Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Running upstream opengraph PHP library maintenance #30

Open
wants to merge 15 commits into
base: master
Choose a base branch
from
Open

Running upstream opengraph PHP library maintenance #30

Changes from all commits
Commits
Show all changes
15 commits
Select commit Hold shift + click to select a range
3d2180e
Merge pull request #2 from scottmac/master
AramZS Jan 8, 2015
5f37fbf
Images always failing - results in no images
AramZS Jan 8, 2015
0be94f5
Use the facebook user-agent instead of $_SERVER
Aug 4, 2014
9fae605
Update OpenGraph.php
mamadoo Dec 8, 2015
cc30c6d
Merging changes from @danilor to support cURL cookies
danilor Feb 20, 2015
461aa9c
Abstracting cookie path to make this easier to integrate into other p…
AramZS Feb 7, 2016
362caf4
Pulling in the Twitter tags on the OG object
AramZS Jul 12, 2016
5f80fa2
Use empty instead of isset to avoid fatal errors when cookie set with…
AramZS Jul 12, 2016
ab010c6
Provide more options for possible images to use as the master image
AramZS Jul 12, 2016
a76a50a
Remove definition check errors for cURL cookie
AramZS Jul 12, 2016
31e6e60
Added check for existing key, if key already exists, the entry is cha…
GregersBoye Oct 8, 2015
6d71e1c
Retain backcompat in cases of multiple og:image tags
AramZS Jul 12, 2016
c703705
Read in the type-dependent schema.
AramZS Jul 12, 2016
d95a5f4
Also allow for images of width 100% to be valid replacement for missi…
AramZS Jul 12, 2016
cf70323
Provide a public-facing parse function
AramZS Jul 12, 2016
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
132 changes: 101 additions & 31 deletions OpenGraph.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,9 +13,9 @@
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

Original can be found at https://github.com/scottmac/opengraph/blob/master/OpenGraph.php

*/

class OpenGraph implements Iterator
Expand Down Expand Up @@ -50,25 +50,40 @@ class OpenGraph implements Iterator
* @return OpenGraph
*/
static public function fetch($URI) {
$curl = curl_init($URI);
$cookie_path = 'cookie.txt';
if ( defined('COOKIE_PATH_FOR_CURL') && !empty(COOKIE_PATH_FOR_CURL) ){
$cookie_path = COOKIE_PATH_FOR_CURL;
}
$curl = curl_init($URI);

curl_setopt($curl, CURLOPT_FAILONERROR, true);
curl_setopt($curl, CURLOPT_FOLLOWLOCATION, true);
curl_setopt($curl, CURLOPT_RETURNTRANSFER, true);
curl_setopt($curl, CURLOPT_TIMEOUT, 15);
curl_setopt($curl, CURLOPT_SSL_VERIFYHOST, false);
curl_setopt($curl, CURLOPT_SSL_VERIFYPEER, false);
curl_setopt($curl, CURLOPT_USERAGENT, $_SERVER['HTTP_USER_AGENT']);
curl_setopt($curl, CURLOPT_FAILONERROR, true);
curl_setopt($curl, CURLOPT_FOLLOWLOCATION, true);
curl_setopt($curl, CURLOPT_RETURNTRANSFER, true);
curl_setopt($curl, CURLOPT_TIMEOUT, 15);
curl_setopt($curl, CURLOPT_SSL_VERIFYHOST, false);
curl_setopt($curl, CURLOPT_SSL_VERIFYPEER, false);
curl_setopt($curl, CURLOPT_USERAGENT, "facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)");
//The following 2 set up lines work with sites like www.nytimes.com
curl_setopt($curl, CURLOPT_COOKIEFILE, $cookie_path); //you can change this path to whetever you want.
curl_setopt($curl, CURLOPT_COOKIEJAR, $cookie_path); //you can change this path to whetever you want.

$response = curl_exec($curl);
$response = mb_convert_encoding(curl_exec($curl), 'HTML-ENTITIES', 'UTF-8');

curl_close($curl);
curl_close($curl);

if (!empty($response)) {
return self::_parse($response);
} else {
return false;
}
if (!empty($response)) {
return self::_parse($response);
} else {
return false;
}
}

static public function parse($HTML){
if ( empty( $HTML ) ){
return false;
}
$response = mb_convert_encoding($HTML, 'HTML-ENTITIES', 'UTF-8');
return self::_parse($response);
}

/**
Expand All @@ -83,7 +98,7 @@ static private function _parse($HTML) {

$doc = new DOMDocument();
$doc->loadHTML($HTML);

libxml_use_internal_errors($old_libxml_error);

$tags = $doc->getElementsByTagName('meta');
Expand All @@ -94,16 +109,23 @@ static private function _parse($HTML) {
$page = new self();

