Implement Web scraping "HTML + XPath" (#4220)

* More PHP type hints for Fever
Follow-up of https://github.com/FreshRSS/FreshRSS/pull/4201
Related to https://github.com/FreshRSS/FreshRSS/issues/4200

* Detail

* Draft

* Progress

* More draft

* Fix thumbnail PHP type hint
https://github.com/FreshRSS/FreshRSS/issues/4215

* More types

* A bit more

* Refactor FreshRSS_Entry::fromArray

* Progress

* Starts to work

* Categories

* Fonctional

* Layout update

* Fix relative URLs

* Cache system

* Forgotten files

* Remove a debug line

* Automatic form validation of XPath expressions

* data-leave-validation

* Fix reload action

* Simpler examples

* Fix column type for PostgreSQL

* Enforce HTTP encoding

* Readme

* Fix get full content

* target="_blank"

* gitignore

* htmlspecialchars_utf8

* Implement HTML <base>
And fix/revert `xml:base` support in SimplePie https://github.com/simplepie/simplepie/commit/e49c578817aa504d8d05cd7f33857aeda9d41908

* SimplePie upstream PR merged
https://github.com/simplepie/simplepie/pull/723

8 files changed, 270 insertions, 114 deletions

diff --git a/app/Models/Entry.php b/app/Models/Entry.php
index a190e505d..ab88d777a 100644
--- a/app/Models/Entry.php
+++ b/app/Models/Entry.php
@@ -59,6 +59,38 @@ class FreshRSS_Entry extends Minz_Model {
 		$this->_guid($guid);
 	}
 
+	/** @param array<string,mixed> $dao */
+	public static function fromArray(array $dao): FreshRSS_Entry {
+		if (!isset($dao['content'])) {
+			$dao['content'] = '';
+		}
+		if (isset($dao['thumbnail'])) {
+			$dao['content'] .= '<p class="enclosure-content"><img src="' . $dao['thumbnail'] . '" alt="" /></p>';
+		}
+		$entry = new FreshRSS_Entry(
+			$dao['id_feed'] ?? 0,
+			$dao['guid'] ?? '',
+			$dao['title'] ?? '',
+			$dao['author'] ?? '',
+			$dao['content'] ?? '',
+			$dao['link'] ?? '',
+			$dao['date'] ?? 0,
+			$dao['is_read'] ?? false,
+			$dao['is_favorite'] ?? false,
+			$dao['tags'] ?? ''
+		);
+		if (isset($dao['id'])) {
+			$entry->_id($dao['id']);
+		}
+		if (!empty($dao['timestamp'])) {
+			$entry->_date(strtotime($dao['timestamp']));
+		}
+		if (!empty($dao['categories'])) {
+			$entry->_tags($dao['categories']);
+		}
+		return $entry;
+	}
+
 	public function id(): string {
 		return $this->id;
 	}
@@ -83,6 +115,7 @@ class FreshRSS_Entry extends Minz_Model {
 		return $this->content;
 	}
 
+	/** @return array<array<string,string>> */
 	public function enclosures(bool $searchBodyImages = false): array {
 		$results = [];
 		try {
@@ -97,11 +130,20 @@ class FreshRSS_Entry extends Minz_Model {
 			if ($searchEnclosures) {
 				$enclosures = $xpath->query('//div[@class="enclosure"]/p[@class="enclosure-content"]/*[@src]');
 				foreach ($enclosures as $enclosure) {
-					$results[] = [
+					$result = [
 						'url' => $enclosure->getAttribute('src'),
 						'type' => $enclosure->getAttribute('data-type'),
+						'medium' => $enclosure->getAttribute('data-medium'),
 						'length' => $enclosure->getAttribute('data-length'),
 					];
+					if (empty($result['medium'])) {
+						switch (strtolower($enclosure->nodeName)) {
+							case 'img': $result['medium'] = 'image'; break;
+							case 'video': $result['medium'] = 'video'; break;
+							case 'audio': $result['medium'] = 'audio'; break;
+						}
+					}
+					$results[] = $result;
 				}
 			}
 			if ($searchBodyImages) {
@@ -432,52 +474,12 @@ class FreshRSS_Entry extends Minz_Model {
 		}
 	}
 
