From dcc9618e1f88ef7de51d03d01d87739fc28c2b2e Mon Sep 17 00:00:00 2001
From: liquidpurple <67183976+liquidpurple@users.noreply.github.com>
Date: Sat, 21 Mar 2026 14:49:14 -0700
Subject: [PATCH] Fix: stop_crawl() overwrites "completed" status
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Race condition: the cleanup thread calls stop_crawl() after a crawl has already been marked "completed", changing its status back to "stopped".

Added a status check before overwriting — if the crawl is already
"completed", leave it alone.
---
 src/crawler.py | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/src/crawler.py b/src/crawler.py
index ebf871f..79ba622 100644
--- a/src/crawler.py
+++ b/src/crawler.py
@@ -331,8 +331,11 @@ def stop_crawl(self):
         # Save final data to database
         if self.db_save_enabled and self.crawl_id:
             self._save_batch_to_db(force=True)
-            from src.crawl_db import set_crawl_status
-            set_crawl_status(self.crawl_id, 'stopped')
+            from src.crawl_db import get_crawl_by_id, set_crawl_status
+            crawl = get_crawl_by_id(self.crawl_id)
+            # Don't overwrite 'completed' status (e.g. from cleanup thread)
+            if crawl and crawl['status'] != 'completed':
+                set_crawl_status(self.crawl_id, 'stopped')
 
         # Clean up JavaScript resources if enabled
         if self.js_renderer: