Merge remote-tracking branch 'origin/main' into list_images

api_tests/src/post.spec.ts

@@ -18,6 +18,7 @@ import {
   resolveBetaCommunity,
   createComment,
   deletePost,
+  delay,
   removePost,
   getPost,
   unfollowRemotes,
@@ -710,3 +711,25 @@ test("Fetch post via redirect", async () => {
   expect(gammaPost.post?.post.ap_id).toBe(alphaPost.post_view.post.ap_id);
   await unfollowRemotes(alpha);
 });
+
+test("Block post that contains banned URL", async () => {
+  let editSiteForm: EditSite = {
+    blocked_urls: ["https://evil.com/"],
+  };
+
+  await epsilon.editSite(editSiteForm);
+
+  await delay(500);
+
+  if (!betaCommunity) {
+    throw "Missing beta community";
+  }
+
+  expect(
+    createPost(epsilon, betaCommunity.community.id, "https://evil.com"),
+  ).rejects.toStrictEqual(Error("blocked_url"));
+
+  // Later tests need this to be empty
+  editSiteForm.blocked_urls = [];
+  await epsilon.editSite(editSiteForm);
+});

api_tests/src/user.spec.ts

@@ -45,7 +45,7 @@ test("Create user", async () => {
   if (!site.my_user) {
     throw "Missing site user";
   }
-  apShortname = `@${site.my_user.local_user_view.person.name}@lemmy-alpha:8541`;
+  apShortname = `${site.my_user.local_user_view.person.name}@lemmy-alpha:8541`;
 });
 
 test("Set some user settings, check that they are federated", async () => {
@@ -68,7 +68,7 @@ test("Delete user", async () => {
   let user = await registerUser(alpha, alphaUrl);
 
   // make a local post and comment
-  let alphaCommunity = (await resolveCommunity(user, "!main@lemmy-alpha:8541"))
+  let alphaCommunity = (await resolveCommunity(user, "main@lemmy-alpha:8541"))
     .community;
   if (!alphaCommunity) {
     throw "Missing alpha community";
@@ -134,7 +134,7 @@ test("Create user with Arabic name", async () => {
   if (!site.my_user) {
     throw "Missing site user";
   }
-  apShortname = `@${site.my_user.local_user_view.person.name}@lemmy-alpha:8541`;
+  apShortname = `${site.my_user.local_user_view.person.name}@lemmy-alpha:8541`;
 
   let alphaPerson = (await resolvePerson(alpha, apShortname)).person;
   expect(alphaPerson).toBeDefined();

crates/api/src/local_user/save_settings.rs

@@ -3,6 +3,7 @@ use lemmy_api_common::{
   context::LemmyContext,
   person::SaveUserSettings,
   utils::{
+    get_url_blocklist,
     local_site_to_slur_regex,
     process_markdown_opt,
     proxy_image_link_opt_api,
@@ -35,7 +36,10 @@ pub async fn save_user_settings(
   let site_view = SiteView::read_local(&mut context.pool()).await?;
 
   let slur_regex = local_site_to_slur_regex(&site_view.local_site);
-  let bio = diesel_option_overwrite(process_markdown_opt(&data.bio, &slur_regex, &context).await?);
+  let url_blocklist = get_url_blocklist(&context).await?;
+  let bio = diesel_option_overwrite(
+    process_markdown_opt(&data.bio, &slur_regex, &url_blocklist, &context).await?,
+  );
 
   let avatar = proxy_image_link_opt_api(&data.avatar, &context).await?;
   let banner = proxy_image_link_opt_api(&data.banner, &context).await?;

crates/api/src/site/leave_admin.rs

@@ -4,6 +4,7 @@ use lemmy_db_schema::{
   source::{
     actor_language::SiteLanguage,
     language::Language,
+    local_site_url_blocklist::LocalSiteUrlBlocklist,
     local_user::{LocalUser, LocalUserUpdateForm},
     moderator::{ModAdd, ModAddForm},
     tagline::Tagline,
@@ -62,6 +63,7 @@ pub async fn leave_admin(
   let taglines = Tagline::get_all(&mut context.pool(), site_view.local_site.id).await?;
   let custom_emojis =
     CustomEmojiView::get_all(&mut context.pool(), site_view.local_site.id).await?;
+  let blocked_urls = LocalSiteUrlBlocklist::get_all(&mut context.pool()).await?;
 
   Ok(Json(GetSiteResponse {
     site_view,
@@ -72,5 +74,6 @@ pub async fn leave_admin(
     discussion_languages,
     taglines,
     custom_emojis,
+    blocked_urls,
   }))
 }

crates/api_common/Cargo.toml

@@ -59,6 +59,8 @@ uuid = { workspace = true, optional = true }
 tokio = { workspace = true, optional = true }
 reqwest = { workspace = true, optional = true }
 ts-rs = { workspace = true, optional = true }
+moka.workspace = true
+anyhow.workspace = true
 once_cell = { workspace = true, optional = true }
 actix-web = { workspace = true, optional = true }
 enum-map = { workspace = true }

crates/api_common/src/request.rs

@@ -59,14 +59,8 @@ pub async fn fetch_link_metadata(
   let opengraph_data = extract_opengraph_data(&html_bytes, url)
     .map_err(|e| info!("{e}"))
     .unwrap_or_default();
-  let thumbnail = extract_thumbnail_from_opengraph_data(
+  let thumbnail =
-    url,
+    extract_thumbnail_from_opengraph_data(url, &opengraph_data, generate_thumbnail, context).await;
-    &opengraph_data,
-    &content_type,
-    generate_thumbnail,
-    context,
-  )
-  .await;
 
   Ok(LinkMetadata {
     opengraph_data,
@@ -158,23 +152,21 @@ fn extract_opengraph_data(html_bytes: &[u8], url: &Url) -> Result<OpenGraphData,
 pub async fn extract_thumbnail_from_opengraph_data(
   url: &Url,
   opengraph_data: &OpenGraphData,
-  content_type: &Option<Mime>,
   generate_thumbnail: bool,
   context: &LemmyContext,
 ) -> Option<DbUrl> {
-  let is_image = content_type.as_ref().unwrap_or(&mime::TEXT_PLAIN).type_() == mime::IMAGE;
+  if generate_thumbnail {
-  if generate_thumbnail && is_image {
     let image_url = opengraph_data
       .image
       .as_ref()
-      .map(lemmy_db_schema::newtypes::DbUrl::inner)
+      .map(DbUrl::inner)
       .unwrap_or(url);
     generate_pictrs_thumbnail(image_url, context)
       .await
       .ok()
       .map(Into::into)
   } else {
-    None
+    opengraph_data.image.clone()
   }
 }
 
@@ -363,7 +355,7 @@ mod tests {
       Some(mime::TEXT_HTML_UTF_8.to_string()),
       sample_res.content_type
     );
-    assert_eq!(None, sample_res.thumbnail);
+    assert!(sample_res.thumbnail.is_some());
   }
 