-	public static function getContentByParsing(string $url, string $path, array $attributes = array(), int $maxRedirs = 3): string {
-		$limits = FreshRSS_Context::$system_conf->limits;
-		$feed_timeout = empty($attributes['timeout']) ? 0 : intval($attributes['timeout']);
-
-		if (FreshRSS_Context::$system_conf->simplepie_syslog_enabled) {
-			syslog(LOG_INFO, 'FreshRSS GET ' . SimplePie_Misc::url_remove_credentials($url));
-		}
-
-		$ch = curl_init();
-		curl_setopt_array($ch, [
-			CURLOPT_URL => $url,
-			CURLOPT_REFERER => SimplePie_Misc::url_remove_credentials($url),
-			CURLOPT_HTTPHEADER => array('Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8'),
-			CURLOPT_USERAGENT => FRESHRSS_USERAGENT,
-			CURLOPT_CONNECTTIMEOUT => $feed_timeout > 0 ? $feed_timeout : $limits['timeout'],
-			CURLOPT_TIMEOUT => $feed_timeout > 0 ? $feed_timeout : $limits['timeout'],
-			//CURLOPT_FAILONERROR => true;
-			CURLOPT_MAXREDIRS => 4,
-			CURLOPT_RETURNTRANSFER => true,
-			CURLOPT_FOLLOWLOCATION => true,
-			CURLOPT_ENCODING => '',	//Enable all encodings
-		]);
-
-		curl_setopt_array($ch, FreshRSS_Context::$system_conf->curl_options);
-
-		if (isset($attributes['curl_params']) && is_array($attributes['curl_params'])) {
-			curl_setopt_array($ch, $attributes['curl_params']);
-		}
-
-		if (isset($attributes['ssl_verify'])) {
-			curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, $attributes['ssl_verify'] ? 2 : 0);
-			curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, $attributes['ssl_verify'] ? true : false);
-			if (!$attributes['ssl_verify']) {
-				curl_setopt($ch, CURLOPT_SSL_CIPHER_LIST, 'DEFAULT@SECLEVEL=1');
-			}
-		}
-		$html = curl_exec($ch);
-		$c_status = curl_getinfo($ch, CURLINFO_HTTP_CODE);
-		$c_error = curl_error($ch);
-		curl_close($ch);
-
-		if ($c_status != 200 || $c_error != '') {
-			Minz_Log::warning('Error fetching content: HTTP code ' . $c_status . ': ' . $c_error . ' ' . $url);
-		}
-
-		if (is_string($html) && strlen($html) > 0) {
+	/**
+	 * @param array<string,mixed> $attributes
+	 */
+	public static function getContentByParsing(string $url, string $path, array $attributes = [], int $maxRedirs = 3): string {
+		$html = getHtml($url, $attributes);
+		if (strlen($html) > 0) {
 			require_once(LIB_PATH . '/lib_phpQuery.php');
 			/**
 			 * @var phpQueryObject @doc
diff --git a/app/Models/EntryDAO.php b/app/Models/EntryDAO.php
index a10440edb..8f248e20f 100644
--- a/app/Models/EntryDAO.php
+++ b/app/Models/EntryDAO.php
@@ -164,7 +164,7 @@ INSERT IGNORE INTO `_entry` (
 )
 SELECT @rank:=@rank+1 AS id, guid, title, author, content_bin, link, date, `lastSeen`, hash, is_read, is_favorite, id_feed, tags
 FROM `_entrytmp`
-ORDER BY date;
+ORDER BY date, id;
 
 DELETE FROM `_entrytmp` WHERE id <= @rank;
 SQL;
@@ -658,6 +658,7 @@ SQL;
 		}
 	}
 
+	/** @return FreshRSS_Entry|null */
 	public function searchByGuid($id_feed, $guid) {
 		// un guid est unique pour un flux donné
 		$sql = 'SELECT id, guid, title, author, '
@@ -669,9 +670,10 @@ SQL;
 		$stm->bindParam(':guid', $guid);
 		$stm->execute();
 		$res = $stm->fetchAll(PDO::FETCH_ASSOC);
-		return isset($res[0]) ? self::daoToEntry($res[0]) : null;
+		return isset($res[0]) ? FreshRSS_Entry::fromArray($res[0]) : null;
 	}
 
+	/** @return FreshRSS_Entry|null */
 	public function searchById($id) {
 		$sql = 'SELECT id, guid, title, author, '
 			. ($this->isCompressed() ? 'UNCOMPRESS(content_bin) AS content' : 'content')
@@ -681,7 +683,7 @@ SQL;
 		$stm->bindParam(':id', $id, PDO::PARAM_INT);
 		$stm->execute();
 		$res = $stm->fetchAll(PDO::FETCH_ASSOC);
-		return isset($res[0]) ? self::daoToEntry($res[0]) : null;
+		return isset($res[0]) ? FreshRSS_Entry::fromArray($res[0]) : null;
 	}
 
 	public function searchIdByGuid($id_feed, $guid) {
@@ -1061,7 +1063,7 @@ SQL;
 		$stm = $this->listWhereRaw($type, $id, $state, $order, $limit, $firstId, $filters, $date_min);
 		if ($stm) {
 			while ($row = $stm->fetch(PDO::FETCH_ASSOC)) {
-				yield self::daoToEntry($row);
+				yield FreshRSS_Entry::fromArray($row);
 			}
 		} else {
 			yield false;
@@ -1092,7 +1094,7 @@ SQL;
 		$stm = $this->pdo->prepare($sql);
 		$stm->execute($ids);
 		while ($row = $stm->fetch(PDO::FETCH_ASSOC)) {
-			yield self::daoToEntry($row);
+			yield FreshRSS_Entry::fromArray($row);
 		}
 	}
 
@@ -1251,23 +1253,4 @@ SQL;
 		$unread = empty($res[1]) ? 0 : intval($res[1]);
 		return array('all' => $all, 'unread' => $unread, 'read' => $all - $unread);
 	}
-
-	public static function daoToEntry($dao) {
-		$entry = new FreshRSS_Entry(
-				$dao['id_feed'],
-				$dao['guid'],
-				$dao['title'],
-				$dao['author'],
-				$dao['content'],
-				$dao['link'],
-				$dao['date'],
-				$dao['is_read'],
-				$dao['is_favorite'],
-				isset($dao['tags']) ? $dao['tags'] : ''
-			);
-		if (isset($dao['id'])) {
-			$entry->_id($dao['id']);
-		}
-		return $entry;
-	}
 }
diff --git a/app/Models/EntryDAOPGSQL.php b/app/Models/EntryDAOPGSQL.php
index 7a46670fc..b97417a7c 100644
--- a/app/Models/EntryDAOPGSQL.php
+++ b/app/Models/EntryDAOPGSQL.php
@@ -45,13 +45,13 @@ rank bigint := (SELECT maxrank - COUNT(*) FROM `_entrytmp`);
 BEGIN
 	INSERT INTO `_entry`
 		(id, guid, title, author, content, link, date, `lastSeen`, hash, is_read, is_favorite, id_feed, tags)
-		(SELECT rank + row_number() OVER(ORDER BY date) AS id, guid, title, author, content,
+		(SELECT rank + row_number() OVER(ORDER BY date, id) AS id, guid, title, author, content,
 			link, date, `lastSeen`, hash, is_read, is_favorite, id_feed, tags
 			FROM `_entrytmp` AS etmp
 			WHERE NOT EXISTS (
 				SELECT 1 FROM `_entry` AS ereal
 				WHERE (etmp.id = ereal.id) OR (etmp.id_feed = ereal.id_feed AND etmp.guid = ereal.guid))
-			ORDER BY date);
+			ORDER BY date, id);
 	DELETE FROM `_entrytmp` WHERE id <= maxrank;
 END $$;';
 		$hadTransaction = $this->pdo->inTransaction();
diff --git a/app/Models/EntryDAOSQLite.php b/app/Models/EntryDAOSQLite.php
index 8b0f2d252..16d15f899 100644
--- a/app/Models/EntryDAOSQLite.php
+++ b/app/Models/EntryDAOSQLite.php
@@ -41,13 +41,13 @@ DROP TABLE IF EXISTS `tmp`;
 CREATE TEMP TABLE `tmp` AS
 	SELECT id, guid, title, author, content, link, date, `lastSeen`, hash, is_read, is_favorite, id_feed, tags
 	FROM `_entrytmp`
-	ORDER BY date;
+	ORDER BY date, id;
 INSERT OR IGNORE INTO `_entry`
 	(id, guid, title, author, content, link, date, `lastSeen`, hash, is_read, is_favorite, id_feed, tags)
 	SELECT rowid + (SELECT MAX(id) - COUNT(*) FROM `tmp`) AS id,
 	guid, title, author, content, link, date, `lastSeen`, hash, is_read, is_favorite, id_feed, tags
 	FROM `tmp`
-	ORDER BY date;
+	ORDER BY date, id;
 DELETE FROM `_entrytmp` WHERE id <= (SELECT MAX(id) FROM `tmp`);
 DROP TABLE IF EXISTS `tmp`;
 ';
diff --git a/app/Models/Feed.php b/app/Models/Feed.php
index 3425f4bce..0e02194ef 100644
--- a/app/Models/Feed.php
+++ b/app/Models/Feed.php
@@ -1,6 +1,28 @@
 <?php
 
