Improve CSS filter to exclude content (#6807)

* Improve CSS filter to exclude content Improve https://github.com/FreshRSS/FreshRSS/pull/6786 Fix bug when changing the CSS filter and reloading article, because original content was lost. Fix similar bug with pathEntries. * Forgotten i18n * Type check
author: Alexandre Alapetite <alexandre@alapetite.fr> 2024-09-14 00:55:41 +0200
committer: GitHub <noreply@github.com> 2024-09-14 00:55:41 +0200
commit: bea4ed69fccc7de2375363d8940af9a96e986650 (patch)
tree: 77b3656064eb010e706cb53883e3dd97aaa47667
parent: fd1b5e9343b6fe92b4e5dfbbc2f01ddfcd010af9 (diff)
24 files changed, 45 insertions, 33 deletions
diff --git a/app/Models/Entry.php b/app/Models/Entry.php
index 35d1bed42..c1ae66f65 100644
--- a/app/Models/Entry.php
+++ b/app/Models/Entry.php
@@ -176,7 +176,8 @@ class FreshRSS_Entry extends Minz_Model {
 	 * Provides the original content without additional content potentially added by loadCompleteContent().
 	 */
 	public function originalContent(): string {
-		return preg_replace('#<!-- FULLCONTENT start //-->.*<!-- FULLCONTENT end //-->#s', '', $this->content) ?? '';
+		return $this->attributeString('original_content') ??
+			preg_replace('#<!-- FULLCONTENT start //-->.*<!-- FULLCONTENT end //-->#s', '', $this->content) ?? '';
 	}
 
 	/**
@@ -897,13 +898,16 @@ HTML;
 						$originalContent = $this->originalContent();
 						switch ($feed->attributeString('content_action')) {
 							case 'prepend':
+								$this->_attribute('original_content');
 								$this->content = $fullContent . $originalContent;
 								break;
 							case 'append':
+								$this->_attribute('original_content');
 								$this->content = $originalContent . $fullContent;
 								break;
 							case 'replace':
 							default:
+								$this->_attribute('original_content', $originalContent);
 								$this->content = $fullContent;
 								break;
 						}
@@ -915,12 +919,12 @@ HTML;
 				}
 			}
 		} elseif (trim($feed->attributeString('path_entries_filter') ?? '') !== '') {
+			$originalContent = $this->attributeString('original_content') ?? $this->content;
 			$doc = new DOMDocument();
 			$utf8BOM = "\xEF\xBB\xBF";
-			if (!$doc->loadHTML($utf8BOM . $this->content, LIBXML_NONET | LIBXML_NOERROR | LIBXML_NOWARNING)) {
+			if (!$doc->loadHTML($utf8BOM . $originalContent, LIBXML_NONET | LIBXML_NOERROR | LIBXML_NOWARNING)) {
 				return false;
 			}
-			$modified = false;
 			$xpath = new DOMXPath($doc);
 			$filterednodes = $xpath->query((new Gt\CssXPath\Translator($feed->attributeString('path_entries_filter') ?? '', '//'))->asXPath()) ?: [];
 			foreach ($filterednodes as $filterednode) {
@@ -928,16 +932,24 @@ HTML;
 					continue;
 				}
 				$filterednode->parentNode->removeChild($filterednode);
-				$modified = true;
 			}
-			if ($modified) {
-				$html = $doc->saveHTML();
-				if (!is_string($html)) {
-					return false;
-				}
+			$html = $doc->saveHTML($doc->getElementsByTagName('body')->item(0) ?? $doc->firstElementChild);
+			if (!is_string($html)) {
+				return false;
+			}
+			$html = preg_replace('%^\s*<body>\s*|\s*</body>\s*$%i', '', $html);
+			$this->_attribute('original_content');
+			if (is_string($html) && $this->content !== $html) {
+				$this->_attribute('original_content', $originalContent);
 				$this->content = $html;
+				return true;
+			}
+		} else {
+			$originalContent = $this->originalContent();
+			if ($originalContent !== $this->content) {
+				$this->content = $originalContent;
+				return true;
 			}