   // #[test]

crates/api_common/src/site.rs

@@ -6,6 +6,7 @@ use lemmy_db_schema::{
     federation_queue_state::FederationQueueState,
     instance::Instance,
     language::Language,
+    local_site_url_blocklist::LocalSiteUrlBlocklist,
     tagline::Tagline,
   },
   ListingType,
@@ -268,6 +269,8 @@ pub struct EditSite {
   pub allowed_instances: Option<Vec<String>>,
   /// A list of blocked instances.
   pub blocked_instances: Option<Vec<String>>,
+  /// A list of blocked URLs
+  pub blocked_urls: Option<Vec<String>>,
   /// A list of taglines shown at the top of the front page.
   pub taglines: Option<Vec<String>>,
   pub registration_mode: Option<RegistrationMode>,
@@ -305,6 +308,7 @@ pub struct GetSiteResponse {
   pub taglines: Vec<Tagline>,
   /// A list of custom emojis your site supports.
   pub custom_emojis: Vec<CustomEmojiView>,
+  pub blocked_urls: Vec<LocalSiteUrlBlocklist>,
 }
 
 #[skip_serializing_none]

crates/api_common/src/utils.rs

@@ -17,6 +17,7 @@ use lemmy_db_schema::{
     instance_block::InstanceBlock,
     local_site::LocalSite,
     local_site_rate_limit::LocalSiteRateLimit,
+    local_site_url_blocklist::LocalSiteUrlBlocklist,
     password_reset_request::PasswordResetRequest,
     person::{Person, PersonUpdateForm},
     person_block::PersonBlock,
@@ -38,18 +39,24 @@ use lemmy_utils::{
   rate_limit::{ActionType, BucketConfig},
   settings::structs::{PictrsImageMode, Settings},
   utils::{
-    markdown::markdown_rewrite_image_links,
+    markdown::{markdown_check_for_blocked_urls, markdown_rewrite_image_links},
     slurs::{build_slur_regex, remove_slurs},
   },
 };
-use regex::Regex;
+use moka::future::Cache;
+use once_cell::sync::Lazy;
+use regex::{escape, Regex, RegexSet};
 use rosetta_i18n::{Language, LanguageId};
-use std::collections::HashSet;
+use std::{collections::HashSet, time::Duration};
 use tracing::warn;
 use url::{ParseError, Url};
 use urlencoding::encode;
 
 pub static AUTH_COOKIE_NAME: &str = "jwt";
+#[cfg(debug_assertions)]
+static URL_BLOCKLIST_RECHECK_DELAY: Duration = Duration::from_millis(500);
+#[cfg(not(debug_assertions))]
+static URL_BLOCKLIST_RECHECK_DELAY: Duration = Duration::from_secs(60);
 
 #[tracing::instrument(skip_all)]
 pub async fn is_mod_or_admin(
@@ -516,6 +523,47 @@ pub fn local_site_opt_to_sensitive(local_site: &Option<LocalSite>) -> bool {
     .unwrap_or(false)
 }
 
+pub async fn get_url_blocklist(context: &LemmyContext) -> LemmyResult<RegexSet> {
+  static URL_BLOCKLIST: Lazy<Cache<(), RegexSet>> = Lazy::new(|| {
+    Cache::builder()
+      .max_capacity(1)
+      .time_to_live(URL_BLOCKLIST_RECHECK_DELAY)
+      .build()
+  });
+
+  Ok(
+    URL_BLOCKLIST
+      .try_get_with::<_, LemmyError>((), async {
+        let urls = LocalSiteUrlBlocklist::get_all(&mut context.pool()).await?;
+
+        let regexes = urls.iter().map(|url| {
+          let url = &url.url;
+          let parsed = Url::parse(url).expect("Coundln't parse URL.");
+          if url.ends_with('/') {
+            format!(
+              "({}://)?{}{}?",
+              parsed.scheme(),
+              escape(parsed.domain().expect("No domain.")),
+              escape(parsed.path())
+            )
+          } else {
+            format!(
+              "({}://)?{}{}",
+              parsed.scheme(),
+              escape(parsed.domain().expect("No domain.")),
+              escape(parsed.path())
+            )
+          }
+        });
+
+        let set = RegexSet::new(regexes)?;
+        Ok(set)
+      })
+      .await
+      .map_err(|e| anyhow::anyhow!("Failed to build URL blocklist due to `{}`", e))?,
+  )
+}
+
 pub async fn send_application_approved_email(
   user: &LocalUserView,
   settings: &Settings,
@@ -867,9 +915,13 @@ fn limit_expire_time(expires: DateTime<Utc>) -> LemmyResult<Option<DateTime<Utc>
 pub async fn process_markdown(
   text: &str,
   slur_regex: &Option<Regex>,
+  url_blocklist: &RegexSet,
   context: &LemmyContext,
 ) -> LemmyResult<String> {
   let text = remove_slurs(text, slur_regex);
+
+  markdown_check_for_blocked_urls(&text, url_blocklist)?;
+
   if context.settings().pictrs_config()?.image_mode() == PictrsImageMode::ProxyAllImages {
     let (text, links) = markdown_rewrite_image_links(text);
     RemoteImage::create(&mut context.pool(), links).await?;
@@ -882,10 +934,13 @@ pub async fn process_markdown(
 pub async fn process_markdown_opt(
   text: &Option<String>,
   slur_regex: &Option<Regex>,
+  url_blocklist: &RegexSet,
   context: &LemmyContext,
 ) -> LemmyResult<Option<String>> {
   match text {
-    Some(t) => process_markdown(t, slur_regex, context).await.map(Some),
+    Some(t) => process_markdown(t, slur_regex, url_blocklist, context)
+      .await
+      .map(Some),
     None => Ok(None),
   }
 }

crates/api_crud/src/comment/create.rs

@@ -10,6 +10,7 @@ use lemmy_api_common::{
     check_post_deleted_or_removed,
     generate_local_apub_endpoint,
     get_post,
+    get_url_blocklist,
     is_mod_or_admin,
     local_site_to_slur_regex,
     process_markdown,
@@ -44,7 +45,8 @@ pub async fn create_comment(
   let local_site = LocalSite::read(&mut context.pool()).await?;
 
   let slur_regex = local_site_to_slur_regex(&local_site);
-  let content = process_markdown(&data.content, &slur_regex, &context).await?;
+  let url_blocklist = get_url_blocklist(&context).await?;
+  let content = process_markdown(&data.content, &slur_regex, &url_blocklist, &context).await?;
   is_valid_body_field(&Some(content.clone()), false)?;
 
   // Check for a community ban

crates/api_crud/src/comment/update.rs

@@ -5,7 +5,12 @@ use lemmy_api_common::{
   comment::{CommentResponse, EditComment},
   context::LemmyContext,
   send_activity::{ActivityChannel, SendActivityData},
-  utils::{check_community_user_action, local_site_to_slur_regex, process_markdown_opt},
+  utils::{
+    check_community_user_action,
+    get_url_blocklist,
+    local_site_to_slur_regex,
+    process_markdown_opt,
+  },
 };
 use lemmy_db_schema::{
   source::{
@@ -54,7 +59,8 @@ pub async fn update_comment(
   .await?;
 
   let slur_regex = local_site_to_slur_regex(&local_site);
-  let content = process_markdown_opt(&data.content, &slur_regex, &context).await?;
+  let url_blocklist = get_url_blocklist(&context).await?;
+  let content = process_markdown_opt(&data.content, &slur_regex, &url_blocklist, &context).await?;
   is_valid_body_field(&content, false)?;
 
   let comment_id = data.comment_id;

crates/api_crud/src/community/create.rs

@@ -9,6 +9,7 @@ use lemmy_api_common::{
     generate_inbox_url,
     generate_local_apub_endpoint,
     generate_shared_inbox_url,
+    get_url_blocklist,
     is_admin,
     local_site_to_slur_regex,
     process_markdown_opt,
@@ -53,9 +54,11 @@ pub async fn create_community(
   }
 
   let slur_regex = local_site_to_slur_regex(&local_site);
+  let url_blocklist = get_url_blocklist(&context).await?;
   check_slurs(&data.name, &slur_regex)?;
   check_slurs(&data.title, &slur_regex)?;
-  let description = process_markdown_opt(&data.description, &slur_regex, &context).await?;
+  let description =
+    process_markdown_opt(&data.description, &slur_regex, &url_blocklist, &context).await?;
   let icon = proxy_image_link_api(&data.icon, &context).await?;
   let banner = proxy_image_link_api(&data.banner, &context).await?;
 

crates/api_crud/src/community/update.rs

@@ -7,6 +7,7 @@ use lemmy_api_common::{
   send_activity::{ActivityChannel, SendActivityData},
   utils::{
     check_community_mod_action,
+    get_url_blocklist,
     local_site_to_slur_regex,
     process_markdown_opt,
     proxy_image_link_opt_api,
@@ -36,8 +37,10 @@ pub async fn update_community(
   let local_site = LocalSite::read(&mut context.pool()).await?;
 
   let slur_regex = local_site_to_slur_regex(&local_site);
+  let url_blocklist = get_url_blocklist(&context).await?;
   check_slurs_opt(&data.title, &slur_regex)?;
-  let description = process_markdown_opt(&data.description, &slur_regex, &context).await?;
+  let description =
+    process_markdown_opt(&data.description, &slur_regex, &url_blocklist, &context).await?;
   is_valid_body_field(&data.description, false)?;
 
   let description = diesel_option_overwrite(description);

crates/api_crud/src/post/create.rs

@@ -9,6 +9,7 @@ use lemmy_api_common::{
   utils::{
     check_community_user_action,
     generate_local_apub_endpoint,
+    get_url_blocklist,
     honeypot_check,
     local_site_to_slur_regex,
     mark_post_as_read,
@@ -29,7 +30,7 @@ use lemmy_db_schema::{
   CommunityVisibility,
 };
 use lemmy_db_views::structs::LocalUserView;
-use lemmy_db_views_actor::structs::CommunityView;
+use lemmy_db_views_actor::structs::CommunityModeratorView;
 use lemmy_utils::{
   error::{LemmyError, LemmyErrorExt, LemmyErrorType},
   spawn_try_task,
@@ -38,6 +39,7 @@ use lemmy_utils::{
     validation::{
       check_url_scheme,
       clean_url_params,
+      is_url_blocked,
       is_valid_alt_text_field,
       is_valid_body_field,
       is_valid_post_title,
@@ -60,8 +62,9 @@ pub async fn create_post(
 
   let slur_regex = local_site_to_slur_regex(&local_site);
   check_slurs(&data.name, &slur_regex)?;
+  let url_blocklist = get_url_blocklist(&context).await?;
 
-  let body = process_markdown_opt(&data.body, &slur_regex, &context).await?;
+  let body = process_markdown_opt(&data.body, &slur_regex, &url_blocklist, &context).await?;
   let data_url = data.url.as_ref();
   let url = data_url.map(clean_url_params); // TODO no good way to handle a "clear"
   let custom_thumbnail = data.custom_thumbnail.as_ref().map(clean_url_params);
@@ -69,6 +72,7 @@ pub async fn create_post(
   is_valid_post_title(&data.name)?;
   is_valid_body_field(&body, true)?;
   is_valid_alt_text_field(&data.alt_text)?;
+  is_url_blocked(&url, &url_blocklist)?;
   check_url_scheme(&url)?;
   check_url_scheme(&custom_thumbnail)?;
 
@@ -83,10 +87,10 @@ pub async fn create_post(
   let community = Community::read(&mut context.pool(), community_id).await?;
   if community.posting_restricted_to_mods {
     let community_id = data.community_id;
-    let is_mod = CommunityView::is_mod_or_admin(
+    let is_mod = CommunityModeratorView::is_community_moderator(
       &mut context.pool(),
-      local_user_view.local_user.person_id,
       community_id,
+      local_user_view.local_user.person_id,
     )
     .await?;
     if !is_mod {

crates/api_crud/src/post/update.rs

@@ -8,6 +8,7 @@ use lemmy_api_common::{
   send_activity::{ActivityChannel, SendActivityData},
   utils::{
     check_community_user_action,
+    get_url_blocklist,
     local_site_to_slur_regex,
     process_markdown_opt,
     proxy_image_link_opt_apub,
@@ -30,6 +31,7 @@ use lemmy_utils::{
     validation::{
       check_url_scheme,
       clean_url_params,
+      is_url_blocked,
       is_valid_alt_text_field,
       is_valid_body_field,
       is_valid_post_title,
@@ -51,9 +53,11 @@ pub async fn update_post(
   let url = data.url.as_ref().map(clean_url_params);
   let custom_thumbnail = data.custom_thumbnail.as_ref().map(clean_url_params);
 
+  let url_blocklist = get_url_blocklist(&context).await?;
+
   let slur_regex = local_site_to_slur_regex(&local_site);
   check_slurs_opt(&data.name, &slur_regex)?;
-  let body = process_markdown_opt(&data.body, &slur_regex, &context).await?;
+  let body = process_markdown_opt(&data.body, &slur_regex, &url_blocklist, &context).await?;
 
   if let Some(name) = &data.name {
     is_valid_post_title(name)?;
@@ -61,6 +65,7 @@ pub async fn update_post(
 
   is_valid_body_field(&body, true)?;
   is_valid_alt_text_field(&data.alt_text)?;
+  is_url_blocked(&url, &url_blocklist)?;
   check_url_scheme(&url)?;
   check_url_scheme(&custom_thumbnail)?;
 

crates/api_crud/src/private_message/create.rs

@@ -8,6 +8,7 @@ use lemmy_api_common::{
     check_person_block,
     generate_local_apub_endpoint,
     get_interface_language,
+    get_url_blocklist,
     local_site_to_slur_regex,
     process_markdown,
     send_email_to_user,
@@ -36,7 +37,8 @@ pub async fn create_private_message(
   let local_site = LocalSite::read(&mut context.pool()).await?;
 
   let slur_regex = local_site_to_slur_regex(&local_site);
-  let content = process_markdown(&data.content, &slur_regex, &context).await?;
+  let url_blocklist = get_url_blocklist(&context).await?;
+  let content = process_markdown(&data.content, &slur_regex, &url_blocklist, &context).await?;
   is_valid_body_field(&Some(content.clone()), false)?;
 
   check_person_block(

crates/api_crud/src/private_message/update.rs

@@ -4,7 +4,7 @@ use lemmy_api_common::{
   context::LemmyContext,
   private_message::{EditPrivateMessage, PrivateMessageResponse},
   send_activity::{ActivityChannel, SendActivityData},
-  utils::{local_site_to_slur_regex, process_markdown},
+  utils::{get_url_blocklist, local_site_to_slur_regex, process_markdown},
 };
 use lemmy_db_schema::{
   source::{
@@ -37,7 +37,8 @@ pub async fn update_private_message(
 
