From 895a520afdcfac61b6d59e79c2cb998e4490484c Mon Sep 17 00:00:00 2001
From: Abdessamad Derraz <3028866+Abdess@users.noreply.github.com>
Date: Fri, 6 Nov 2026 08:12:37 +0100
Subject: [PATCH] fix: verify --platform names uncached data dirs

A platform entry that resolves only through a data directory tree read as missing with no cause on a checkout where data/ was never refreshed, while an online pack build refreshes the trees and packs it. verify_platform now reports the uncached trees and prints them when files are missing.
---
 scripts/verify.py                    | 14 ++++++
 tests/test_verify_data_dir_notice.py | 74 ++++++++++++++++++++++++++++
 2 files changed, 88 insertions(+)
 create mode 100644 tests/test_verify_data_dir_notice.py

diff --git a/scripts/verify.py b/scripts/verify.py
index 9fbdfa9a3..8689a4dcf 100644
--- a/scripts/verify.py
+++ b/scripts/verify.py
@@ -1575,6 +1575,13 @@ def verify_platform(
         "exclusion_notes": exclusions,
         "region_fallbacks": sorted(region_fallbacks),
         **({"left_out": left_out} if no_game_data else {}),
+        # Every tree of the registry answers a declaration by its tail, so
+        # one that was never refreshed leaves its files missing here while
+        # an online pack build refreshes it and packs them.
+        "data_dir_notices": sorted(
+            key for key, entry in (data_dir_registry or {}).items()
+            if not os.path.isdir(entry.get("local_cache", ""))
+        ),
         "details": details,
         "ground_truth_coverage": {
             "with_validation": with_validation,
@@ -1952,6 +1959,13 @@ def print_platform_result(
         # them missing, as an offline build leaves them out.
         shown = ", ".join(fetched[:10]) + (f" (+{len(fetched) - 10})" if len(fetched) > 10 else "")
         print(f"  Release assets fetched when the pack is built ({len(fetched)}): {shown}")
+    uncached = result.get("data_dir_notices") or []
+    if uncached and result.get("status_counts", {}).get(Status.MISSING):
+        shown = ", ".join(uncached[:10]) + (f" (+{len(uncached) - 10})" if len(uncached) > 10 else "")
+        print(
+            f"  Data directories not cached ({len(uncached)}), their files count missing "
+            f"until refresh_data_dirs.py runs (an online pack build refreshes them): {shown}"
+        )
     for line in result.get("left_out") or []:
         print(f"  {line}")
     fallbacks = result.get("region_fallbacks") or []
diff --git a/tests/test_verify_data_dir_notice.py b/tests/test_verify_data_dir_notice.py
new file mode 100644
index 000000000..b9e4a71f9
--- /dev/null
+++ b/tests/test_verify_data_dir_notice.py
@@ -0,0 +1,74 @@
+"""verify --platform names the data directories that were never refreshed.
+
+A platform entry can resolve only through a data directory tree (AmberELEC's
+Machines/*/config.ini in the blueMSX tree). On a checkout where data/ was
+never refreshed, verify counted 200 such entries missing with no word on the
+cause, while an online pack build refreshes the trees and packs them.
+verify --emulator already named its uncached trees.
+"""
+
+from __future__ import annotations
+
+import contextlib
+import io
+import os
+import sys
+import tempfile
+import unittest
+from pathlib import Path
+
+REPO_ROOT = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(REPO_ROOT / "scripts"))
+
+import generate_db  # noqa: E402
+from verify import print_platform_result, verify_platform  # noqa: E402
+
+
+class UncachedDataDirectoryNotice(unittest.TestCase):
+    def setUp(self):
+        self._tmp = tempfile.TemporaryDirectory(dir=REPO_ROOT / "tmp")
+        self._cwd = os.getcwd()
+        os.chdir(self._tmp.name)
+        Path("emulators").mkdir()
+        Path("data/present").mkdir(parents=True)
+        self.db = {"files": {}, "indexes": generate_db.build_indexes({}, {})}
+        self.config = {
+            "platform": "Tree", "verification_mode": "md5",
+            "systems": {"msx": {"files": [{
+                "name": "config.ini", "destination": "Machines/MSX/config.ini",
+            }]}},
+        }
+        self.registry = {
+            "absent": {"local_cache": "data/absent"},
+            "present": {"local_cache": "data/present"},
+        }
+
+    def tearDown(self):
+        os.chdir(self._cwd)
+        self._tmp.cleanup()
+
+    def _run(self) -> tuple[dict, str]:
+        result = verify_platform(
+            self.config, self.db, "emulators", {}, data_dir_registry=self.registry,
+        )
+        out = io.StringIO()
+        with contextlib.redirect_stdout(out):
+            print_platform_result(result, ["Tree"])
+        return result, out.getvalue()
+
+    def test_an_uncached_tree_is_named(self):
+        result, printed = self._run()
+        self.assertEqual(result["data_dir_notices"], ["absent"])
+        self.assertIn("Data directories not cached (1)", printed)
+        self.assertIn("refresh_data_dirs.py", printed)
+        self.assertIn(": absent", printed)
+
+    def test_no_note_when_every_tree_is_cached(self):
+        Path("data/absent").mkdir()
+        result, printed = self._run()
+        self.assertEqual(result["data_dir_notices"], [])
+        self.assertNotIn("refresh_data_dirs", printed)
+
+
+if __name__ == "__main__":
+    unittest.main()