From 0ff229de13f2bd5aa213e8741fbbe82027946273 Mon Sep 17 00:00:00 2001
From: Michael SCHAL <michaelschal@gmail.com>
Date: Mon, 22 Dec 2025 15:24:09 +0100
Subject: [PATCH] feat: add price verification script and update task.md to
 investigate comparator price extraction issues.

---
 task.md                |  9 +++++++--
 verify_search_price.py | 45 ++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 52 insertions(+), 2 deletions(-)
 create mode 100644 verify_search_price.py

diff --git a/task.md b/task.md
index 523f804..52b5469 100644
--- a/task.md
+++ b/task.md
@@ -2,9 +2,8 @@
 
 ## 🚀 Current Focus
 
+- [ ] Investigate Comparator Price Extraction (User reported false prices)
 - [x] Fix Catalog Images (Generic/Missing icons)
-- [x] Address "PDF vs Image" question
-- [x] Implement "Delete Images" (Cleanup bad data)
 
 ## 📋 Master Plan
 
@@ -27,6 +26,11 @@
   - [x] **Cleanup Bad Data**:
     - [x] Run cleanup endpoint.
   - [x] Verify fix with new scrape.
+- [/] **Comparator Price Fix**:
+  - [ ] Reproduce "False Price" issue with script.
+  - [ ] Audit `improved_search_service.py` extraction logic (Lowest Price vs AI).
+  - [ ] Implement robust price validation/filtering.
+  - [ ] Verify fix.
 
 ## 📝 Progress Log
 
@@ -34,3 +38,4 @@
 - **2025-12-22**: Fixed Catalog Retrieval using HTTP fallback.
 - **2025-12-22**: Investigating generic icon issue. Found images are served via Thumbor.
 - **2025-12-22**: **FIXED**: Scraper now targets Thumbor images. Wiped bad catalogs.
+- **2025-12-22**: User reported incorrect prices in Comparator. Investigating.
diff --git a/verify_search_price.py b/verify_search_price.py
new file mode 100644
index 0000000..433821a
--- /dev/null
+++ b/verify_search_price.py
@@ -0,0 +1,45 @@
+import asyncio
+import logging
+import sys
+import os
+
+# Identify workspace root
+sys.path.append(os.getcwd())
+
+# Mock Env Vars
+os.environ["DATABASE_URL"] = "postgresql://user:password@localhost:5432/pricewatch"
+os.environ["BROWSERLESS_URL"] = "ws://localhost:3012"  # Ignored by patch, but good for completeness
+
+from app.services.improved_search_service import ImprovedSearchService
+from playwright.async_api import async_playwright
+
+# Setup logging
+logging.basicConfig(level=logging.DEBUG)  # DEBUG level to see price extraction logic
+logger = logging.getLogger(__name__)
+
+
+# Monkey patch _connect_browser to use local launch
+async def _connect_browser_local(p):
+    logger.info("Launching local browser (headless)...")
+    return await p.chromium.launch(headless=True)
+
+
+ImprovedSearchService._connect_browser = _connect_browser_local
+
+
+async def test_search():
+    await ImprovedSearchService.initialize()
+
+    query = "nintendo switch"
+    target_site = "stokomani.fr"
+
+    print(f"Searching for: {query} on {target_site}")
+
+    async for result in ImprovedSearchService.search_site_generator(target_site, query):
+        print(f"[{result.source}] {result.title}\n  -> Price: {result.price}€\n  -> URL: {result.url}")
+
+    await ImprovedSearchService.shutdown()
+
+
+if __name__ == "__main__":
+    asyncio.run(test_search())