   // Doing the update
   let slur_regex = local_site_to_slur_regex(&local_site);
-  let content = process_markdown(&data.content, &slur_regex, &context).await?;
+  let url_blocklist = get_url_blocklist(&context).await?;
+  let content = process_markdown(&data.content, &slur_regex, &url_blocklist, &context).await?;
   is_valid_body_field(&Some(content.clone()), false)?;
 
   let private_message_id = data.private_message_id;

crates/api_crud/src/site/create.rs

@@ -6,6 +6,7 @@ use lemmy_api_common::{
   site::{CreateSite, SiteResponse},
   utils::{
     generate_shared_inbox_url,
+    get_url_blocklist,
     is_admin,
     local_site_rate_limit_to_rate_limit_config,
     local_site_to_slur_regex,
@@ -58,7 +59,8 @@ pub async fn create_site(
   let keypair = generate_actor_keypair()?;
 
   let slur_regex = local_site_to_slur_regex(&local_site);
-  let sidebar = process_markdown_opt(&data.sidebar, &slur_regex, &context).await?;
+  let url_blocklist = get_url_blocklist(&context).await?;
+  let sidebar = process_markdown_opt(&data.sidebar, &slur_regex, &url_blocklist, &context).await?;
   let icon = proxy_image_link_opt_api(&data.icon, &context).await?;
   let banner = proxy_image_link_opt_api(&data.banner, &context).await?;
 

crates/api_crud/src/site/read.rs

@@ -6,6 +6,7 @@ use lemmy_api_common::{
 use lemmy_db_schema::source::{
   actor_language::{LocalUserLanguage, SiteLanguage},
   language::Language,
+  local_site_url_blocklist::LocalSiteUrlBlocklist,
   tagline::Tagline,
 };
 use lemmy_db_views::structs::{CustomEmojiView, LocalUserView, SiteView};
@@ -47,6 +48,7 @@ pub async fn get_site(
       let taglines = Tagline::get_all(&mut context.pool(), site_view.local_site.id).await?;
       let custom_emojis =
         CustomEmojiView::get_all(&mut context.pool(), site_view.local_site.id).await?;
+      let blocked_urls = LocalSiteUrlBlocklist::get_all(&mut context.pool()).await?;
       Ok(GetSiteResponse {
         site_view,
         admins,
@@ -56,6 +58,7 @@ pub async fn get_site(
         discussion_languages,
         taglines,
         custom_emojis,
+        blocked_urls,
       })
     })
     .await

crates/api_crud/src/site/update.rs

@@ -4,6 +4,7 @@ use lemmy_api_common::{
   context::LemmyContext,
   site::{EditSite, SiteResponse},
   utils::{
+    get_url_blocklist,
     is_admin,
     local_site_rate_limit_to_rate_limit_config,
     local_site_to_slur_regex,
@@ -18,6 +19,7 @@ use lemmy_db_schema::{
     federation_blocklist::FederationBlockList,
     local_site::{LocalSite, LocalSiteUpdateForm},
     local_site_rate_limit::{LocalSiteRateLimit, LocalSiteRateLimitUpdateForm},
+    local_site_url_blocklist::LocalSiteUrlBlocklist,
     local_user::LocalUser,
     site::{Site, SiteUpdateForm},
     tagline::Tagline,
@@ -34,6 +36,7 @@ use lemmy_utils::{
     validation::{
       build_and_check_regex,
       check_site_visibility_valid,
+      check_urls_are_valid,
       is_valid_body_field,
       site_description_length_check,
       site_name_length_check,
@@ -61,7 +64,8 @@ pub async fn update_site(
   }
 
   let slur_regex = local_site_to_slur_regex(&local_site);
-  let sidebar = process_markdown_opt(&data.sidebar, &slur_regex, &context).await?;
+  let url_blocklist = get_url_blocklist(&context).await?;
+  let sidebar = process_markdown_opt(&data.sidebar, &slur_regex, &url_blocklist, &context).await?;
   let icon = proxy_image_link_opt_api(&data.icon, &context).await?;
   let banner = proxy_image_link_opt_api(&data.banner, &context).await?;
 
@@ -137,6 +141,11 @@ pub async fn update_site(
   let blocked = data.blocked_instances.clone();
   FederationBlockList::replace(&mut context.pool(), blocked).await?;
 
+  if let Some(url_blocklist) = data.blocked_urls.clone() {
+    let parsed_urls = check_urls_are_valid(&url_blocklist)?;
+    LocalSiteUrlBlocklist::replace(&mut context.pool(), parsed_urls).await?;
+  }
+
   // TODO can't think of a better way to do this.
   // If the server suddenly requires email verification, or required applications, no old users
   // will be able to log in. It really only wants this to be a requirement for NEW signups.
@@ -578,6 +587,7 @@ mod tests {
       captcha_difficulty: None,
       allowed_instances: None,
       blocked_instances: None,
+      blocked_urls: None,
       taglines: None,
       registration_mode: site_registration_mode,
       reports_email_admins: None,

crates/apub/src/api/list_comments.rs

@@ -27,7 +27,10 @@ pub async fn list_comments(
   check_private_instance(&local_user_view, &local_site)?;
 
   let community_id = if let Some(name) = &data.community_name {
-    Some(resolve_actor_identifier::<ApubCommunity, Community>(name, &context, &None, true).await?)
+    Some(
+      resolve_actor_identifier::<ApubCommunity, Community>(name, &context, &local_user_view, true)
+        .await?,
+    )
     .map(|c| c.id)
   } else {
     data.community_id

crates/apub/src/api/list_posts.rs

@@ -30,7 +30,10 @@ pub async fn list_posts(
   let page = data.page;
   let limit = data.limit;
   let community_id = if let Some(name) = &data.community_name {
-    Some(resolve_actor_identifier::<ApubCommunity, Community>(name, &context, &None, true).await?)
+    Some(
+      resolve_actor_identifier::<ApubCommunity, Community>(name, &context, &local_user_view, true)
+        .await?,
+    )
     .map(|c| c.id)
   } else {
     data.community_id

crates/apub/src/api/resolve_object.rs

@@ -1,7 +1,6 @@
-use crate::fetcher::search::{
+use crate::fetcher::{
-  search_query_to_object_id,
+  search::{search_query_to_object_id, search_query_to_object_id_local, SearchableObjects},
-  search_query_to_object_id_local,
+  user_or_community::UserOrCommunity,
-  SearchableObjects,
 };
 use activitypub_federation::config::Data;
 use actix_web::web::{Json, Query};
@@ -31,7 +30,7 @@ pub async fn resolve_object(
 
