " ".trim($node->nodeValue);
$text .= ' ' . trim($node->nodeValue);
}
}
if ($text == "") {
$list = $xpath->query("//div[@class='content']");
if ($text == '') {
$list = $xpath->query('//div[@class="content"]');
foreach ($list as $node) {
if (strlen($node->nodeValue) > 40) {
$text .= " ".trim($node->nodeValue);
$text .= ' ' . trim($node->nodeValue);
}
}
}
// If none text was found then take the paragraph content
if ($text == "") {
$list = $xpath->query("//p");
if ($text == '') {
$list = $xpath->query('//p');
foreach ($list as $node) {
if (strlen($node->nodeValue) > 40) {
$text .= " ".trim($node->nodeValue);
$text .= ' ' . trim($node->nodeValue);
}
}
}
if ($text != "") {
$text = trim(str_replace(["\n", "\r"], [" ", " "], $text));
if ($text != '') {
$text = trim(str_replace(["\n", "\r"], [' ', ' '], $text));
while (strpos($text, " ")) {
$text = trim(str_replace(" ", " ", $text));
while (strpos($text, ' ')) {
$text = trim(str_replace(' ', ' ', $text));
}
$siteinfo["text"] = trim(html_entity_decode(substr($text, 0, 350), ENT_QUOTES, "UTF-8").'...');
$siteinfo['text'] = trim(html_entity_decode(substr($text, 0, 350), ENT_QUOTES, 'UTF-8') . '...');
}
}
logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
logger('Siteinfo for ' . $url . ' ' . print_r($siteinfo, true), LOGGER_DEBUG);
Addon::callHooks("getsiteinfo", $siteinfo);
Addon::callHooks('getsiteinfo', $siteinfo);
return($siteinfo);
return $siteinfo;
}
/**