From 94ca509d3e10885c749e11efb7d5ddfbce13daf8 Mon Sep 17 00:00:00 2001
From: Abdessamad Derraz <3028866+Abdess@users.noreply.github.com>
Date: Sun, 18 Oct 2026 03:55:51 +0200
Subject: [PATCH] fix: hold the data trees for the whole pipeline

The pipeline held its output directory for the run but let each step
lock the data directory trees on its own, so another session's
refresh could swap a tree between the packs and the manifests. The
trees are now held shared from right after the pipeline's own
refresh until it exits.
---
 scripts/pipeline.py         | 10 ++++++++++
 tests/test_artifact_lock.py | 14 ++++++++++++++
 2 files changed, 24 insertions(+)

diff --git a/scripts/pipeline.py b/scripts/pipeline.py
index 88b06cf7..56d101a0 100644
--- a/scripts/pipeline.py
+++ b/scripts/pipeline.py
@@ -46,8 +46,10 @@ from common import (
     artifact_lock,
     hold_artifact_lock,
     list_registered_platforms,
+    load_data_dir_registry,
     truth_target_dir,
 )
+from refresh_data_dirs import hold_data_dirs
 
 
 def run(cmd: list[str], label: str) -> tuple[bool, str]:
@@ -478,6 +480,14 @@ def main() -> None:
 
     all_ok = _truth_steps(args, results) and all_ok
 
+    # Every step from here reads the data directory trees: verify, the packs,
+    # the manifests and the integrity check. Held shared until the run ends,
+    # a refresh from another session waits instead of swapping a tree
+    # between the packs and the manifests, which would then describe two
+    # revisions of it. Taken after step 2, whose own refresh needs them
+    # exclusive.
+    hold_data_dirs(held, load_data_dir_registry("platforms"))
+
     # Step 3: Verify
     verify_cmd = [sys.executable, "scripts/verify.py", "--all", "--strict"]
     if args.include_archived:
diff --git a/tests/test_artifact_lock.py b/tests/test_artifact_lock.py
index abd4173c..748b109e 100644
--- a/tests/test_artifact_lock.py
+++ b/tests/test_artifact_lock.py
@@ -207,5 +207,19 @@ class HeldLockTest(unittest.TestCase):
         self.assertIn("hold_artifact_lock(args.output_dir)", source)
 
 
+class PipelineHoldsTheDataTrees(unittest.TestCase):
+    """The pipeline held dist/ for the whole run but not the data directory
+    trees: a refresh from another session could swap one between the packs
+    and the manifests, which then described two revisions of it."""
+
+    def test_the_trees_are_held_after_the_refresh_and_before_verify(self):
+        source = (Path(__file__).resolve().parents[1] / "scripts" / "pipeline.py").read_text()
+        refresh = source.index('"scripts/refresh_data_dirs.py"')
+        hold = source.index("hold_data_dirs(held,")
+        verify = source.index('"scripts/verify.py", "--all"')
+        self.assertLess(refresh, hold)
+        self.assertLess(hold, verify)
+
+
 if __name__ == "__main__":
     unittest.main()