From 257d928ab385a6520040688ed599676735e206a8 Mon Sep 17 00:00:00 2001
From: Zentino <52699319+zent1n0@users.noreply.github.com>
Date: Fri, 20 Dec 2024 15:19:34 +0800
Subject: [PATCH] Resolve relative URLs within RSS article description

PR #21943.


---------

Co-authored-by: Vladimir Golovnev <glassez@yandex.ru>
---
 src/gui/rss/rsswidget.cpp | 47 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 47 insertions(+)

diff --git a/src/gui/rss/rsswidget.cpp b/src/gui/rss/rsswidget.cpp
index d6e7439ab..5d05df0b8 100644
--- a/src/gui/rss/rsswidget.cpp
+++ b/src/gui/rss/rsswidget.cpp
@@ -54,6 +54,48 @@
 #include "feedlistwidget.h"
 #include "ui_rsswidget.h"
 
+namespace
+{
+    void convertRelativeUrlToAbsolute(QString &html, const QString &baseUrl)
+    {
+        const QRegularExpression rx {uR"(((<a\s+[^>]*?href|<img\s+[^>]*?src)\s*=\s*["'])((https?|ftp):)?(\/\/[^\/]*)?(\/?[^\/"].*?)(["']))"_s
+            , QRegularExpression::CaseInsensitiveOption};
+
+        const QString normalizedBaseUrl = baseUrl.endsWith(u'/') ? baseUrl : (baseUrl + u'/');
+        const QUrl url {normalizedBaseUrl};
+        const QString defaultScheme = url.scheme();
+        QRegularExpressionMatchIterator iter = rx.globalMatch(html);
+
+        while (iter.hasNext())
+        {
+            const QRegularExpressionMatch match = iter.next();
+            const QString scheme = match.captured(4);
+            const QString host = match.captured(5);
+            if (!scheme.isEmpty())
+            {
+                if (host.isEmpty())
+                    break; // invalid URL, should never happen
+
+                // already absolute URL
+                continue;
+            }
+
+            QString relativePath = match.captured(6);
+            if (relativePath.startsWith(u'/'))
+                relativePath = relativePath.mid(1);
+
+            const QString absoluteUrl = !host.isEmpty()
+                    ? QString(defaultScheme + u':' + host) : (normalizedBaseUrl + relativePath);
+            const QString fullMatch = match.captured(0);
+            const QString prefix = match.captured(1);
+            const QString suffix = match.captured(7);
+
+            html.replace(fullMatch, (prefix + absoluteUrl + suffix));
+        }
+    }
+}
+
+
 RSSWidget::RSSWidget(IGUIApplication *app, QWidget *parent)
     : GUIApplicationComponent(app, parent)
     , m_ui {new Ui::RSSWidget}
@@ -605,6 +647,11 @@ void RSSWidget::renderArticle(const RSS::Article *article) const
 
         html += u"<pre>" + description + u"</pre>";
     }
+
+    // Supplement relative URLs to absolute ones
+    const QUrl url {article->link()};
+    const QString baseUrl = url.toString(QUrl::RemovePath | QUrl::RemoveQuery);
+    convertRelativeUrlToAbsolute(html, baseUrl);
     html += u"</div>";
     m_ui->textBrowser->setHtml(html);
 }