   let res = if is_authenticated {
     // user is fully authenticated; allow remote lookups as well.
-    search_query_to_object_id(&data.q, &context).await
+    search_query_to_object_id(data.q.clone(), &context).await
   } else {
     // user isn't authenticated only allow a local search.
     search_query_to_object_id_local(&data.q, &context).await
@@ -52,14 +51,6 @@ async fn convert_response(
   let removed_or_deleted;
   let mut res = ResolveObjectResponse::default();
   match object {
-    Person(p) => {
-      removed_or_deleted = p.deleted;
-      res.person = Some(PersonView::read(pool, p.id).await?)
-    }
-    Community(c) => {
-      removed_or_deleted = c.deleted || c.removed;
-      res.community = Some(CommunityView::read(pool, c.id, user_id, false).await?)
-    }
     Post(p) => {
       removed_or_deleted = p.deleted || p.removed;
       res.post = Some(PostView::read(pool, p.id, user_id, false).await?)
@@ -68,6 +59,16 @@ async fn convert_response(
       removed_or_deleted = c.deleted || c.removed;
       res.comment = Some(CommentView::read(pool, c.id, user_id).await?)
     }
+    PersonOrCommunity(p) => match *p {
+      UserOrCommunity::User(u) => {
+        removed_or_deleted = u.deleted;
+        res.person = Some(PersonView::read(pool, u.id).await?)
+      }
+      UserOrCommunity::Community(c) => {
+        removed_or_deleted = c.deleted || c.removed;
+        res.community = Some(CommunityView::read(pool, c.id, user_id, false).await?)
+      }
+    },
   };
   // if the object was deleted from database, dont return it
   if removed_or_deleted {

crates/apub/src/fetcher/search.rs

@@ -1,6 +1,7 @@
 use crate::{
+  fetcher::user_or_community::{PersonOrGroup, UserOrCommunity},
   objects::{comment::ApubComment, community::ApubCommunity, person::ApubPerson, post::ApubPost},
-  protocol::objects::{group::Group, note::Note, page::Page, person::Person},
+  protocol::objects::{note::Note, page::Page},
 };
 use activitypub_federation::{
   config::Data,
@@ -9,7 +10,7 @@ use activitypub_federation::{
 };
 use chrono::{DateTime, Utc};
 use lemmy_api_common::context::LemmyContext;
-use lemmy_utils::error::{LemmyError, LemmyErrorType};
+use lemmy_utils::error::LemmyError;
 use serde::Deserialize;
 use url::Url;
 
@@ -18,28 +19,22 @@ use url::Url;
 /// which gets resolved to an URL.
 #[tracing::instrument(skip_all)]
 pub(crate) async fn search_query_to_object_id(
-  query: &str,
+  mut query: String,
   context: &Data<LemmyContext>,
 ) -> Result<SearchableObjects, LemmyError> {
-  Ok(match Url::parse(query) {
+  Ok(match Url::parse(&query) {
     Ok(url) => {
       // its already an url, just go with it
       ObjectId::from(url).dereference(context).await?
     }
     Err(_) => {
       // not an url, try to resolve via webfinger
-      let mut chars = query.chars();
+      if query.starts_with('!') || query.starts_with('@') {
-      let kind = chars.next();
+        query.remove(0);
-      let identifier = chars.as_str();
-      match kind {
-        Some('@') => SearchableObjects::Person(
-          webfinger_resolve_actor::<LemmyContext, ApubPerson>(identifier, context).await?,
-        ),
-        Some('!') => SearchableObjects::Community(
-          webfinger_resolve_actor::<LemmyContext, ApubCommunity>(identifier, context).await?,
-        ),
-        _ => return Err(LemmyErrorType::InvalidQuery)?,
       }
+      SearchableObjects::PersonOrCommunity(Box::new(
+        webfinger_resolve_actor::<LemmyContext, UserOrCommunity>(&query, context).await?,
+      ))
     }
   })
 }
@@ -59,19 +54,17 @@ pub(crate) async fn search_query_to_object_id_local(
 /// The types of ActivityPub objects that can be fetched directly by searching for their ID.
 #[derive(Debug)]
 pub(crate) enum SearchableObjects {
-  Person(ApubPerson),
-  Community(ApubCommunity),
   Post(ApubPost),
   Comment(ApubComment),
+  PersonOrCommunity(Box<UserOrCommunity>),
 }
 
 #[derive(Deserialize)]
 #[serde(untagged)]
 pub(crate) enum SearchableKinds {
-  Group(Group),
+  Page(Box<Page>),
-  Person(Person),
-  Page(Page),
   Note(Note),
+  PersonOrGroup(Box<PersonOrGroup>),
 }
 
 #[async_trait::async_trait]
@@ -82,10 +75,9 @@ impl Object for SearchableObjects {
 
   fn last_refreshed_at(&self) -> Option<DateTime<Utc>> {
     match self {
-      SearchableObjects::Person(p) => p.last_refreshed_at(),
-      SearchableObjects::Community(c) => c.last_refreshed_at(),
       SearchableObjects::Post(p) => p.last_refreshed_at(),
       SearchableObjects::Comment(c) => c.last_refreshed_at(),
+      SearchableObjects::PersonOrCommunity(p) => p.last_refreshed_at(),
     }
   }
 
@@ -99,13 +91,9 @@ impl Object for SearchableObjects {
     object_id: Url,
     context: &Data<Self::DataType>,
   ) -> Result<Option<Self>, LemmyError> {
-    let c = ApubCommunity::read_from_id(object_id.clone(), context).await?;
+    let uc = UserOrCommunity::read_from_id(object_id.clone(), context).await?;
-    if let Some(c) = c {
+    if let Some(uc) = uc {
-      return Ok(Some(SearchableObjects::Community(c)));
+      return Ok(Some(SearchableObjects::PersonOrCommunity(Box::new(uc))));
-    }
-    let p = ApubPerson::read_from_id(object_id.clone(), context).await?;
-    if let Some(p) = p {
-      return Ok(Some(SearchableObjects::Person(p)));
     }
     let p = ApubPost::read_from_id(object_id.clone(), context).await?;
     if let Some(p) = p {
@@ -121,10 +109,12 @@ impl Object for SearchableObjects {
   #[tracing::instrument(skip_all)]
   async fn delete(self, data: &Data<Self::DataType>) -> Result<(), LemmyError> {
     match self {
-      SearchableObjects::Person(p) => p.delete(data).await,
-      SearchableObjects::Community(c) => c.delete(data).await,
       SearchableObjects::Post(p) => p.delete(data).await,
       SearchableObjects::Comment(c) => c.delete(data).await,
+      SearchableObjects::PersonOrCommunity(pc) => match *pc {
+        UserOrCommunity::User(p) => p.delete(data).await,
+        UserOrCommunity::Community(c) => c.delete(data).await,
+      },
     }
   }
 
@@ -139,10 +129,12 @@ impl Object for SearchableObjects {
     data: &Data<Self::DataType>,
   ) -> Result<(), LemmyError> {
     match apub {
-      SearchableKinds::Group(a) => ApubCommunity::verify(a, expected_domain, data).await,
-      SearchableKinds::Person(a) => ApubPerson::verify(a, expected_domain, data).await,
       SearchableKinds::Page(a) => ApubPost::verify(a, expected_domain, data).await,
       SearchableKinds::Note(a) => ApubComment::verify(a, expected_domain, data).await,
+      SearchableKinds::PersonOrGroup(pg) => match pg.as_ref() {
+        PersonOrGroup::Person(a) => ApubPerson::verify(a, expected_domain, data).await,
+        PersonOrGroup::Group(a) => ApubCommunity::verify(a, expected_domain, data).await,
+      },
     }
   }
 