 class FreshRSS_Feed extends Minz_Model {
+
+	/**
+	 * Normal RSS or Atom feed
+	 * @var int
+	 */
+	const KIND_RSS = 0;
+	/**
+	 * Invalid RSS or Atom feed
+	 * @var int
+	 */
+	const KIND_RSS_FORCED = 2;
+	/**
+	 * Normal HTML with XPath scraping
+	 * @var int
+	 */
+	const KIND_HTML_XPATH = 10;
+	/**
+	 * Normal JSON with XPath scraping
+	 * @var int
+	 */
+	const KIND_JSON_XPATH = 20;
+
 	const PRIORITY_MAIN_STREAM = 10;
 	const PRIORITY_NORMAL = 0;
 	const PRIORITY_ARCHIVED = -10;
@@ -10,33 +32,50 @@ class FreshRSS_Feed extends Minz_Model {
 	const ARCHIVING_RETENTION_COUNT_LIMIT = 10000;
 	const ARCHIVING_RETENTION_PERIOD = 'P3M';
 
-	/**
-	 * @var int
-	 */
+	/** @var int */
 	private $id = 0;
-	private $url;
-	/**
-	 * @var int
-	 */
+	/** @var string */
+	private $url = '';
+	/** @var int */
+	private $kind = 0;
+	/** @var int */
 	private $category = 1;
+	/** @var int */
 	private $nbEntries = -1;
+	/** @var int */
 	private $nbNotRead = -1;
+	/** @var int */
 	private $nbPendingNotRead = 0;
+	/** @var string */
 	private $name = '';
+	/** @var string */
 	private $website = '';
+	/** @var string */
 	private $description = '';
+	/** @var int */
 	private $lastUpdate = 0;
+	/** @var int */
 	private $priority = self::PRIORITY_MAIN_STREAM;
+	/** @var string */
 	private $pathEntries = '';
+	/** @var string */
 	private $httpAuth = '';
+	/** @var bool */
 	private $error = false;
+	/** @var int */
 	private $ttl = self::TTL_DEFAULT;
 	private $attributes = [];
+	/** @var bool */
 	private $mute = false;
+	/** @var string */
 	private $hash = '';
+	/** @var string */
 	private $lockPath = '';
+	/** @var string */
 	private $hubUrl = '';
+	/** @var string */
 	private $selfUrl = '';
+	/** @var array<FreshRSS_FilterAction> $filterActions */
 	private $filterActions = null;
 
 	public function __construct(string $url, bool $validate = true) {
@@ -47,6 +86,9 @@ class FreshRSS_Feed extends Minz_Model {
 		}
 	}
 
+	/**
+	 * @return FreshRSS_Feed
+	 */
 	public static function example() {
 		$f = new FreshRSS_Feed('http://example.net/', false);
 		$f->faviconPrepare();
@@ -71,6 +113,9 @@ class FreshRSS_Feed extends Minz_Model {
 	public function selfUrl(): string {
 		return $this->selfUrl;
 	}
+	public function kind(): int {
+		return $this->kind;
+	}
 	public function hubUrl(): string {
 		return $this->hubUrl;
 	}
@@ -200,6 +245,9 @@ class FreshRSS_Feed extends Minz_Model {
 		}
 		$this->url = $value;
 	}
+	public function _kind($value) {
+		$this->kind = $value;
+	}
 	public function _category($value) {
 		$value = intval($value);
 		$this->category = $value >= 0 ? $value : 0;
@@ -267,7 +315,7 @@ class FreshRSS_Feed extends Minz_Model {
 	 * @return SimplePie|null
 	 */
 	public function load(bool $loadDetails = false, bool $noCache = false) {
-		if ($this->url !== null) {
+		if ($this->url != '') {
 			// @phpstan-ignore-next-line
 			if (CACHE_PATH === false) {
 				throw new Minz_FileNotExistException(
@@ -347,6 +395,7 @@ class FreshRSS_Feed extends Minz_Model {
 		$guids = [];
 		$hasBadGuids = $this->attributes('hasBadGuids');
 
+		// TODO: Replace very slow $simplePie->get_item($i) by getting all items at once
 		for ($i = $simplePie->get_item_quantity() - 1; $i >= 0; $i--) {
 			$item = $simplePie->get_item($i);
 			if ($item == null) {
@@ -375,6 +424,7 @@ class FreshRSS_Feed extends Minz_Model {
 		$hasBadGuids = $this->attributes('hasBadGuids');
 
