refactor: deduplicate articles at insertion time instead of query time

Change deduplication strategy from query-time (ROW_NUMBER window function)
to insertion-time (global guid check before insert).

Benefits:
- Simpler queries without CTE/window functions
- Consistent read state (no duplicate articles to manage)
- Better query performance (no per-query deduplication overhead)

Changes:
- Add CheckArticleExistsByGUID query for global guid lookup
- Add migration to remove existing duplicate articles
- Modify fetchOneFeed and AddFeed to skip duplicates on insert
- Revert GetUnreadArticles/GetReadArticles to simple queries

1 files changed, 24 insertions, 26 deletions

diff --git a/backend/db/queries/articles.sql b/backend/db/queries/articles.sql
index bd4cd91..134b967 100644
--- a/backend/db/queries/articles.sql
+++ b/backend/db/queries/articles.sql
@@ -7,35 +7,23 @@ INNER JOIN feeds AS f ON a.feed_id = f.id
 WHERE a.id = ?;
 
 -- name: GetUnreadArticles :many
-WITH ranked AS (
-    SELECT
-        a.id, a.feed_id, a.guid, a.title, a.url, a.is_read,
-        f.id as feed_id_2, f.url as feed_url, f.title as feed_title, f.is_subscribed as feed_is_subscribed,
-        ROW_NUMBER() OVER (PARTITION BY a.guid ORDER BY a.id) as rn
-    FROM articles AS a
-    INNER JOIN feeds AS f ON a.feed_id = f.id
-    WHERE a.is_read = 0 AND f.is_subscribed = 1 AND f.user_id = ?
-)
-SELECT id, feed_id, guid, title, url, is_read, feed_id_2, feed_url, feed_title, feed_is_subscribed
-FROM ranked
-WHERE rn = 1
-ORDER BY id DESC
+SELECT
+    a.id, a.feed_id, a.guid, a.title, a.url, a.is_read,
+    f.id as feed_id_2, f.url as feed_url, f.title as feed_title, f.is_subscribed as feed_is_subscribed
+FROM articles AS a
+INNER JOIN feeds AS f ON a.feed_id = f.id
+WHERE a.is_read = 0 AND f.is_subscribed = 1 AND f.user_id = ?
+ORDER BY a.id DESC
 LIMIT 100;
 
 -- name: GetReadArticles :many
-WITH ranked AS (
-    SELECT
-        a.id, a.feed_id, a.guid, a.title, a.url, a.is_read,
-        f.id as feed_id_2, f.url as feed_url, f.title as feed_title, f.is_subscribed as feed_is_subscribed,
-        ROW_NUMBER() OVER (PARTITION BY a.guid ORDER BY a.id) as rn
-    FROM articles AS a
-    INNER JOIN feeds AS f ON a.feed_id = f.id
-    WHERE a.is_read = 1 AND f.is_subscribed = 1 AND f.user_id = ?
-)
-SELECT id, feed_id, guid, title, url, is_read, feed_id_2, feed_url, feed_title, feed_is_subscribed
-FROM ranked
-WHERE rn = 1
-ORDER BY id DESC
+SELECT
+    a.id, a.feed_id, a.guid, a.title, a.url, a.is_read,
+    f.id as feed_id_2, f.url as feed_url, f.title as feed_title, f.is_subscribed as feed_is_subscribed
+FROM articles AS a
+INNER JOIN feeds AS f ON a.feed_id = f.id
+WHERE a.is_read = 1 AND f.is_subscribed = 1 AND f.user_id = ?
+ORDER BY a.id DESC
 LIMIT 100;
 
 -- name: GetArticlesByFeed :many
@@ -83,3 +71,13 @@ SELECT EXISTS(
     SELECT 1 FROM articles
     WHERE feed_id = ? AND guid = ?
 ) as article_exists;
+
+-- name: CheckArticleExistsByGUID :one
+SELECT EXISTS(
+    SELECT 1 FROM articles
+    WHERE guid = ?
+) as article_exists;
+
+-- name: GetAllArticleGUIDs :many
+SELECT DISTINCT guid
+FROM articles;