-			return $modified;
 		}
 		return false;
 	}
diff --git a/app/i18n/cs/sub.php b/app/i18n/cs/sub.php
index c031e5efb..352189e90 100644
--- a/app/i18n/cs/sub.php
+++ b/app/i18n/cs/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Přepínač CSS článku na původních webových stránkách',
 		'css_path_filter' => array(
 			'_' => 'selektor CSS prvků, které mají být odstraněny',
-			'help' => 'Selektor CSS může být seznam, například: <kbd>.footer, .aside</kbd>',
+			'help' => 'Selektor CSS může být seznam, například: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Popis',
 		'empty' => 'Tento kanál je prázdný. Ověřte prosím, zda je stále udržován.',
diff --git a/app/i18n/de/sub.php b/app/i18n/de/sub.php
index ad622d911..f1dbed6fb 100644
--- a/app/i18n/de/sub.php
+++ b/app/i18n/de/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'CSS-Selektor des Artikelinhaltes auf der Original-Webseite',
 		'css_path_filter' => array(
 			'_' => 'CSS-Selector für die Elemente, die entfernt werden sollen',
-			'help' => 'CSS-Selector könnte eine Liste sein, wie z.B.: <kbd>.footer, .aside</kbd>',
+			'help' => 'CSS-Selector könnte eine Liste sein, wie z.B.: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Beschreibung',
 		'empty' => 'Dieser Feed ist leer. Bitte stellen Sie sicher, dass er noch gepflegt wird.',
diff --git a/app/i18n/el/sub.php b/app/i18n/el/sub.php
index b5bcc2eba..b69bd8f4c 100644
--- a/app/i18n/el/sub.php
+++ b/app/i18n/el/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Article CSS selector on original website',	// TODO
 		'css_path_filter' => array(
 			'_' => 'CSS selector of the elements to remove',	// TODO
-			'help' => 'A CSS selector may be a list such as: <kbd>.footer, .aside</kbd>',	// TODO
+			'help' => 'A CSS selector may be a list such as: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',	// TODO
 		),
 		'description' => 'Description',	// TODO
 		'empty' => 'This feed is empty. Please verify that it is still maintained.',	// TODO
diff --git a/app/i18n/en-us/sub.php b/app/i18n/en-us/sub.php
index 057f5158c..84fdb2e4d 100644
--- a/app/i18n/en-us/sub.php
+++ b/app/i18n/en-us/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Article CSS selector on original website',	// IGNORE
 		'css_path_filter' => array(
 			'_' => 'CSS selector of the elements to remove',	// IGNORE
-			'help' => 'A CSS selector may be a list such as: <kbd>.footer, .aside</kbd>',	// IGNORE
+			'help' => 'A CSS selector may be a list such as: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',	// IGNORE
 		),
 		'description' => 'Description',	// IGNORE
 		'empty' => 'This feed is empty. Please verify that it is still maintained.',	// IGNORE
diff --git a/app/i18n/en/sub.php b/app/i18n/en/sub.php
index 9b3f9e9e8..397cdb99c 100644
--- a/app/i18n/en/sub.php
+++ b/app/i18n/en/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Article CSS selector on original website',
 		'css_path_filter' => array(
 			'_' => 'CSS selector of the elements to remove',
-			'help' => 'A CSS selector may be a list such as: <kbd>.footer, .aside</kbd>',
+			'help' => 'A CSS selector may be a list such as: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Description',
 		'empty' => 'This feed is empty. Please verify that it is still maintained.',
diff --git a/app/i18n/es/sub.php b/app/i18n/es/sub.php
index a134beb57..21a21a879 100644
--- a/app/i18n/es/sub.php
+++ b/app/i18n/es/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Ruta a la CSS de los artículos en la web original',
 		'css_path_filter' => array(
