From dc10052c1688c5790495a9bb1716e21e64c4f434 Mon Sep 17 00:00:00 2001
From: JMARyA <jmarya@hydrar.de>
Date: Fri, 3 Jan 2025 00:20:22 +0100
Subject: [PATCH] update

---
 src/archive.rs   | 19 ++++++++++++++++++-
 src/pages/mod.rs | 18 ++++++++++++++++--
 2 files changed, 34 insertions(+), 3 deletions(-)

diff --git a/src/archive.rs b/src/archive.rs
index 9c03191..fdd09ed 100644
--- a/src/archive.rs
+++ b/src/archive.rs
@@ -1,4 +1,4 @@
-use std::{io::Read, path::PathBuf};
+use std::{collections::HashSet, io::Read, path::PathBuf};
 
 use based::{request::RequestContext, result::LogAndIgnore};
 use maud::html;
@@ -39,6 +39,23 @@ fn internalize_urls(input: &str) -> String {
     .to_string()
 }
 
+/// Extract all domains
+pub fn extract_domains(input: &str) -> Vec<String> {
+    let url_pattern = r"https?://([a-zA-Z0-9.-]+)(/[\w./-]*)?";
+    let re = regex::Regex::new(url_pattern).unwrap();
+
+    let mut domains = HashSet::new();
+    for caps in re.captures_iter(input) {
+        let domain = caps[1].trim_start_matches("www.");
+        domains.insert(domain.to_string());
+    }
+
+    let mut domains: Vec<_> = domains.into_iter().collect();
+    domains.sort();
+
+    domains
+}
+
 /// Represents a directory containg archived websites
 #[derive(Debug, Clone)]
 pub struct WebsiteArchive {
diff --git a/src/pages/mod.rs b/src/pages/mod.rs
index f3a8680..003325b 100644
--- a/src/pages/mod.rs
+++ b/src/pages/mod.rs
@@ -15,7 +15,7 @@ use serde_json::json;
 
 use webarc::{
     ai::{generate_embedding, remove_data_urls, EmbedStore, SearchResult},
-    archive::WebsiteArchive,
+    archive::{extract_domains, WebsiteArchive},
     conf::get_config,
     render_page,
 };
@@ -88,7 +88,7 @@ pub async fn domain_info_route(
     let (path_entries, is_doc) = domain.paths(paths.to_str().unwrap());
     let path_seperations: Vec<&str> = paths.to_str().unwrap().split('/').collect();
 
-    // TODO : Show domains beeing linked on the page
+    let domains = extract_domains(&document.render_local(None).await.unwrap_or_default());
 
     let content = html! {
         h2 class="text-xl font-bold mb-4 flex items-center" {
@@ -134,6 +134,20 @@ pub async fn domain_info_route(
             };
         };
     };
+
+    @if !domains.is_empty() {
+        div class="max-w-md mx-auto p-4 bg-neutral-900 rounded-lg shadow-md" {
+            h3 class="font-bold mb-2" { "Domains linked on this page:" };
+            ul class="space-y-2 p-4" {
+                @for domain in domains {
+                    a href=(format!("/d/{domain}")) class="flex items-center gap-2 p-3 border bg-neutral-800 rounded hover:shadow-lg transition" {
+                        (favicon(&domain));
+                        span class="font-medium" { (domain) };
+                    };
+                };
+            };
+        };
+    };
     };
 
     render_page(content, ctx).await