 		// We want chronological order and SimplePie uses reverse order.
+		// TODO: Replace very slow $simplePie->get_item($i) by getting all items at once
 		for ($i = $simplePie->get_item_quantity() - 1; $i >= 0; $i--) {
 			$item = $simplePie->get_item($i);
 			if ($item == null) {
@@ -428,15 +478,18 @@ class FreshRSS_Feed extends Minz_Model {
 						} elseif ($medium === 'audio' || strpos($mime, 'audio') === 0) {
 							$enclosureContent .= '<p class="enclosure-content"><audio preload="none" src="' . $elink
 								. ($length == null ? '' : '" data-length="' . intval($length))
-								. '" data-type="' . htmlspecialchars($mime, ENT_COMPAT, 'UTF-8')
+								. ($mime == '' ? '' : '" data-type="' . htmlspecialchars($mime, ENT_COMPAT, 'UTF-8'))
 								. '" controls="controls"></audio> <a download="" href="' . $elink . '">💾</a></p>';
 						} elseif ($medium === 'video' || strpos($mime, 'video') === 0) {
 							$enclosureContent .= '<p class="enclosure-content"><video preload="none" src="' . $elink
 								. ($length == null ? '' : '" data-length="' . intval($length))
-								. '" data-type="' . htmlspecialchars($mime, ENT_COMPAT, 'UTF-8')
+								. ($mime == '' ? '' : '" data-type="' . htmlspecialchars($mime, ENT_COMPAT, 'UTF-8'))
 								. '" controls="controls"></video> <a download="" href="' . $elink . '">💾</a></p>';
 						} else {	//e.g. application, text, unknown
-							$enclosureContent .= '<p class="enclosure-content"><a download="" href="' . $elink . '">💾</a></p>';
+							$enclosureContent .= '<p class="enclosure-content"><a download="" href="' . $elink
+								. ($mime == '' ? '' : '" data-type="' . htmlspecialchars($mime, ENT_COMPAT, 'UTF-8'))
+								. ($medium == '' ? '' : '" data-medium="' . htmlspecialchars($medium, ENT_COMPAT, 'UTF-8'))
+								. '">💾</a></p>';
 						}
 
 						$thumbnailContent = '';
@@ -490,6 +543,97 @@ class FreshRSS_Feed extends Minz_Model {
 	}
 
 	/**
+	 * @param array<string,mixed> $attributes
+	 * @return SimplePie|null
+	 */
+	public function loadHtmlXpath(bool $loadDetails = false, bool $noCache = false, array $attributes = []) {
+		if ($this->url == '') {
+			return null;
+		}
+		$feedSourceUrl = htmlspecialchars_decode($this->url, ENT_QUOTES);
+		if ($this->httpAuth != '') {
+			$feedSourceUrl = preg_replace('#((.+)://)(.+)#', '${1}' . $this->httpAuth . '@${3}', $feedSourceUrl);
+		}
+
+		// Same naming conventions than https://github.com/RSS-Bridge/rss-bridge/wiki/XPathAbstract
+		// https://github.com/RSS-Bridge/rss-bridge/wiki/The-collectData-function
+		/** @var array<string,string> */
+		$xPathSettings = $this->attributes('xpath');
+		$xPathFeedTitle = $xPathSettings['feedTitle'] ?? '';
+		$xPathItem = $xPathSettings['item'] ?? '';
+		$xPathItemTitle = $xPathSettings['itemTitle'] ?? '';
+		$xPathItemContent = $xPathSettings['itemContent'] ?? '';
+		$xPathItemUri = $xPathSettings['itemUri'] ?? '';
+		$xPathItemAuthor = $xPathSettings['itemAuthor'] ?? '';
+		$xPathItemTimestamp = $xPathSettings['itemTimestamp'] ?? '';
+		$xPathItemThumbnail = $xPathSettings['itemThumbnail'] ?? '';
+		$xPathItemCategories = $xPathSettings['itemCategories'] ?? '';
+		if ($xPathItem == '') {
+			return null;
+		}
+
+		$html = getHtml($feedSourceUrl, $attributes);
+		if (strlen($html) <= 0) {
+			return null;
+		}
+
+		$view = new FreshRSS_View();
+		$view->_path('index/rss.phtml');
+		$view->internal_rendering = true;
+		$view->rss_url = $feedSourceUrl;
+		$view->entries = [];
+
+		try {
+			$doc = new DOMDocument();
+			$doc->recover = true;
+			$doc->strictErrorChecking = false;
+			$doc->loadHTML($html, LIBXML_NONET | LIBXML_NOERROR | LIBXML_NOWARNING);
+			$xpath = new DOMXPath($doc);
+			$view->rss_title = $xPathFeedTitle == '' ? '' : htmlspecialchars(@$xpath->evaluate('normalize-space(' . $xPathFeedTitle . ')'), ENT_COMPAT, 'UTF-8');
+			$view->rss_base = htmlspecialchars(trim($xpath->evaluate('normalize-space(//base/@href)')), ENT_COMPAT, 'UTF-8');
+			$nodes = $xpath->query($xPathItem);
+			if (empty($nodes)) {
+				return null;
+			}
+
+			foreach ($nodes as $node) {
+				$item = [];
+				$item['title'] = $xPathItemTitle == '' ? '' : @$xpath->evaluate('normalize-space(' . $xPathItemTitle . ')', $node);
+				$item['content'] = $xPathItemContent == '' ? '' : @$xpath->evaluate('normalize-space(' . $xPathItemContent . ')', $node);
+				$item['link'] = $xPathItemUri == '' ? '' : @$xpath->evaluate('normalize-space(' . $xPathItemUri . ')', $node);
+				$item['author'] = $xPathItemAuthor == '' ? '' : @$xpath->evaluate('normalize-space(' . $xPathItemAuthor . ')', $node);
+				$item['timestamp'] = $xPathItemTimestamp == '' ? '' : @$xpath->evaluate('normalize-space(' . $xPathItemTimestamp . ')', $node);
+				$item['thumbnail'] = $xPathItemThumbnail == '' ? '' : @$xpath->evaluate('normalize-space(' . $xPathItemThumbnail . ')', $node);
+				if ($xPathItemCategories != '') {
+					$itemCategories = @$xpath->query($xPathItemCategories);
+					if ($itemCategories) {
+						foreach ($itemCategories as $itemCategory) {
+							$item['categories'][] = $itemCategory->textContent;
+						}
+					}
+				}
+				if ($item['title'] . $item['content'] . $item['link'] != '') {
+					$item['guid'] = 'urn:sha1:' . sha1($item['title'] . $item['content'] . $item['link']);
+					$item = Minz_Helper::htmlspecialchars_utf8($item);
+					$view->entries[] = FreshRSS_Entry::fromArray($item);
+				}
+			}
+		} catch (Exception $ex) {
+			Minz_Log::warning($ex->getMessage());
+			return null;
+		}
+
+		if (count($view->entries) < 1) {
+			return null;
+		}
+
+		$simplePie = customSimplePie();
+		$simplePie->set_raw_data($view->renderToString());
+		$simplePie->init();
+		return $simplePie;
+	}
+
+	/**
 	 * To keep track of some new potentially unread articles since last commit+fetch from database
 	 */
 	public function incPendingUnread(int $n = 1) {
@@ -532,18 +676,23 @@ class FreshRSS_Feed extends Minz_Model {
 		return false;
 	}
 
