From eb4f2d8b5dd1e806ba75e326026ab97c92f676b8 Mon Sep 17 00:00:00 2001
From: Bill Thiede <git@xinu.tv>
Date: Wed, 29 Jan 2025 16:53:38 -0800
Subject: [PATCH] server: filter out bad urls when indexing search summary

---
 server/src/newsreader.rs | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

diff --git a/server/src/newsreader.rs b/server/src/newsreader.rs
index 692d4b6..1e4e7ed 100644
--- a/server/src/newsreader.rs
+++ b/server/src/newsreader.rs
@@ -2,7 +2,7 @@ use std::collections::HashMap;
 
 use cacher::FilesystemCacher;
 use futures::{stream::FuturesUnordered, StreamExt};
-use log::info;
+use log::{error, info};
 use maplit::hashmap;
 use scraper::Selector;
 use shared::compute_color;
@@ -293,11 +293,17 @@ pub async fn refresh<'ctx>(pool: &PgPool, cacher: &FilesystemCacher) -> Result<b
         .fetch_all(pool)
         .await?
         .into_iter()
-        .map(|r| {
-            let link = Url::parse(&r.link).expect("failed to parse link");
-            let body = r.clean_summary.unwrap_or("NO SUMMARY".to_string());
+        .filter_map(|r| {
+            let Ok(link) = Url::parse(&r.link) else {
+                error!("failed to parse link: {}", r.link);
+                return None;
+            };
+            let Some(body) = r.clean_summary else {
+                error!("clean_summary missing for {}", r.link);
+                return None;
+            };
             let id = r.id;
-            update_search_summary(pool, cacher, link, body, id)
+            Some(update_search_summary(pool, cacher, link, body, id))
         })
         .collect();