From 7e238155450fea04dac835678eec83d581de7209 Mon Sep 17 00:00:00 2001
From: Abdessamad Derraz <3028866+Abdess@users.noreply.github.com>
Date: Tue, 20 Oct 2026 00:32:26 +0200
Subject: [PATCH] fix: reports hold the data trees they read

generate_readme, generate_site and slots resolve against data/ but took no hold when run on their own, as the documented commands do. A refresh swapping a tree mid-read then gave counts from half of it. Each now holds the trees shared, like verify and the builder.
---
 scripts/generate_readme.py      | 11 ++++++++++-
 scripts/generate_site.py        |  9 ++++++++-
 scripts/slots.py                | 10 ++++++++++
 tests/test_audit_regressions.py | 22 ++++++++++++++++++++++
 4 files changed, 50 insertions(+), 2 deletions(-)

diff --git a/scripts/generate_readme.py b/scripts/generate_readme.py
index cabf8aed..574ed11a 100644
--- a/scripts/generate_readme.py
+++ b/scripts/generate_readme.py
@@ -772,9 +772,18 @@ def main() -> None:
     )
     args = parser.parse_args()
 
+    import contextlib
+
+    from common import load_data_dir_registry
+    from refresh_data_dirs import hold_data_dirs
+
     db = load_database(args.db)
 
-    readme = generate_readme(db, args.platforms_dir)
+    # Coverage resolves against data/: a refresh swapping a tree mid-read
+    # published counts taken from half of it.
+    with contextlib.ExitStack() as held:
+        hold_data_dirs(held, load_data_dir_registry(args.platforms_dir))
+        readme = generate_readme(db, args.platforms_dir)
     status = "Generated" if write_if_changed("README.md", readme) else "Unchanged"
     print(f"{status} ./README.md")
 
diff --git a/scripts/generate_site.py b/scripts/generate_site.py
index 03ca6935..d57ea385 100644
--- a/scripts/generate_site.py
+++ b/scripts/generate_site.py
@@ -4395,7 +4395,14 @@ def main() -> None:
     try:
         # Two runs over one docs/ delete each other's pages and scratch
         # files in the sweep, so the second is refused instead.
-        with artifact_lock(args.docs_dir):
+        import contextlib
+
+        from common import load_data_dir_registry
+        from refresh_data_dirs import hold_data_dirs
+
+        # The coverage pages resolve against data/, held while they are read.
+        with artifact_lock(args.docs_dir), contextlib.ExitStack() as held:
+            hold_data_dirs(held, load_data_dir_registry(args.platforms_dir))
             _generate(args)
     except ArtifactLockBusy as exc:
         print(f"ERROR: {exc}", file=sys.stderr)
diff --git a/scripts/slots.py b/scripts/slots.py
index a810b2a6..1e91c6b8 100644
--- a/scripts/slots.py
+++ b/scripts/slots.py
@@ -894,6 +894,16 @@ def main() -> int:
     )
     args = parser.parse_args()
 
+    import contextlib
+
+    from common import load_data_dir_registry
+    from refresh_data_dirs import hold_data_dirs
+
+    # Claims resolve against data/: the trees stay put until the process
+    # ends, which releases the hold.
+    held = contextlib.ExitStack()
+    hold_data_dirs(held, load_data_dir_registry(args.platforms_dir))
+
     db = load_database(args.db)
     profiles = load_emulator_profiles(args.emulators_dir)
     if args.platform:
diff --git a/tests/test_audit_regressions.py b/tests/test_audit_regressions.py
index b57c1316..0209b03f 100644
--- a/tests/test_audit_regressions.py
+++ b/tests/test_audit_regressions.py
@@ -5265,5 +5265,27 @@ class AHashScanningCoreContradictsNoPlatform(unittest.TestCase):
         self.assertEqual(claimed, [])
 
 
+class ReportsHoldTheDataTrees(unittest.TestCase):
+    """Run on their own, the README, site and slot reports resolved against
+    data/ without the shared hold, so a concurrent refresh could swap a tree
+    while they read it and the published counts came from half of it."""
+
+    def test_each_entry_point_holds_them(self):
+        import ast
+
+        for script in ("generate_readme.py", "generate_site.py", "slots.py"):
+            tree = ast.parse((ROOT / "scripts" / script).read_text(encoding="utf-8"))
+            main = next(
+                node for node in tree.body
+                if isinstance(node, ast.FunctionDef) and node.name == "main"
+            )
+            with self.subTest(script=script):
+                self.assertTrue(any(
+                    isinstance(call, ast.Call)
+                    and getattr(call.func, "id", None) == "hold_data_dirs"
+                    for call in ast.walk(main)
+                ))
+
+
 if __name__ == "__main__":
     unittest.main()