-	protected function cacheFilename(): string {
-		$simplePie = customSimplePie($this->attributes());
-		$filename = $simplePie->get_cache_filename($this->url);
-		return CACHE_PATH . '/' . $filename . '.spc';
+	public static function cacheFilename(string $url, array $attributes, int $kind = FreshRSS_Feed::KIND_RSS): string {
+		$simplePie = customSimplePie($attributes);
+		$filename = $simplePie->get_cache_filename($url);
+		if ($kind == FreshRSS_Feed::KIND_HTML_XPATH) {
+			return CACHE_PATH . '/' . $filename . '.html';
+		} else {
+			return CACHE_PATH . '/' . $filename . '.spc';
+		}
 	}
 
 	public function clearCache(): bool {
-		return @unlink($this->cacheFilename());
+		return @unlink(FreshRSS_Feed::cacheFilename($this->url, $this->attributes(), $this->kind));
 	}
 
+	/** @return int|false */
 	public function cacheModifiedTime() {
-		return @filemtime($this->cacheFilename());
+		return @filemtime(FreshRSS_Feed::cacheFilename($this->url, $this->attributes(), $this->kind));
 	}
 
 	public function lock(): bool {
@@ -567,7 +716,7 @@ class FreshRSS_Feed extends Minz_Model {
 	 * @return array<FreshRSS_FilterAction>
 	 */
 	public function filterActions(): array {
-		if ($this->filterActions == null) {
+		if (empty($this->filterActions)) {
 			$this->filterActions = array();
 			$filters = $this->attributes('filters');
 			if (is_array($filters)) {
@@ -582,6 +731,9 @@ class FreshRSS_Feed extends Minz_Model {
 		return $this->filterActions;
 	}
 