@@ -151,10 +143,11 @@ impl Object for SearchableObjects {
     use SearchableKinds as SAT;
     use SearchableObjects as SO;
     Ok(match apub {
-      SAT::Group(g) => SO::Community(ApubCommunity::from_json(g, context).await?),
+      SAT::Page(p) => SO::Post(ApubPost::from_json(*p, context).await?),
-      SAT::Person(p) => SO::Person(ApubPerson::from_json(p, context).await?),
-      SAT::Page(p) => SO::Post(ApubPost::from_json(p, context).await?),
       SAT::Note(n) => SO::Comment(ApubComment::from_json(n, context).await?),
+      SAT::PersonOrGroup(pg) => {
+        SO::PersonOrCommunity(Box::new(UserOrCommunity::from_json(*pg, context).await?))
+      }
     })
   }
 }

crates/apub/src/objects/comment.rs

@@ -18,7 +18,7 @@ use activitypub_federation::{
 use chrono::{DateTime, Utc};
 use lemmy_api_common::{
   context::LemmyContext,
-  utils::{is_mod_or_admin, local_site_opt_to_slur_regex, process_markdown},
+  utils::{get_url_blocklist, is_mod_or_admin, local_site_opt_to_slur_regex, process_markdown},
 };
 use lemmy_db_schema::{
   source::{
@@ -165,7 +165,8 @@ impl Object for ApubComment {
 
     let local_site = LocalSite::read(&mut context.pool()).await.ok();
     let slur_regex = &local_site_opt_to_slur_regex(&local_site);
-    let content = process_markdown(&content, slur_regex, context).await?;
+    let url_blocklist = get_url_blocklist(context).await?;
+    let content = process_markdown(&content, slur_regex, &url_blocklist, context).await?;
     let language_id =
       LanguageTag::to_language_id_single(note.language, &mut context.pool()).await?;
 

crates/apub/src/objects/community.rs

@@ -21,6 +21,7 @@ use lemmy_api_common::{
     generate_featured_url,
     generate_moderators_url,
     generate_outbox_url,
+    get_url_blocklist,
     local_site_opt_to_slur_regex,
     process_markdown_opt,
     proxy_image_link_opt_apub,
@@ -141,8 +142,10 @@ impl Object for ApubCommunity {
 
     let local_site = LocalSite::read(&mut context.pool()).await.ok();
     let slur_regex = &local_site_opt_to_slur_regex(&local_site);
+    let url_blocklist = get_url_blocklist(context).await?;
     let description = read_from_string_or_source_opt(&group.summary, &None, &group.source);
-    let description = process_markdown_opt(&description, slur_regex, context).await?;
+    let description =
+      process_markdown_opt(&description, slur_regex, &url_blocklist, context).await?;
     let icon = proxy_image_link_opt_apub(group.icon.map(|i| i.url), context).await?;
     let banner = proxy_image_link_opt_apub(group.image.map(|i| i.url), context).await?;
 
@@ -177,18 +180,21 @@ impl Object for ApubCommunity {
 
     let community: ApubCommunity = community.into();
 
-    // Fetching mods and outbox is not necessary for Lemmy to work, so ignore errors. Besides,
+    // These collections are not necessary for Lemmy to work, so ignore errors.
-    // we need to ignore these errors so that tests can work entirely offline.
     let community_ = community.clone();
     let context_ = context.reset_request_count();
     spawn_try_task(async move {
-      group.outbox.dereference(&community_, &context_).await?;
+      group.outbox.dereference(&community_, &context_).await.ok();
-      group.followers.dereference(&community_, &context_).await?;
+      group
+        .followers
+        .dereference(&community_, &context_)
+        .await
+        .ok();
       if let Some(featured) = group.featured {
-        featured.dereference(&community_, &context_).await?;
+        featured.dereference(&community_, &context_).await.ok();
       }
       if let Some(moderators) = group.attributed_to {
-        moderators.dereference(&community_, &context_).await?;
+        moderators.dereference(&community_, &context_).await.ok();
       }
       Ok(())
     });

crates/apub/src/objects/instance.rs

@@ -19,7 +19,12 @@ use activitypub_federation::{
 use chrono::{DateTime, Utc};
 use lemmy_api_common::{
   context::LemmyContext,
-  utils::{local_site_opt_to_slur_regex, process_markdown_opt, proxy_image_link_opt_apub},
+  utils::{
+    get_url_blocklist,
+    local_site_opt_to_slur_regex,
+    process_markdown_opt,
+    proxy_image_link_opt_apub,
+  },
 };
 use lemmy_db_schema::{
   newtypes::InstanceId,
@@ -138,8 +143,9 @@ impl Object for ApubSite {
 
     let local_site = LocalSite::read(&mut context.pool()).await.ok();
     let slur_regex = &local_site_opt_to_slur_regex(&local_site);
+    let url_blocklist = get_url_blocklist(context).await?;
     let sidebar = read_from_string_or_source_opt(&apub.content, &None, &apub.source);
-    let sidebar = process_markdown_opt(&sidebar, slur_regex, context).await?;
+    let sidebar = process_markdown_opt(&sidebar, slur_regex, &url_blocklist, context).await?;
     let icon = proxy_image_link_opt_apub(apub.icon.map(|i| i.url), context).await?;
     let banner = proxy_image_link_opt_apub(apub.image.map(|i| i.url), context).await?;
 

crates/apub/src/objects/person.rs

@@ -22,6 +22,7 @@ use lemmy_api_common::{
   context::LemmyContext,
   utils::{
     generate_outbox_url,
+    get_url_blocklist,
     local_site_opt_to_slur_regex,
     process_markdown_opt,
     proxy_image_link_opt_apub,
@@ -152,8 +153,9 @@ impl Object for ApubPerson {
 
     let local_site = LocalSite::read(&mut context.pool()).await.ok();
     let slur_regex = &local_site_opt_to_slur_regex(&local_site);
+    let url_blocklist = get_url_blocklist(context).await?;
     let bio = read_from_string_or_source_opt(&person.summary, &None, &person.source);
-    let bio = process_markdown_opt(&bio, slur_regex, context).await?;
+    let bio = process_markdown_opt(&bio, slur_regex, &url_blocklist, context).await?;
     let avatar = proxy_image_link_opt_apub(person.icon.map(|i| i.url), context).await?;
     let banner = proxy_image_link_opt_apub(person.image.map(|i| i.url), context).await?;
 

crates/apub/src/objects/post.rs

@@ -26,7 +26,7 @@ use lemmy_api_common::{
   context::LemmyContext,
   request::fetch_link_metadata_opt,
   utils::{
-    is_mod_or_admin,
+    get_url_blocklist,
     local_site_opt_to_sensitive,
     local_site_opt_to_slur_regex,
     process_markdown_opt,
@@ -43,6 +43,7 @@ use lemmy_db_schema::{
   },
   traits::Crud,
 };
+use lemmy_db_views_actor::structs::CommunityModeratorView;
 use lemmy_utils::{
   error::LemmyError,
   utils::{markdown::markdown_to_html, slurs::check_slurs_opt, validation::check_url_scheme},
@@ -185,7 +186,8 @@ impl Object for ApubPost {
     let creator = page.creator()?.dereference(context).await?;
     let community = page.community(context).await?;
     if community.posting_restricted_to_mods {
-      is_mod_or_admin(&mut context.pool(), &creator, community.id).await?;
+      CommunityModeratorView::is_community_moderator(&mut context.pool(), community.id, creator.id)
+        .await?;
     }
     let mut name = page
       .name
@@ -245,9 +247,10 @@ impl Object for ApubPost {
       let thumbnail_url = proxy_image_link_opt_apub(thumbnail_url, context).await?;
 