 			'_' => 'Selector CSS de los elementos a remover',
-			'help' => 'Un selector CSS puede ser una lista, por ejemplo: <kbd>.footer, .aside</kbd>',
+			'help' => 'Un selector CSS puede ser una lista, por ejemplo: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Descripción',
 		'empty' => 'La fuente está vacía. Por favor, verifica que siga activa.',
diff --git a/app/i18n/fr/sub.php b/app/i18n/fr/sub.php
index 9bdbdbe5d..c8ba35d89 100644
--- a/app/i18n/fr/sub.php
+++ b/app/i18n/fr/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Sélecteur CSS des articles sur le site d’origine',
 		'css_path_filter' => array(
 			'_' => 'Sélecteur CSS des éléments à supprimer',
-			'help' => 'Un sélecteur CSS peut être une liste comme : <kbd>.footer, .aside</kbd>',
+			'help' => 'Un sélecteur CSS peut être une liste comme : <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Description',	// IGNORE
 		'empty' => 'Ce flux est vide. Veuillez vérifier qu’il est toujours maintenu.',
diff --git a/app/i18n/he/sub.php b/app/i18n/he/sub.php
index 7ac401530..b7d67a308 100644
--- a/app/i18n/he/sub.php
+++ b/app/i18n/he/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'נתיב הCSS של המאמר באתר המקורי',
 		'css_path_filter' => array(
 			'_' => 'CSS selector of the elements to remove',	// TODO
-			'help' => 'A CSS selector may be a list such as: <kbd>.footer, .aside</kbd>',	// TODO
+			'help' => 'A CSS selector may be a list such as: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',	// TODO
 		),
 		'description' => 'תיאור',
 		'empty' => 'הזנה זו ריקה. אנא ודאו שהיא עדיין מתוחזקת.',
diff --git a/app/i18n/hu/sub.php b/app/i18n/hu/sub.php
index 1b942f340..84e47fde9 100644
--- a/app/i18n/hu/sub.php
+++ b/app/i18n/hu/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Cikk CSS selector az eredeti weblapon',
 		'css_path_filter' => array(
 			'_' => 'A törlendő elemek CSS selectora',
-			'help' => 'Egy CSS selector lehet egy lista például: <kbd>.footer, .aside</kbd>',
+			'help' => 'Egy CSS selector lehet egy lista például: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Leírás',
 		'empty' => 'Ez a hírforrás üres. Ellenőrizd hogy van e tartalom rajta.',
diff --git a/app/i18n/id/sub.php b/app/i18n/id/sub.php
index 2cc0b8685..09354ab73 100644
--- a/app/i18n/id/sub.php
+++ b/app/i18n/id/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Article CSS selector on original website',	// TODO
 		'css_path_filter' => array(
 			'_' => 'CSS selector of the elements to remove',	// TODO
-			'help' => 'A CSS selector may be a list such as: <kbd>.footer, .aside</kbd>',	// TODO
+			'help' => 'A CSS selector may be a list such as: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',	// TODO
 		),
 		'description' => 'Description',	// TODO
 		'empty' => 'This feed is empty. Please verify that it is still maintained.',	// TODO
diff --git a/app/i18n/it/sub.php b/app/i18n/it/sub.php
index ae0a16123..b6f4506c9 100644
--- a/app/i18n/it/sub.php
+++ b/app/i18n/it/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Percorso del foglio di stile CSS del sito di origine',
 		'css_path_filter' => array(
 			'_' => 'Il selettore CSS degli elementi da rimuovere',
-			'help' => 'Il selettore CSS potrebbe essere una lista, ad esempio: <kbd>.footer, .aside</kbd>',
+			'help' => 'Il selettore CSS potrebbe essere una lista, ad esempio: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Descrizione',
 		'empty' => 'Questo feed non contiene articoli. Per favore verifica il sito direttamente.',