+	/**
+	 * @param array<FreshRSS_FilterAction> $filterActions
+	 */
 	private function _filterActions($filterActions) {
 		$this->filterActions = $filterActions;
 		if (is_array($this->filterActions) && !empty($this->filterActions)) {
diff --git a/app/Models/FeedDAO.php b/app/Models/FeedDAO.php
index ab73b2ec2..c4a0b1429 100644
--- a/app/Models/FeedDAO.php
+++ b/app/Models/FeedDAO.php
@@ -5,7 +5,9 @@ class FreshRSS_FeedDAO extends Minz_ModelPdo implements FreshRSS_Searchable {
 	protected function addColumn(string $name) {
 		Minz_Log::warning(__method__ . ': ' . $name);
 		try {
-			if ($name === 'attributes') {	//v1.11.0
+			if ($name === 'kind') {	//v1.20.0
+				return $this->pdo->exec('ALTER TABLE `_feed` ADD COLUMN kind SMALLINT DEFAULT 0') !== false;
+			} elseif ($name === 'attributes') {	//v1.11.0
 				return $this->pdo->exec('ALTER TABLE `_feed` ADD COLUMN attributes TEXT') !== false;
 			}
 		} catch (Exception $e) {
@@ -17,7 +19,7 @@ class FreshRSS_FeedDAO extends Minz_ModelPdo implements FreshRSS_Searchable {
 	protected function autoUpdateDb(array $errorInfo) {
 		if (isset($errorInfo[0])) {
 			if ($errorInfo[0] === FreshRSS_DatabaseDAO::ER_BAD_FIELD_ERROR || $errorInfo[0] === FreshRSS_DatabaseDAOPGSQL::UNDEFINED_COLUMN) {
-				foreach (['attributes'] as $column) {
+				foreach (['attributes', 'kind'] as $column) {
 					if (stripos($errorInfo[2], $column) !== false) {
 						return $this->addColumn($column);
 					}
@@ -32,6 +34,7 @@ class FreshRSS_FeedDAO extends Minz_ModelPdo implements FreshRSS_Searchable {
 			INSERT INTO `_feed`
 				(
 					url,
+					kind,
 					category,
 					name,
 					website,
@@ -45,7 +48,7 @@ class FreshRSS_FeedDAO extends Minz_ModelPdo implements FreshRSS_Searchable {
 					attributes
 				)
 				VALUES
-				(?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)';
+				(?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)';
 		$stm = $this->pdo->prepare($sql);
 
 		$valuesTmp['url'] = safe_ascii($valuesTmp['url']);
@@ -59,6 +62,7 @@ class FreshRSS_FeedDAO extends Minz_ModelPdo implements FreshRSS_Searchable {
 
 		$values = array(
 			substr($valuesTmp['url'], 0, 511),
+			$valuesTmp['kind'] ?? FreshRSS_Feed::KIND_RSS,
 			$valuesTmp['category'],
 			mb_strcut(trim($valuesTmp['name']), 0, FreshRSS_DatabaseDAO::LENGTH_INDEX_UNICODE, 'UTF-8'),
 			substr($valuesTmp['website'], 0, 255),
@@ -84,7 +88,7 @@ class FreshRSS_FeedDAO extends Minz_ModelPdo implements FreshRSS_Searchable {
 		}
 	}
 
-	public function addFeedObject($feed): int {
+	public function addFeedObject(FreshRSS_Feed $feed): int {
 		// TODO: not sure if we should write this method in DAO since DAO
 		// should not be aware about feed class
 
@@ -94,6 +98,7 @@ class FreshRSS_FeedDAO extends Minz_ModelPdo implements FreshRSS_Searchable {
 			$values = array(
 				'id' => $feed->id(),
 				'url' => $feed->url(),
+				'kind' => $feed->kind(),
 				'category' => $feed->category(),
 				'name' => $feed->name(),
 				'website' => $feed->website(),
@@ -252,7 +257,7 @@ class FreshRSS_FeedDAO extends Minz_ModelPdo implements FreshRSS_Searchable {
 
 	public function selectAll() {
 		$sql = <<<'SQL'
-SELECT id, url, category, name, website, description, `lastUpdate`,
+SELECT id, url, kind, category, name, website, description, `lastUpdate`,
 	priority, `pathEntries`, `httpAuth`, error, ttl, attributes
 FROM `_feed`
 SQL;
@@ -346,7 +351,7 @@ SQL;
 	 */
 	public function listFeedsOrderUpdate(int $defaultCacheDuration = 3600, int $limit = 0) {
 		$this->updateTTL();
-		$sql = 'SELECT id, url, name, website, `lastUpdate`, `pathEntries`, `httpAuth`, ttl, attributes '
+		$sql = 'SELECT id, url, kind, name, website, `lastUpdate`, `pathEntries`, `httpAuth`, ttl, attributes '
 			. 'FROM `_feed` '
 			. ($defaultCacheDuration < 0 ? '' : 'WHERE ttl >= ' . FreshRSS_Feed::TTL_DEFAULT
 			. ' AND `lastUpdate` < (' . (time() + 60)
@@ -557,20 +562,21 @@ SQL;
 				$category = $catID;
 			}
 