       let slur_regex = &local_site_opt_to_slur_regex(&local_site);
+      let url_blocklist = get_url_blocklist(context).await?;
 
       let body = read_from_string_or_source_opt(&page.content, &page.media_type, &page.source);
-      let body = process_markdown_opt(&body, slur_regex, context).await?;
+      let body = process_markdown_opt(&body, slur_regex, &url_blocklist, context).await?;
       let language_id =
         LanguageTag::to_language_id_single(page.language, &mut context.pool()).await?;
 

crates/apub/src/objects/private_message.rs

@@ -14,7 +14,7 @@ use activitypub_federation::{
 use chrono::{DateTime, Utc};
 use lemmy_api_common::{
   context::LemmyContext,
-  utils::{check_person_block, local_site_opt_to_slur_regex, process_markdown},
+  utils::{check_person_block, get_url_blocklist, local_site_opt_to_slur_regex, process_markdown},
 };
 use lemmy_db_schema::{
   source::{
@@ -127,8 +127,9 @@ impl Object for ApubPrivateMessage {
 
     let local_site = LocalSite::read(&mut context.pool()).await.ok();
     let slur_regex = &local_site_opt_to_slur_regex(&local_site);
+    let url_blocklist = get_url_blocklist(context).await?;
     let content = read_from_string_or_source(&note.content, &None, &note.source);
-    let content = process_markdown(&content, slur_regex, context).await?;
+    let content = process_markdown(&content, slur_regex, &url_blocklist, context).await?;
 
     let form = PrivateMessageInsertForm {
       creator_id: creator.id,

crates/db_schema/src/impls/local_site_url_blocklist.rs

@@ -0,0 +1,49 @@
+use crate::{
+  schema::local_site_url_blocklist,
+  source::local_site_url_blocklist::{LocalSiteUrlBlocklist, LocalSiteUrlBlocklistForm},
+  utils::{get_conn, DbPool},
+};
+use diesel::{dsl::insert_into, result::Error};
+use diesel_async::{AsyncPgConnection, RunQueryDsl};
+
+impl LocalSiteUrlBlocklist {
+  pub async fn replace(pool: &mut DbPool<'_>, url_blocklist: Vec<String>) -> Result<(), Error> {
+    let conn = &mut get_conn(pool).await?;
+
+    conn
+      .build_transaction()
+      .run(|conn| {
+        Box::pin(async move {
+          use crate::schema::local_site_url_blocklist::dsl::local_site_url_blocklist;
+
+          Self::clear(conn).await?;
+
+          let forms = url_blocklist
+            .into_iter()
+            .map(|url| LocalSiteUrlBlocklistForm { url, updated: None })
+            .collect::<Vec<_>>();
+
+          insert_into(local_site_url_blocklist)
+            .values(forms)
+            .execute(conn)
+            .await?;
+
+          Ok(())
+        }) as _
+      })
+      .await
+  }
+
+  async fn clear(conn: &mut AsyncPgConnection) -> Result<usize, Error> {
+    diesel::delete(local_site_url_blocklist::table)
+      .execute(conn)
+      .await
+  }
+
+  pub async fn get_all(pool: &mut DbPool<'_>) -> Result<Vec<Self>, Error> {
+    let conn = &mut get_conn(pool).await?;
+    local_site_url_blocklist::table
+      .get_results::<Self>(conn)
+      .await
+  }
+}

crates/db_schema/src/impls/mod.rs

@@ -17,6 +17,7 @@ pub mod instance_block;
 pub mod language;
 pub mod local_site;
 pub mod local_site_rate_limit;
+pub mod local_site_url_blocklist;
 pub mod local_user;
 pub mod local_user_vote_display_mode;
 pub mod login_token;

crates/db_schema/src/schema.rs

@@ -409,6 +409,15 @@ diesel::table! {
     }
 }
 
+diesel::table! {
+    local_site_url_blocklist (id) {
+        id -> Int4,
+        url -> Text,
+        published -> Timestamptz,
+        updated -> Nullable<Timestamptz>,
+    }
+}
+
 diesel::table! {
     use diesel::sql_types::*;
     use super::sql_types::SortTypeEnum;
@@ -1052,6 +1061,7 @@ diesel::allow_tables_to_appear_in_same_query!(
     local_image,
     local_site,
     local_site_rate_limit,
+    local_site_url_blocklist,
     local_user,
     local_user_language,
     local_user_vote_display_mode,

crates/db_schema/src/source/local_site_url_blocklist.rs

@@ -0,0 +1,28 @@
+#[cfg(feature = "full")]
+use crate::schema::local_site_url_blocklist;
+use chrono::{DateTime, Utc};
+use serde::{Deserialize, Serialize};
+use serde_with::skip_serializing_none;
+#[cfg(feature = "full")]
+use ts_rs::TS;
+
+#[skip_serializing_none]
+#[derive(Clone, PartialEq, Eq, Debug, Serialize, Deserialize)]
+#[cfg_attr(feature = "full", derive(Queryable, Selectable, Identifiable, TS))]
+#[cfg_attr(feature = "full", diesel(table_name = local_site_url_blocklist))]
+#[cfg_attr(feature = "full", diesel(check_for_backend(diesel::pg::Pg)))]
+#[cfg_attr(feature = "full", ts(export))]
+pub struct LocalSiteUrlBlocklist {
+  pub id: i32,
+  pub url: String,
+  pub published: DateTime<Utc>,
+  pub updated: Option<DateTime<Utc>>,
+}
+
+#[derive(Default, Clone)]
+#[cfg_attr(feature = "full", derive(Insertable, AsChangeset))]
+#[cfg_attr(feature = "full", diesel(table_name = local_site_url_blocklist))]
+pub struct LocalSiteUrlBlocklistForm {
+  pub url: String,
+  pub updated: Option<DateTime<Utc>>,
+}

crates/db_schema/src/source/mod.rs

@@ -22,6 +22,7 @@ pub mod instance_block;
 pub mod language;
 pub mod local_site;
 pub mod local_site_rate_limit;
+pub mod local_site_url_blocklist;
 pub mod local_user;
 pub mod local_user_vote_display_mode;
 pub mod login_token;

crates/utils/src/error.rs

@@ -135,6 +135,7 @@ pub enum LemmyErrorType {
   CouldntSetAllRegistrationsAccepted,
   CouldntSetAllEmailVerified,
   Banned,
+  BlockedUrl,
   CouldntGetComments,
   CouldntGetPosts,
   InvalidUrl,

crates/utils/src/utils/markdown/mod.rs

@@ -1,6 +1,7 @@
-use crate::settings::SETTINGS;
+use crate::{error::LemmyResult, settings::SETTINGS, LemmyErrorType};
 use markdown_it::{plugins::cmark::inline::image::Image, MarkdownIt};
 use once_cell::sync::Lazy;
+use regex::RegexSet;
 use url::Url;
 use urlencoding::encode;
 
@@ -98,6 +99,13 @@ pub fn markdown_rewrite_image_links(mut src: String) -> (String, Vec<Url>) {
   (src, links)
 }
 
