From fa8f60035b507ca17dc1f3b711b9725b1e80107d Mon Sep 17 00:00:00 2001
From: mwisnowski <93788087+mwisnowski@users.noreply.github.com>
Date: Fri, 20 Feb 2026 12:22:47 -0800
Subject: [PATCH] fix: delete stale raw parquet before auto-refresh (#53)

* fix: delete stale raw parquet before auto-refresh

When auto-refresh triggers, remove cached raw file to ensure fresh
MTGJSON download instead of reprocessing old data.

* docs: update changelog and release notes for stale data fix
---
 CHANGELOG.md                      |  4 ++++
 RELEASE_NOTES_TEMPLATE.md         |  3 +++
 code/web/services/orchestrator.py | 10 ++++++++++
 3 files changed, 17 insertions(+)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 4ef30dd..fdc2fb7 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -92,6 +92,10 @@ This format follows Keep a Changelog principles and aims for Semantic Versioning
   - Optimized linting rules for development workflow
 
 ### Fixed
+- **Card Data Auto-Refresh**: Fixed stale data issue when new sets are released
+  - Auto-refresh now deletes cached raw parquet file before downloading fresh data
+  - Ensures new sets are included instead of reprocessing old cached data
+  - Resolves issue where Docker volumes would retain outdated raw files
 - **Template Quality**: Resolved HTML structure issues found by validation tests
   - Fixed duplicate ID attributes in build wizard and theme picker templates
   - Removed erroneous block tags from component documentation
diff --git a/RELEASE_NOTES_TEMPLATE.md b/RELEASE_NOTES_TEMPLATE.md
index cf89adf..b04af1e 100644
--- a/RELEASE_NOTES_TEMPLATE.md
+++ b/RELEASE_NOTES_TEMPLATE.md
@@ -89,6 +89,9 @@ Web UI improvements with Tailwind CSS migration, TypeScript conversion, componen
 _None_
 
 ### Fixed
+- **Card Data Auto-Refresh**: Fixed stale data issue when new sets are released
+  - Auto-refresh now deletes cached raw parquet file before downloading fresh data
+  - Ensures new sets are included instead of reprocessing old cached data
 - **Template Quality**: Resolved HTML structure issues
   - Fixed duplicate ID attributes in templates
   - Removed erroneous template block tags
diff --git a/code/web/services/orchestrator.py b/code/web/services/orchestrator.py
index 654d5ac..9e64821 100644
--- a/code/web/services/orchestrator.py
+++ b/code/web/services/orchestrator.py
@@ -1417,6 +1417,16 @@ def _ensure_setup_ready(out, force: bool = False) -> None:
             
             try:
                 from file_setup.setup import initial_setup
+                from path_util import card_files_raw_dir
+                # Delete stale raw file to force fresh download from MTGJSON
+                # This ensures we get the latest card data (e.g., new sets like ECL)
+                raw_cards_path = os.path.join(card_files_raw_dir(), "cards.parquet")
+                if os.path.exists(raw_cards_path):
+                    try:
+                        os.remove(raw_cards_path)
+                        out(f"Removed stale raw file: {raw_cards_path}")
+                    except Exception as e:
+                        out(f"Warning: Could not remove stale raw file: {e}")
                 # Always run initial_setup when forced or when cards are missing/stale
                 initial_setup()
             except Exception as e: