X-Git-Url: https://git.ao2.it/tweeper.git/blobdiff_plain/50842e9d4c8ba313313710d0b4472548390cd77b..cc55224c919a727b16897879d7c35f54ea11aaf8:/tweeper.php?ds=sidebyside
diff --git a/tweeper.php b/tweeper.php
index 17fff8b..293595f 100644
--- a/tweeper.php
+++ b/tweeper.php
@@ -18,6 +18,12 @@
* along with this program. If not, see .
*/
+require_once 'Symfony/Component/Serializer/autoload.php';
+
+use Symfony\Component\Serializer\Serializer;
+use Symfony\Component\Serializer\Encoder\XmlEncoder;
+use Symfony\Component\Serializer\Normalizer\ObjectNormalizer;
+
date_default_timezone_set('UTC');
class Tweeper {
@@ -30,12 +36,20 @@ class Tweeper {
public static function epoch_to_gmdate($timestamp)
{
+ if (!is_numeric($timestamp) || is_nan($timestamp)) {
+ $timestamp = 0;
+ }
+
return gmdate('D, d M Y H:i:s', $timestamp) . ' GMT';
}
public static function str_to_gmdate($date)
{
$timestamp = strtotime($date);
+ if (FALSE === $timestamp) {
+ $timestamp = 0;
+ }
+
return Tweeper::epoch_to_gmdate($timestamp);
}
@@ -96,7 +110,11 @@ class Tweeper {
"video/ogg",
);
- $url_info = Tweeper::get_info($url);
+ // The RSS specification says that the enclosure element url must be http.
+ // See http://sourceforge.net/p/feedvalidator/bugs/72/
+ $http_url = preg_replace("/^https/", "http", $url);
+
+ $url_info = Tweeper::get_info($http_url);
$supported = in_array($url_info['content_type'], $supported_content_types);
if (!$supported) {
@@ -147,7 +165,7 @@ class Tweeper {
private function load_stylesheet($host) {
$stylesheet = "file://" . __DIR__ . "/rss_converter_" . $host . ".xsl";
if (FALSE === file_exists($stylesheet)) {
- trigger_error("Conversion to RSS not supported: $host", E_USER_ERROR);
+ trigger_error("Conversion to RSS not supported for $host ($stylesheet not found)", E_USER_ERROR);
return NULL;
}
@@ -164,31 +182,106 @@ class Tweeper {
return $xsltProcessor;
}
- public function tweep($src_url) {
- $url = parse_url($src_url);
- if (FALSE === $url || empty($url["host"])) {
- trigger_error("Invalid url: $src_url", E_USER_ERROR);
+ private function json_to_xml($json, $root_node_name) {
+ // Apparenty the ObjectNormalizer used afterwards is not able to handle
+ // the stdClass object created by json_decode() with the default setting
+ // $assoc = false; so use $assoc = true
+ $data = json_decode($json, $assoc = true);
+ if (!$data) {
return NULL;
}
- $xsltProcessor = $this->load_stylesheet($url["host"]);
- if (NULL === $xsltProcessor) {
+ $encoder = new XmlEncoder();
+ $normalizer = new ObjectNormalizer();
+ $serializer = new Serializer(array($normalizer), array($encoder));
+
+ $serializer_options = array (
+ 'xml_encoding' => "UTF-8",
+ 'xml_format_output' => TRUE,
+ 'xml_root_node_name' => $root_node_name,
+ );
+
+ $xml_data = $serializer->serialize($data, 'xml', $serializer_options);
+ if (!$xml_data) {
+ trigger_error("Cannot serialize data", E_USER_ERROR);
return NULL;
}
- $html = Tweeper::get_contents($src_url);
+ return $xml_data;
+ }
+ private function get_xml_instagram_com($html) {
+ // extract the json data from the html code
+ $json_match_expr = '/window._sharedData = (.*);/';
+ $ret = preg_match($json_match_expr, $html, $matches);
+ if ($ret !== 1) {
+ trigger_error("Cannot match expression: $json_match_expr\n", E_USER_ERROR);
+ return NULL;
+ }
+
+ return $this->json_to_xml($matches[1], 'instagram');
+ }
+
+ private function preprocess_html_facebook_com($html) {
+ $html = str_replace('', '', $html);
+ return $html;
+ }
+
+ private function html_to_xml($html, $host) {
$xmlDoc = new DOMDocument();
// Handle warnings and errors when loading invalid HTML.
$xml_errors_value = libxml_use_internal_errors(true);
- $xmlDoc->loadHTML($html);
+
+ // If there is a host-specific method to get the xml data, use it!
+ $get_xml_host_method = 'get_xml_' . str_replace(".", "_", $host);
+ if (method_exists($this, $get_xml_host_method)) {
+ $xml_data = call_user_func_array(array($this, $get_xml_host_method), array($html));
+ $xmlDoc->loadXML($xml_data);
+ } else {
+ $xmlDoc->loadHTML($html);
+ }
+
foreach (libxml_get_errors() as $xml_error) {
$this->log_xml_error($xml_error);
}
libxml_clear_errors();
libxml_use_internal_errors($xml_errors_value);
+ return $xmlDoc;
+ }
+
+ public function tweep($src_url) {
+ $url = parse_url($src_url);
+ if (FALSE === $url || empty($url["host"])) {
+ trigger_error("Invalid url: $src_url", E_USER_ERROR);
+ return NULL;
+ }
+
+ // Strip the leading www. to be more forgiving on input URLs
+ $host = preg_replace('/^www\./', '', $url["host"]);
+
+ $xsltProcessor = $this->load_stylesheet($host);
+ if (NULL === $xsltProcessor) {
+ return NULL;
+ }
+
+ $html = $this->get_contents($src_url);
+ if (FALSE === $html) {
+ return NULL;
+ }
+
+ $preprocess_html_host_method = 'preprocess_html_' . str_replace(".", "_", $host);
+ if (method_exists($this, $preprocess_html_host_method)) {
+ $html = call_user_func_array(array($this, $preprocess_html_host_method), array($html));
+ }
+
+ $xmlDoc = $this->html_to_xml($html, $host);
+ if (NULL === $xmlDoc) {
+ return NULL;
+ }
+
$output = $xsltProcessor->transformToXML($xmlDoc);
if (FALSE === $output) {
@@ -199,12 +292,18 @@ class Tweeper {
}
}
+function is_cli()
+{
+ return (php_sapi_name() === "cli");
+}
+
function usage($argv)
{
- if (php_sapi_name() != 'cli')
- $usage = htmlentities("{$_SERVER['SCRIPT_NAME']}?src_url=&generate_enclosure=<0|1>");
- else
+ if (is_cli()) {
$usage = "{$argv[0]} [-e|-h|--help] \n";
+ } else {
+ $usage = htmlentities("{$_SERVER['SCRIPT_NAME']}?src_url=&generate_enclosure=<0|1>");
+ }
return "usage: $usage";
}
@@ -255,16 +354,16 @@ function parse_options_query_string()
}
-if (php_sapi_name() != 'cli') {
- $options = parse_options_query_string();
- $ERROR_STREAM = fopen('php://output', 'w');
-} else {
+if (is_cli()) {
$options = parse_options_cli($argv, $argc);
$ERROR_STREAM = fopen('php://stderr', 'w');
+} else {
+ $options = parse_options_query_string();
+ $ERROR_STREAM = fopen('php://output', 'w');
}
if (!isset($options['src_url'])) {
- fwrite($ERROR_STREAM, usage($argv));
+ fwrite($ERROR_STREAM, usage(is_cli() ? $argv : NULL));
exit(1);
}