+pub fn markdown_check_for_blocked_urls(text: &str, blocklist: &RegexSet) -> LemmyResult<()> {
+  if blocklist.is_match(text) {
+    Err(LemmyErrorType::BlockedUrl)?
+  }
+  Ok(())
+}
+
 #[cfg(test)]
 mod tests {
   #![allow(clippy::unwrap_used)]
@@ -237,6 +245,69 @@ mod tests {
     });
   }
 
+  #[test]
+  fn test_url_blocking() {
+    let set = RegexSet::new(vec![r"(https://)?example\.com/?"]).unwrap();
+
+    assert!(
+      markdown_check_for_blocked_urls(&String::from("[](https://example.com)"), &set).is_err()
+    );
+
+    assert!(markdown_check_for_blocked_urls(
+      &String::from("Go to https://example.com to get free Robux"),
+      &set
+    )
+    .is_err());
+
+    assert!(
+      markdown_check_for_blocked_urls(&String::from("[](https://example.blog)"), &set).is_ok()
+    );
+
+    assert!(markdown_check_for_blocked_urls(&String::from("example.com"), &set).is_err());
+
+    assert!(markdown_check_for_blocked_urls(
+      "Odio exercitationem culpa sed sunt
+      et. Sit et similique tempora deserunt doloremque. Cupiditate iusto
+      repellat et quis qui. Cum veritatis facere quasi repellendus sunt
+      eveniet nemo sint. Cumque sit unde est. https://example.com Alias
+      repellendus at quos.",
+      &set
+    )
+    .is_err());
+
+    let set = RegexSet::new(vec![r"(https://)?example\.com/spam\.jpg"]).unwrap();
+    assert!(markdown_check_for_blocked_urls(
+      &String::from("![](https://example.com/spam.jpg)"),
+      &set
+    )
+    .is_err());
+
+    let set = RegexSet::new(vec![
+      r"(https://)?quo\.example\.com/?",
+      r"(https://)?foo\.example\.com/?",
+      r"(https://)?bar\.example\.com/?",
+    ])
+    .unwrap();
+
+    assert!(
+      markdown_check_for_blocked_urls(&String::from("https://baz.example.com"), &set).is_ok()
+    );
+
+    assert!(
+      markdown_check_for_blocked_urls(&String::from("https://bar.example.com"), &set).is_err()
+    );
+
+    let set = RegexSet::new(vec![r"(https://)?example\.com/banned_page"]).unwrap();
+
+    assert!(
+      markdown_check_for_blocked_urls(&String::from("https://example.com/page"), &set).is_ok()
+    );
+
+    let set = RegexSet::new(vec![r"(https://)?ex\.mple\.com/?"]).unwrap();
+
+    assert!(markdown_check_for_blocked_urls("example.com", &set).is_ok());
+  }
+
   #[test]
   fn test_sanitize_html() {
     let sanitized = sanitize_html("<script>alert('xss');</script> hello &\"'");

crates/utils/src/utils/validation.rs

@@ -1,8 +1,8 @@
 use crate::error::{LemmyErrorExt, LemmyErrorType, LemmyResult};
 use itertools::Itertools;
 use once_cell::sync::Lazy;
-use regex::{Regex, RegexBuilder};
+use regex::{Regex, RegexBuilder, RegexSet};
-use url::Url;
+use url::{ParseError, Url};
 
 // From here: https://github.com/vector-im/element-android/blob/develop/matrix-sdk-android/src/main/java/org/matrix/android/sdk/api/MatrixPatterns.kt#L35
 static VALID_MATRIX_ID_REGEX: Lazy<Regex> = Lazy::new(|| {
@@ -299,6 +299,33 @@ pub fn check_url_scheme(url: &Option<Url>) -> LemmyResult<()> {
   }
 }
 
+pub fn is_url_blocked(url: &Option<Url>, blocklist: &RegexSet) -> LemmyResult<()> {
+  if let Some(url) = url {
+    if blocklist.is_match(url.as_str()) {
+      Err(LemmyErrorType::BlockedUrl)?
+    }
+  }
+
+  Ok(())
+}
+
+pub fn check_urls_are_valid(urls: &Vec<String>) -> LemmyResult<Vec<String>> {
+  let mut parsed_urls = vec![];
+  for url in urls {
+    let url = Url::parse(url).or_else(|e| {
+      if e == ParseError::RelativeUrlWithoutBase {
+        Url::parse(&format!("https://{url}"))
+      } else {
+        Err(e)
+      }
+    })?;
+
+    parsed_urls.push(url.to_string());
+  }
+
+  Ok(parsed_urls)
+}
+
 #[cfg(test)]
 mod tests {
   #![allow(clippy::unwrap_used)]
@@ -310,7 +337,9 @@ mod tests {
       build_and_check_regex,
       check_site_visibility_valid,
       check_url_scheme,
+      check_urls_are_valid,
       clean_url_params,
+      is_url_blocked,
       is_valid_actor_name,
       is_valid_bio_field,
       is_valid_display_name,
@@ -550,4 +579,38 @@ mod tests {
     let magnet_link="magnet:?xt=urn:btih:4b390af3891e323778959d5abfff4b726510f14c&dn=Ravel%20Complete%20Piano%20Sheet%20Music%20-%20Public%20Domain&tr=udp%3A%2F%2Fopen.tracker.cl%3A1337%2Fannounce";
     assert!(check_url_scheme(&Some(Url::parse(magnet_link).unwrap())).is_ok());
   }
+
+  #[test]
+  fn test_url_block() {
+    let set = regex::RegexSet::new(vec![
+      r"(https://)?example\.org/page/to/article",
+      r"(https://)?example\.net/?",
+      r"(https://)?example\.com/?",
+    ])
+    .unwrap();
+
+    assert!(is_url_blocked(&Some(Url::parse("https://example.blog").unwrap()), &set).is_ok());
+
+    assert!(is_url_blocked(&Some(Url::parse("https://example.org").unwrap()), &set).is_ok());
+
+    assert!(is_url_blocked(&None, &set).is_ok());
+
+    assert!(is_url_blocked(&Some(Url::parse("https://example.com").unwrap()), &set).is_err());
+  }
+
+  #[test]
+  fn test_url_parsed() {
+    assert_eq!(
+      vec![String::from("https://example.com/")],
+      check_urls_are_valid(&vec![String::from("example.com")]).unwrap()
+    );
+
+    assert!(check_urls_are_valid(&vec![
+      String::from("example.com"),
+      String::from("https://example.blog")
+    ])
+    .is_ok());
+
+    assert!(check_urls_are_valid(&vec![String::from("https://example .com"),]).is_err());
+  }
 }

migrations/2024-03-06-201637_url_blocklist/down.sql

@@ -0,0 +1,3 @@
+-- This file should undo anything in `up.sql`
+DROP TABLE local_site_url_blocklist;
+

migrations/2024-03-06-201637_url_blocklist/up.sql

@@ -0,0 +1,7 @@
+CREATE TABLE local_site_url_blocklist (
+    id serial NOT NULL PRIMARY KEY,
+    url text NOT NULL UNIQUE,
+    published timestamp with time zone NOT NULL DEFAULT now(),
+    updated timestamp with time zone
+);
+