From 0715ba889e4c37280cd223bb9620883c203f9f03 Mon Sep 17 00:00:00 2001
From: Timothy Flynn <trflynn89@pm.me>
Date: Tue, 17 Oct 2023 11:42:55 -0400
Subject: [PATCH] LibWebView: Add method to break a URL into parts for eTLD+1
 highlighting

This is meant to serve as the method all Ladybird chromes can use to
highlight the eTLD+1 substring of the URL. It uses the Public Suffix
List to break the URL into 3 parts: the scheme and subdomain, the
eTLD+1, and all remaining parts (port, path, query, etc.).
---
 Userland/Libraries/LibWebView/URL.cpp | 47 +++++++++++++++++++++++++++
 Userland/Libraries/LibWebView/URL.h   |  8 ++++-
 2 files changed, 54 insertions(+), 1 deletion(-)

diff --git a/Userland/Libraries/LibWebView/URL.cpp b/Userland/Libraries/LibWebView/URL.cpp
index 5dd93de60b7..a210f0fc578 100644
--- a/Userland/Libraries/LibWebView/URL.cpp
+++ b/Userland/Libraries/LibWebView/URL.cpp
@@ -83,4 +83,51 @@ Optional<URL> sanitize_url(StringView url, Optional<StringView> search_engine, A
     return result.release_value();
 }
 
+static URLParts break_file_url_into_parts(URL const& url, StringView url_string)
+{
+    auto scheme = url_string.substring_view(0, url.scheme().bytes_as_string_view().length() + "://"sv.length());
+    auto path = url_string.substring_view(scheme.length());
+
+    return URLParts { scheme, path, {} };
+}
+
+static URLParts break_web_url_into_parts(URL const& url, StringView url_string)
+{
+    auto host = MUST(url.serialized_host());
+
+    auto public_suffix = MUST(PublicSuffixData::the()->get_public_suffix(host));
+    if (!public_suffix.has_value())
+        return {};
+
+    auto public_suffix_start = url_string.find(*public_suffix);
+    auto public_suffix_end = public_suffix_start.value() + public_suffix->bytes_as_string_view().length();
+
+    auto scheme_and_subdomain = url_string.substring_view(0, *public_suffix_start);
+    scheme_and_subdomain = scheme_and_subdomain.trim("."sv, TrimMode::Right);
+
+    if (auto index = scheme_and_subdomain.find_last('.'); index.has_value())
+        scheme_and_subdomain = scheme_and_subdomain.substring_view(0, *index + 1);
+    else
+        scheme_and_subdomain = scheme_and_subdomain.substring_view(0, url.scheme().bytes_as_string_view().length() + "://"sv.length());
+
+    auto effective_tld_plus_one = url_string.substring_view(scheme_and_subdomain.length(), public_suffix_end - scheme_and_subdomain.length());
+    auto remainder = url_string.substring_view(public_suffix_end);
+
+    return URLParts { scheme_and_subdomain, effective_tld_plus_one, remainder };
+}
+
+Optional<URLParts> break_url_into_parts(StringView url_string)
+{
+    auto url = URL::create_with_url_or_path(url_string);
+    if (!url.is_valid())
+        return {};
+
+    if (url.scheme() == "file"sv)
+        return break_file_url_into_parts(url, url_string);
+    if (url.scheme().is_one_of("http"sv, "https"sv, "gemini"sv))
+        return break_web_url_into_parts(url, url_string);
+
+    return {};
+}
+
 }
diff --git a/Userland/Libraries/LibWebView/URL.h b/Userland/Libraries/LibWebView/URL.h
index ea7b004ca1e..fe5e36c6dfd 100644
--- a/Userland/Libraries/LibWebView/URL.h
+++ b/Userland/Libraries/LibWebView/URL.h
@@ -16,7 +16,13 @@ enum class AppendTLD {
     No,
     Yes,
 };
-
 Optional<URL> sanitize_url(StringView, Optional<StringView> search_engine = {}, AppendTLD = AppendTLD::No);
 
+struct URLParts {
+    StringView scheme_and_subdomain;
+    StringView effective_tld_plus_one;
+    StringView remainder;
+};
+Optional<URLParts> break_url_into_parts(StringView url);
+
 }