$nonOgDescription = null;

foreach ($tags AS $tag) {
if ($tag->hasAttribute('property') &&
strpos($tag->getAttribute('property'), 'og:') === 0) {
if ($tag->hasAttribute('property') && strpos($tag->getAttribute('property'), 'og:') === 0) {
$key = strtr(substr($tag->getAttribute('property'), 3), '-', '_');
$page->_values[$key] = $tag->getAttribute('content');

if( array_key_exists($key, $page->_values) ){
if ( !array_key_exists($key.'_additional', $page->_values) ){
$page->_values[$key.'_additional'] = array();
}
$page->_values[$key.'_additional'][] = $tag->getAttribute('content');
}else{
$page->_values[$key] = $tag->getAttribute('content');
}
}
//Added this if loop to retrieve description values from sites like the New York Times who have malformed it.
if ($tag ->hasAttribute('value') && $tag->hasAttribute('property') &&

//Added this if loop to retrieve description values from sites like the New York Times who have malformed it.
if ($tag->hasAttribute('value') && $tag->hasAttribute('property') &&
strpos($tag->getAttribute('property'), 'og:') === 0) {
$key = strtr(substr($tag->getAttribute('property'), 3), '-', '_');
$page->_values[$key] = $tag->getAttribute('value');
Expand All @@ -112,8 +134,46 @@ static private function _parse($HTML) {
if ($tag->hasAttribute('name') && $tag->getAttribute('name') === 'description') {
$nonOgDescription = $tag->getAttribute('content');
}


if ($tag->hasAttribute('property') &&
strpos($tag->getAttribute('property'), 'twitter:') === 0) {
$key = strtr($tag->getAttribute('property'), '-:', '__');
$page->_values[$key] = $tag->getAttribute('content');
}

if ($tag->hasAttribute('name') &&
strpos($tag->getAttribute('name'), 'twitter:') === 0) {
$key = strtr($tag->getAttribute('name'), '-:', '__');
if( array_key_exists($key, $page->_values) ){
if (!array_key_exists($key.'_additional', $page->_values)){
$page->_values[$key.'_additional'] = array();
}
$page->_values[$key.'_additional'][] = $tag->getAttribute('content');
} else {
$page->_values[$key] = $tag->getAttribute('content');
}
}

// Notably this will not work if you declare type after you declare type values on a page.
if ( array_key_exists('type', $page->_values) ){
$meta_key = $page->_values['type'].':';
if ($tag->hasAttribute('property') && strpos($tag->getAttribute('property'), $meta_key) === 0) {
$meta_key_len = strlen($meta_key);
$key = strtr(substr($tag->getAttribute('property'), $meta_key_len), '-', '_');
$key = $page->_values['type'].'_'.$key;

if( array_key_exists($key, $page->_values) ){
if ( !array_key_exists($key.'_additional', $page->_values) ){
$page->_values[$key.'_additional'] = array();
}
$page->_values[$key.'_additional'][] = $tag->getAttribute('content');
}else{
$page->_values[$key] = $tag->getAttribute('content');
}
}
}
}

//Based on modifications at https://github.com/bashofmann/opengraph/blob/master/src/OpenGraph/OpenGraph.php
if (!isset($page->_values['title'])) {
$titles = $doc->getElementsByTagName('title');
Expand All @@ -126,7 +186,7 @@ static private function _parse($HTML) {
}

//Fallback to use image_src if ogp::image isn't set.
if (!isset($page->values['image'])) {
if (!isset($page->_values['image'])) {
$domxpath = new DOMXPath($doc);
$elements = $domxpath->query("//link[@rel='image_src']");

Expand All @@ -136,11 +196,21 @@ static private function _parse($HTML) {
$page->_values['image'] = $domattr->value;
$page->_values['image_src'] = $domattr->value;
}
}
} else if (!empty($page->_values['twitter_image'])){
$page->_values['image'] = $page->_values['twitter_image'];
} else {
$elements = $doc->getElementsByTagName("img");
foreach ( $elements as $tag ){
if ($tag->hasAttribute('width') && ( ($tag->getAttribute('width') > 300) || ($tag->getAttribute('width') == '100%') ) ){
$page->_values['image'] = $tag->getAttribute('src');
break;
}
}
}
}

if (empty($page->_values)) { return false; }

return $page;
}

Expand All @@ -155,7 +225,7 @@ public function __get($key) {
if (array_key_exists($key, $this->_values)) {
return $this->_values[$key];
}

if ($key === 'schema') {
foreach (self::$TYPES AS $schema => $types) {
if (array_search($this->_values['type'], $types)) {
Expand Down Expand Up @@ -192,7 +262,7 @@ public function hasLocation() {
if (array_key_exists('latitude', $this->_values) && array_key_exists('longitude', $this->_values)) {
return true;
}

$address_keys = array('street_address', 'locality', 'region', 'postal_code', 'country_name');
$valid_address = true;
foreach ($address_keys AS $key) {
Expand Down