diff --git a/app/i18n/ja/sub.php b/app/i18n/ja/sub.php
index 614269caf..ad1b18622 100644
--- a/app/i18n/ja/sub.php
+++ b/app/i18n/ja/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => '元のWebサイトから記事を抽出するCSSセレクタ',
 		'css_path_filter' => array(
 			'_' => '要素を削除するCSSセレクタ',
-			'help' => 'CSSセレクタは次のようなリストです: <kbd>.footer, .aside</kbd>',
+			'help' => 'CSSセレクタは次のようなリストです: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => '説明',
 		'empty' => 'このフィードは空です。サイトが運営されているかどうかを確認してみてください。',
diff --git a/app/i18n/ko/sub.php b/app/i18n/ko/sub.php
index 57eb72ab7..42a263fe8 100644
--- a/app/i18n/ko/sub.php
+++ b/app/i18n/ko/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => '웹사이트 상의 글 본문에 해당하는 CSS 경로',
 		'css_path_filter' => array(
 			'_' => '제거할 요소의 CSS 선택자',
-			'help' => 'CSS 선택자는 다음과 같은 목록일 수 있습니다: <kbd>.footer, .aside</kbd>',
+			'help' => 'CSS 선택자는 다음과 같은 목록일 수 있습니다: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => '설명',
 		'empty' => '이 피드는 비어있습니다. 피드가 계속 운영되고 있는지 확인하세요.',
diff --git a/app/i18n/lv/sub.php b/app/i18n/lv/sub.php
index cc739194d..de86a6e08 100644
--- a/app/i18n/lv/sub.php
+++ b/app/i18n/lv/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Raksta CSS selektors sākotnējā vietnē',
 		'css_path_filter' => array(
 			'_' => 'Noņemamo elementu CSS selektors',
-			'help' => 'CSS selektors var būt saraksts, piemēram.: <kbd>.footer, .aside</kbd>',
+			'help' => 'CSS selektors var būt saraksts, piemēram.: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Apraksts',
 		'empty' => 'Šī barotne ir tukša. Lūdzu, pārbaudiet, vai tā joprojām tiek uzturēta.',
diff --git a/app/i18n/nl/sub.php b/app/i18n/nl/sub.php
index 036e631f0..c14ea5587 100644
--- a/app/i18n/nl/sub.php
+++ b/app/i18n/nl/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'CSS-pad van artikelen op originele website',
 		'css_path_filter' => array(
 			'_' => 'CSS selector van de elementen om te verwijderen',
-			'help' => 'Een CSS selector kan een lijst zijn, zoals: <kbd>.footer, .aside</kbd>',
+			'help' => 'Een CSS selector kan een lijst zijn, zoals: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Omschrijving',
 		'empty' => 'Deze feed is leeg. Controleer of deze nog actueel is.',
diff --git a/app/i18n/oc/sub.php b/app/i18n/oc/sub.php
index d503f0ead..12f14c00f 100644
--- a/app/i18n/oc/sub.php
+++ b/app/i18n/oc/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Selector CSS dels articles sul site d’origina',
 		'css_path_filter' => array(
 			'_' => 'Selector CSS de l’element de tirar',
-			'help' => 'Un selector CSS pòt èsser una lista coma : <kbd>.footer, .aside</kbd>',
+			'help' => 'Un selector CSS pòt èsser una lista coma : <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Descripcion',	// IGNORE
 		'empty' => 'Aqueste flux es void. Assegurats-vos qu’es totjorn mantengut.',
diff --git a/app/i18n/pl/sub.php b/app/i18n/pl/sub.php
index 117ec1b88..f8bf7f2ff 100644
--- a/app/i18n/pl/sub.php
+++ b/app/i18n/pl/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Selektor CSS dla wiadomości na pierwotnej stronie',
 		'css_path_filter' => array(
 			'_' => 'Selektor CSS elementów do usunięcia',
-			'help' => 'Selector CSS może być listą, na przykład: <kbd>.footer, .aside</kbd>',
+			'help' => 'Selector CSS może być listą, na przykład: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Opis',
 		'empty' => 'Ten kanał jest pusty. Należy sprawdzić czy kanał w dalszym ciągu działa.',
diff --git a/app/i18n/pt-br/sub.php b/app/i18n/pt-br/sub.php
index db178e10c..581186309 100644
--- a/app/i18n/pt-br/sub.php
+++ b/app/i18n/pt-br/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Caminho do CSS do artigo no site original',
 		'css_path_filter' => array(
 			'_' => 'Seletor CSS dos elementos a serem removidos',
-			'help' => 'O seletor CSS pode ser uma lista com: <kbd>.footer, .aside</kbd>',
+			'help' => 'O seletor CSS pode ser uma lista com: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Descrição',
 		'empty' => 'Este feed está vazio. Por favor verifique ele ainda é mantido.',
diff --git a/app/i18n/ru/sub.php b/app/i18n/ru/sub.php
index 9c7047364..ab747abb4 100644
--- a/app/i18n/ru/sub.php
+++ b/app/i18n/ru/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'CSS селектор статьи на сайте',
 		'css_path_filter' => array(
 			'_' => 'CSS селектор элемента для удаления',
-			'help' => 'CSS селектор может быть списком как: <kbd>.footer, .aside</kbd>',
+			'help' => 'CSS селектор может быть списком как: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Описание',
 		'empty' => 'Лента пустая. Пожалуйста, убедитесь, что её до сих пор обслуживают.',
diff --git a/app/i18n/sk/sub.php b/app/i18n/sk/sub.php
index 8b30f5865..d0b1f5874 100644
--- a/app/i18n/sk/sub.php
+++ b/app/i18n/sk/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Pôvodný CSS súbor článku z webovej stránky',
 		'css_path_filter' => array(
 			'_' => 'CSS selektor elementu na odstránenie',
-			'help' => 'CSS selektor môže byť zoznam ako: <kbd>.footer, .aside</kbd>',
+			'help' => 'CSS selektor môže byť zoznam ako: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Popis',
 		'empty' => 'Tento kanál je prázdny. Overte, prosím, či je ešte spravovaný autorom.',
diff --git a/app/i18n/tr/sub.php b/app/i18n/tr/sub.php
index 6b7a0aaea..a550017ab 100644
--- a/app/i18n/tr/sub.php
+++ b/app/i18n/tr/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => 'Makaleleri kendi CSS görünümü ile göster',
 		'css_path_filter' => array(
 			'_' => 'Kaldırılacak elemana ait CSS seçicisi',
-			'help' => 'CSS seçicisi şu şekilde olabilir: <kbd>.footer, .aside</kbd>',
+			'help' => 'CSS seçicisi şu şekilde olabilir: <kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => 'Tanım',
 		'empty' => 'Bu akış boş. Lütfen akışın aktif olduğuna emin olun.',
diff --git a/app/i18n/zh-cn/sub.php b/app/i18n/zh-cn/sub.php
index f1d77db1b..36bbf115c 100644
--- a/app/i18n/zh-cn/sub.php
+++ b/app/i18n/zh-cn/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => '原文的 CSS 选择器',
 		'css_path_filter' => array(
 			'_' => '需移除元素的 CSS 选择器',
-			'help' => '可设置多个 CSS 选择器，例如：<kbd>.footer, .aside</kbd>',
+			'help' => '可设置多个 CSS 选择器，例如：<kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => '描述',
 		'empty' => '此源为空。请确认它是否正常更新。',
diff --git a/app/i18n/zh-tw/sub.php b/app/i18n/zh-tw/sub.php
index 7d118cd1b..f0997ef8f 100644
--- a/app/i18n/zh-tw/sub.php
+++ b/app/i18n/zh-tw/sub.php
@@ -63,7 +63,7 @@ return array(
 		'css_path' => '原文的 CSS 選擇器',
 		'css_path_filter' => array(
 			'_' => '需移除元素的 CSS 選擇器',
-			'help' => '可設置多個 CSS 選擇器，例如：<kbd>.footer, .aside</kbd>',
+			'help' => '可設置多個 CSS 選擇器，例如：<kbd>.footer, .aside, p[data-sanitized-class="menu"]</kbd>',
 		),
 		'description' => '描述',
 		'empty' => '此源為空。請確認它是否正常更新。',
author	Alexandre Alapetite <alexandre@alapetite.fr>	2024-09-14 00:55:41 +0200
committer	GitHub <noreply@github.com>	2024-09-14 00:55:41 +0200
commit	bea4ed69fccc7de2375363d8940af9a96e986650 (patch)
tree	77b3656064eb010e706cb53883e3dd97aaa47667
parent	fd1b5e9343b6fe92b4e5dfbbc2f01ddfcd010af9 (diff)