-			$myFeed = new FreshRSS_Feed(isset($dao['url']) ? $dao['url'] : '', false);
+			$myFeed = new FreshRSS_Feed($dao['url'] ?? '', false);
+			$myFeed->_kind($dao['kind'] ?? FreshRSS_Feed::KIND_RSS);
 			$myFeed->_category($category);
 			$myFeed->_name($dao['name']);
-			$myFeed->_website(isset($dao['website']) ? $dao['website'] : '', false);
-			$myFeed->_description(isset($dao['description']) ? $dao['description'] : '');
-			$myFeed->_lastUpdate(isset($dao['lastUpdate']) ? $dao['lastUpdate'] : 0);
-			$myFeed->_priority(isset($dao['priority']) ? $dao['priority'] : 10);
-			$myFeed->_pathEntries(isset($dao['pathEntries']) ? $dao['pathEntries'] : '');
-			$myFeed->_httpAuth(isset($dao['httpAuth']) ? base64_decode($dao['httpAuth']) : '');
-			$myFeed->_error(isset($dao['error']) ? $dao['error'] : 0);
-			$myFeed->_ttl(isset($dao['ttl']) ? $dao['ttl'] : FreshRSS_Feed::TTL_DEFAULT);
-			$myFeed->_attributes('', isset($dao['attributes']) ? $dao['attributes'] : '');
-			$myFeed->_nbNotRead(isset($dao['cache_nbUnreads']) ? $dao['cache_nbUnreads'] : 0);
-			$myFeed->_nbEntries(isset($dao['cache_nbEntries']) ? $dao['cache_nbEntries'] : 0);
+			$myFeed->_website($dao['website'] ?? '', false);
+			$myFeed->_description($dao['description'] ?? '');
+			$myFeed->_lastUpdate($dao['lastUpdate'] ?? 0);
+			$myFeed->_priority($dao['priority'] ?? 10);
+			$myFeed->_pathEntries($dao['pathEntries'] ?? '');
+			$myFeed->_httpAuth(base64_decode($dao['httpAuth'] ?? ''));
+			$myFeed->_error($dao['error'] ?? 0);
+			$myFeed->_ttl($dao['ttl'] ?? FreshRSS_Feed::TTL_DEFAULT);
+			$myFeed->_attributes('', $dao['attributes'] ?? '');
+			$myFeed->_nbNotRead($dao['cache_nbUnreads'] ?? 0);
+			$myFeed->_nbEntries($dao['cache_nbEntries'] ?? 0);
 			if (isset($dao['id'])) {
 				$myFeed->_id($dao['id']);
 			}
diff --git a/app/Models/FeedDAOSQLite.php b/app/Models/FeedDAOSQLite.php
index 54146858b..a4432ea62 100644
--- a/app/Models/FeedDAOSQLite.php
+++ b/app/Models/FeedDAOSQLite.php
@@ -5,7 +5,7 @@ class FreshRSS_FeedDAOSQLite extends FreshRSS_FeedDAO {
 	protected function autoUpdateDb(array $errorInfo) {
 		if ($tableInfo = $this->pdo->query("PRAGMA table_info('feed')")) {
 			$columns = $tableInfo->fetchAll(PDO::FETCH_COLUMN, 1);
-			foreach (['attributes'] as $column) {
+			foreach (['attributes', 'kind'] as $column) {
 				if (!in_array($column, $columns)) {
 					return $this->addColumn($column);
 				}
diff --git a/app/Models/View.php b/app/Models/View.php
index e3a591155..365bfd261 100644
--- a/app/Models/View.php
+++ b/app/Models/View.php
@@ -7,12 +7,19 @@ class FreshRSS_View extends Minz_View {
 	public $callbackBeforeFeeds;
 	public $callbackBeforePagination;
 	public $categories;
+	/** @var FreshRSS_Category|null */
 	public $category;
+	/** @var string */
 	public $current_user;
+	/** @var array<FreshRSS_Entry> */
 	public $entries;
+	/** @var FreshRSS_Entry */
 	public $entry;
+	/** @var FreshRSS_Feed|null */
 	public $feed;
+	/** @var array<FreshRSS_Feed> */
 	public $feeds;
+	/** @var int */
 	public $nbUnreadTags;
 	public $tags;
 
@@ -88,8 +95,14 @@ class FreshRSS_View extends Minz_View {
 	public $nbPage;
 
 	// RSS view
-	public $rss_title;
-	public $url;
+	/** @var string */
+	public $rss_title = '';
+	/** @var string */
+	public $rss_url = '';
+	/** @var string */
+	public $rss_base = '';
+	/** @var boolean */
+	public $internal_rendering = false;
 
 	// Content preview
 	public $fatalError;