fix(site-publish): close split migration boundaries

Authored-By: OpenAI (GPT-5) <noreply@openai.com>

README.md

@@ -133,9 +133,6 @@ any route's mutable objects change.
 Mutable default and override partitions receive their final cache policy before
 the matching prefix-scoped stale deletion, so publication never exposes a
 provisional cache policy or a pointer to a missing immutable target.
-Generated Ingress annotations retain each artifact's prior route. When a move
-places that retired prefix inside the new sync scope, only its declared
-immutable subtrees are excluded; a current-file collision fails publication.
 
 Artifact input directories must be pairwise disjoint after filesystem
 resolution. Publication stops before build or upload if one contains another or

scripts/deploy.py

@@ -12,8 +12,6 @@ from pathlib import Path
 from urllib.error import HTTPError, URLError
 from urllib.request import Request, urlopen
 
-import yaml
-
 from utils import (
     NAMESPACE,
     clone_apps,
@@ -178,32 +176,7 @@ def publish_route_immutables(artifact, route, site_dir, credential_env_names=Non
             publish_immutable_rule(artifact, route, rule, html_dir, aws_env)
 
 
-def retired_immutable_filters(artifact, route, html_dir, previous_path):
-    """Protect immutable keys only when an old route falls inside the new scope."""
-    if not previous_path or previous_path == route["path"]:
-        return []
-    current_prefix = route["path"].strip("/")
-    previous_prefix = previous_path.strip("/")
-    if current_prefix:
-        marker = f"{current_prefix}/"
-        if not previous_prefix.startswith(marker):
-            return []
-        previous_prefix = previous_prefix[len(marker):]
-    filters = []
-    for rule in artifact["cache_rules"]:
-        if not _is_immutable(rule):
-            continue
-        retired_path = "/".join(part for part in (previous_prefix, rule["path"]) if part)
-        collision = html_dir / retired_path
-        if collision.exists() and any(path.is_file() for path in collision.rglob("*")):
-            raise RuntimeError(
-                f"current artifact collides with retired immutable partition: {retired_path}"
-            )
-        filters.extend(("--exclude", f"{retired_path}/*"))
-    return filters
-
-
-def s3_sync(artifact, route, site_dir, credential_env_names=None, previous_path=None):
+def s3_sync(artifact, route, site_dir, credential_env_names=None):
     endpoint = artifact["s3_endpoint"]
     html_dir = site_dir / artifact["build_dir"]
     aws_env = publication_aws_env(artifact, credential_env_names)
@@ -231,13 +204,18 @@ def s3_sync(artifact, route, site_dir, credential_env_names=None, previous_path=
     # so a fresh upload always carries the right MIME type.
     specific_paths = [rule["path"] for rule in artifact["cache_rules"] if rule["path"]]
     default_filters = [arg for path in specific_paths for arg in ("--exclude", f"{path}/*")]
-    retired_filters = retired_immutable_filters(artifact, route, html_dir, previous_path)
+    # A move from a nested route to `/` makes the old partition fall inside the
+    # new sync scope. Preserve declared immutable subtrees at every retired
+    # prefix without enumerating the bucket.
+    retired_immutable_filters = [
+        arg for path in immutable_paths for arg in ("--exclude", f"*/{path}/*")
+    ]
     run(["aws", "--endpoint-url", endpoint, "s3", "cp", f"{html_dir}/", destination,
          "--recursive", "--only-show-errors", "--cache-control", default_cache,
-         *default_filters, *retired_filters, *exclude_args], env=aws_env)
+         *default_filters, *retired_immutable_filters, *exclude_args], env=aws_env)
     run(["aws", "--endpoint-url", endpoint, "s3", "sync", f"{html_dir}/", destination,
          "--delete", "--only-show-errors", "--cache-control", default_cache,
-         *default_filters, *retired_filters, *exclude_args], env=aws_env)
+         *default_filters, *retired_immutable_filters, *exclude_args], env=aws_env)
     for rule in artifact["cache_rules"]:
         if not rule["path"]:
             continue
@@ -325,27 +303,6 @@ def render_site_manifests(site_name, action_dir, app_dir, manifests_dir, cfg):
     render_templates(action_dir, template_vars, app_dir, manifests_dir)
 
 
-def previous_route_paths(app_dir):
-    """Read artifact-to-route history from generated Ingress annotations."""
-    paths = {}
-    manifests = app_dir / "manifests"
-    if not manifests.exists():
-        return paths
-    for path in sorted(manifests.glob("ingress*.yaml")):
-        document = yaml.safe_load(path.read_text()) or {}
-        annotations = document.get("metadata", {}).get("annotations", {})
-        artifact = annotations.get("site-publish.fritzlab.net/artifact")
-        route_path = annotations.get("site-publish.fritzlab.net/route-path")
-        if artifact is None and route_path is None:
-            continue
-        if not isinstance(artifact, str) or not isinstance(route_path, str) or not route_path.startswith("/"):
-            raise RuntimeError(f"invalid site-publish route history in {path}")
-        if artifact in paths:
-            raise RuntimeError(f"duplicate site-publish route history for artifact {artifact}")
-        paths[artifact] = route_path
-    return paths
-
-
 def deploy_static(site_name, site_dir, action_dir, token, cfg):
     artifact_by_name = {artifact["name"]: artifact for artifact in cfg["artifacts"]}
     credential_env_names = {
@@ -354,10 +311,6 @@ def deploy_static(site_name, site_dir, action_dir, token, cfg):
     validate_publication_environment(cfg)
     for artifact in cfg["artifacts"]:
         validate_artifact_output(site_dir, artifact)
-    apps_dir = clone_apps(token)
-    app_dir = apps_dir / "sjc001" / "websites" / site_name
-    manifests_dir = app_dir / "manifests"
-    previous_paths = previous_route_paths(app_dir)
     # Complete immutable work across the whole publication before any route's
     # mutable pointers can change. Partial immutable success is safe; mixing a
     # new route with an old route after a later immutable failure is not.
@@ -366,13 +319,14 @@ def deploy_static(site_name, site_dir, action_dir, token, cfg):
             artifact_by_name[route["artifact"]], route, site_dir, credential_env_names,
         )
     for route in cfg["routes"]:
-        s3_sync(
-            artifact_by_name[route["artifact"]], route, site_dir, credential_env_names,
-            previous_paths.get(route["artifact"]),
-        )
+        s3_sync(artifact_by_name[route["artifact"]], route, site_dir, credential_env_names)
     if cfg["compatibility"]:
         ensure_bucket_aliases(site_name, cfg["aliases"], os.environ.get("GARAGE_ADMIN_TOKEN"))
 
+    apps_dir = clone_apps(token)
+    app_dir = apps_dir / "sjc001" / "websites" / site_name
+    manifests_dir = app_dir / "manifests"
+
     render_site_manifests(site_name, action_dir, app_dir, manifests_dir, cfg)
 
     commit_and_push(apps_dir, f"Deploy {site_name}", token)

templates/ingress.yaml.j2

@@ -4,8 +4,6 @@ metadata:
   name: {{ route.resource_name }}
   namespace: {{ namespace }}
   annotations:
-    site-publish.fritzlab.net/artifact: {{ route.artifact }}
-    site-publish.fritzlab.net/route-path: {{ route.path | tojson }}
     traefik.ingress.kubernetes.io/router.middlewares: https-redirect@file,retry-upstream@file{% if route.access_middleware %},{{ route.access_middleware }}@file{% endif %}{% for m in route.middlewares %},{{ m }}@file{% endfor %}
 spec:
   ingressClassName: traefik

tests/test_contract.py

@@ -427,6 +427,7 @@ class PublishingTests(unittest.TestCase):
             rendered = [" ".join(command) for command, _ in commands]
             self.assertIn("s3://baseline-dist/dist/", rendered[0])
             self.assertIn("releases/*", rendered[0])
+            self.assertIn("*/releases/*", rendered[0])
             self.assertNotIn("--delete", rendered[0])
             self.assertIn("--delete", rendered[1])
             self.assertTrue(all("s3://baseline-dist/dist/" in command for command in rendered[1:]))
@@ -439,37 +440,12 @@ class PublishingTests(unittest.TestCase):
                 immutable_publish.call_args.args[2]["cache_control"],
             )
 
-    def test_root_move_preserves_only_actual_retired_immutable_prefix(self):
-        cfg = normalize_site_config(fixture("split-site.yaml"), "baseline.fritzlab.net")
-        artifact = next(item for item in cfg["artifacts"] if item["name"] == "distributions")
-        route = {**next(item for item in cfg["routes"] if item["artifact"] == "distributions"),
-                 "path": "/"}
-        with tempfile.TemporaryDirectory() as tmp:
-            root = Path(tmp)
-            html = root / artifact["build_dir"]
-            (html / "releases").mkdir(parents=True)
-            (html / "channels").mkdir()
-            (html / "docs" / "releases").mkdir(parents=True)
-            (html / "channels" / "stable.json").write_text("channel")
-            (html / "docs" / "releases" / "index.html").write_text("mutable")
-            commands = []
-            with patch.dict(os.environ, {
-                "DIST_S3_ACCESS_KEY": "dist-key", "DIST_S3_SECRET_KEY": "dist-secret"
-            }, clear=False), patch.object(
-                deploy, "run", side_effect=lambda command, **_: commands.append(command)
-            ):
-                deploy.s3_sync(artifact, route, root, previous_path="/foo")
-            rendered = [" ".join(command) for command in commands]
-            self.assertTrue(all("foo/releases/*" in command for command in rendered[:2]))
-            self.assertTrue(all("*/releases/*" not in command for command in rendered))
-
     def test_later_route_immutable_failure_stops_all_mutable_publication(self):
         cfg = normalize_site_config(fixture("split-site.yaml"), "baseline.fritzlab.net")
         with tempfile.TemporaryDirectory() as tmp:
             root = Path(tmp)
             with patch.object(deploy, "validate_publication_environment"), \
-                    patch.object(deploy, "validate_artifact_output"), \
-                    patch.object(deploy, "clone_apps", return_value=root / "apps"), patch.object(
+                    patch.object(deploy, "validate_artifact_output"), patch.object(
                         deploy, "publish_route_immutables",
                         side_effect=[None, RuntimeError("immutable failed")],
                     ) as immutable_publish, patch.object(deploy, "s3_sync") as mutable_sync, \