Skip to content
This repository has been archived by the owner on Mar 26, 2024. It is now read-only.

Commit

Permalink
Fixed parser problem with some websites. Changed project name.
Browse files Browse the repository at this point in the history
  • Loading branch information
Tesura authored and Tesura committed Sep 9, 2022
1 parent 2681c29 commit a367ef1
Show file tree
Hide file tree
Showing 2 changed files with 157 additions and 91 deletions.
55 changes: 32 additions & 23 deletions composer.json
View file
Original file line number Diff line number Diff line change
@@ -1,25 +1,34 @@
{
"name": "anzenkodo/small-rss-php",
"description": "RSS & Atom feed parser for PHP. Very small and easy-to-use library for parsing your feeds.",
"keywords": ["rss", "atom", "feed"],
"homepage": "https://github.com/AnzenKodo/small-rss-php",
"license": ["unlicense"],
"authors": [
{
"name": "AnzenKodo",
"email": "AnzenKodo@altmails.com",
"homepage": "https://AnzenKodo.github.io/AnzenKodo",
"role": "Developer"
}
],
"support": {
"issues": "https://github.com/AnzenKodo/small-rss-php/issues",
"source": "https://github.com/AnzenKodo/small-rss-php"
},
"require": {
"php": ">=8.0"
},
"autoload": {
"classmap": ["src/"]
}
"name": "anzenkodo/rss-atom-parser",
"description": "RSS & Atom feed parser for PHP. Very small and easy-to-use library for parsing your feeds.",
"type": "library",
"keywords": [ "rss", "atom", "feed" ],
"homepage": "https://github.com/AnzenKodo/rss-atom-parser",
"readme": "https://github.com/AnzenKodo/rss-atom-parser/blob/main/README.md",
"license": "MIT",
"authors": [
{
"name": "AnzenKodo",
"email": "AnzenKodo@altmails.com",
"homepage": "https://AnzenKodo.github.io/AnzenKodo",
"role": "Developer"
}
],
"support": {
"issues": "https://github.com/AnzenKodo/rss-atom-parser/issues",
"source": "https://github.com/AnzenKodo/rss-atom-parser",
"docs": "https://github.com/AnzenKodo/rss-atom-parser/blob/main/README.md",
},
"funding": [
{
"type": "website",
"url": "https://AnzenKodo.github.io/AnzenKodo#support"
}
],
"require": {
"php": ">=8.0"
},
"autoload": {
"classmap": [ "src/" ]
}
}
193 changes: 125 additions & 68 deletions src/rss.php
View file
Original file line number Diff line number Diff line change
@@ -1,121 +1,176 @@
<?php

class RSS {
public static function feed(string|array|object $url): array {
public static $useragent = "FeedFetcher-Google";

public static function feed(string|array|object $url): object {
$data = array();

// If there are more then 1 $url in object and array type.
if (is_object($url) or is_array($url))
$data = self::multi_feed_data($url);
$data = self::getFeeds($url);
// If only 1 $url in string type.
else if (is_string($url))
$data = self::feed_data($url);
else {
else if (is_string($url)) {
$data = self::getFeed($url);
} else {
throw new ErrorException("Can't find datatype of {$url}");
}

return $data;
}

// For multiple feeds in array and object type only.
public static function multi_feed_data(array|object $urls): array {
public static function getFeeds(array|object $urls): object {
$data = array();

foreach ($urls as $url) {
$data[] = self::feed($url);
}
// Sort feed by time
usort($data, function($t1, $t2) {
$time1 = $t1['date']." ".$t1['time'];
$time2 = $t2['date']." ".$t2['time'];

return strtotime($time1) + strtotime($time2);
});

return $data;
// Sort feed by time
$ord = array();
foreach ($data as $key){
$ord[] = strtotime($key->item[0]->date);
}
array_multisort($ord, SORT_DESC, $data);

return (object)$data;
}

// For single feed in string type.
public static function feed_data(string $url): array {
public static function getFeed(string $url): object {
ini_set('user_agent', self::$useragent);

// Checks if $url content type is html. If it is HTML then finds the Feed
// url and changes $url to feed url.
$its_html = self::checkContentType($url, "html");
if($its_html) {
$content_html = self::getContent($url);
$url = self::getFeedUrl($content_html);
}

$its_html = self::check_content_type($url, "html");
if($its_html) { $url = self::get_feed_url($content); }
$content = self::getContent($url);

$content = self::get_content($url);
$xml = new SimpleXmlElement($content, true);
$data = (object)array();
if ($xml->channel) { // If feed is xml vesrion 1.0
$data = self::getRSS($xml, $url);
} else if($xml->entry) { // If feed is xml version 2.0
$data = self::getAtom($xml, $url);
} else {
}

$xml = self::get_feed($content);
return (object)$data;
}

return self::format_data($xml);
private static function isHttp(string $str, string $url, bool $boolean = false) {
if (preg_match("/^(https|http):\/\//", $str)) return $str;

if ($str[0] == "/") return preg_replace('/(?<=(\w|\d))\/.*$/', $str, $url);

if ($boolean) return false;

return $url;
}

private static function getAtom(SimpleXMLElement $xml, string $url): array {
$title = $xml->id ? $xml->id : $url;
$title = $xml->title != "" ? $xml->title : $title;

// Find proper name
$replace = "";
if (isset($xml->link)) {
for ($x = 0; $x < 2; $x++) {
if ($xml->link[$x]["rel"] == "alternate") {
$replace = $xml->link[$x]["href"];
break;
} else {
$replace = $url;
}
}
} else if (isset($xml->author->uri)) {
$replace = $xml->author->uri;
} else {
$replace = $url;
}

// Simplifies the feed data.
private static function format_data(object $xml): array {
// Formats time and returns $type of time format.
$time_format = function(object|string $time, string $format = "d-M-Y H:i"):
string {
$data = [
"title" => $title,
"description" => "$xml->subtitle",
"date" => self::timeFormat($xml->updated, "d-M-Y"),
"time" => self::timeFormat($xml->updated, "H:i"),
"link" => self::isHttp($replace, $url, true) ? self::isHttp($replace, $url) : self::isHttp($xml->id, $url),
"feed" => "$url",
"item" => array()
];

// Convert to time format
$timestamp = strtotime("$time");
foreach ($xml->entry as $item) {
$time = $item->updated ? $item->updated : $item->published;

$time = date($format, $timestamp);
$data_item = [
"title" => "$item->title",
"link" => self::isHttp($item->link["href"], $url),
"date" => self::timeFormat($time, "d-M-Y"),
"time" => self::timeFormat($time, "H:i")
];

return $time;
};
array_push($data["item"], (object)$data_item);
}

return $data;
}

private static function getRSS(SimpleXMLElement $xml, string $url): array {
$xml = $xml->channel;
$time = $xml->lastBuildDate ? $xml->lastBuildDate : $xml->pubDate;

// Formats the xml data
$data = [
"title" => "$xml->title",
"title" => $xml->title != "" ? $xml->title : $xml->link,
"description" => "$xml->description",
"date" => $time_format($xml->lastBuildDate, "d-M-Y"),
"time" => $time_format($xml->lastBuildDate, "H:i"),
"link" => "$xml->link",
"date" => self::timeFormat($time, "d-M-Y"),
"time" => self::timeFormat($time, "H:i"),
"link" => $xml->link != "" ? self::isHttp($xml->link, $url) : $url,
"feed" => $url,
"item" => array()
];

// Formats the xml items data
foreach ($xml->item as $item) {
$data_item = [
"title" => "$item->title",
"link" => "$item->link",
"date" => $time_format($item->pubDate, "d-M-Y"),
"time" => $time_format($item->pubDate, "H:i")
"title" => $item->title != "" ? $item->title : $item->link,
"link" => $item->link ? self::isHttp($item->link, $url) : $item->enclosure["url"],
"date" => self::timeFormat($item->pubDate, "d-M-Y"),
"time" => self::timeFormat($item->pubDate, "H:i")
];

array_push($data["item"], $data_item);
array_push($data["item"], (object)$data_item);
}

return $data;
}

// Checks feed xml version and returns content as per feed
private static function get_feed(string $content): object {
// Converts $content to object
$xml_data = new SimpleXmlElement($content, LIBXML_NOCDATA);
private static function timeFormat(object|string $time, string $format =
"d-M-Y H:i"): string {

$xml = (object)array(); // Convert array to object
// Convert to time format
$timestamp = strtotime("$time");

if ($xml_data->channel) // If feed is xml vesrion 1.0
$xml = $xml_data->channel;
else if($xml_data->entry) // If feed is xml version 2.0
$xml = $xml_data;
else {
/* self::console_log("$url: Invalid content type"); */
}
$time = date($format, $timestamp);

return $xml;
return $time;
}

// Returns given $url page content
private static function get_content(string $url, $useragent = "FeedFetcher-Google"): string {
private static function getContent(string $url): string {
$curl = curl_init($url);

// Return the transfer as a string.
curl_setopt($curl, CURLOPT_RETURNTRANSFER, true);
// The maximum number of seconds to allow cURL functions to execute.
curl_setopt($curl, CURLOPT_TIMEOUT, 60);
// Changes curl useragent
curl_setopt($curl, CURLOPT_USERAGENT, $useragent);
curl_setopt($curl, CURLOPT_USERAGENT, self::$useragent);
// To make cURL follow a redirect
curl_setopt($curl, CURLOPT_FOLLOWLOCATION, true);

// Given $url content
$content = curl_exec($curl);
Expand All @@ -126,7 +181,7 @@ private static function get_content(string $url, $useragent = "FeedFetcher-Googl
}

// Finds Feed url from HTML.
private static function get_feed_url(string $content): string {
private static function getFeedUrl(string $content): string {
// Suppresses DOMDocuments errors
libxml_use_internal_errors(true);

Expand All @@ -153,24 +208,26 @@ private static function get_feed_url(string $content): string {

// Checks given $content_type_name value matches given $url header content
// type.
private static function check_content_type(string $url, string
private static function checkContentType(string $url, string
$type_name): bool {
// Get $url header
$header = get_headers($url, true);

$content_type = "Content-Type";
// Contet type in lower case
$content_type_lower = strtolower($content_type);
$content_type = "";

// Find $type_name in header
$content_pos = strpos($header[$content_type], $type_name);
$content_pos_lower = strpos($header[$content_type_lower], $type_name);

if ($content_pos)
if (isset($header["Content-Type"]))
$content_type = $header["Content-Type"];
else
$content_type = $header["content-type"];

if (gettype($content_type) == "array")
$content_type = $content_type[0];

if (strpos($content_type, $type_name))
$checked_header = true;
else if ($content_pos_lower)
else {
$checked_header = false;
else {}
}

return $checked_header;
}
Expand Down

0 comments on commit a367ef1

Please sign in to comment.