fix(site-publish): close split migration boundaries

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

scripts/deploy.py

@@ -12,6 +12,8 @@ from pathlib import Path
 from urllib.error import HTTPError, URLError
 from urllib.request import Request, urlopen
 
+import yaml
+
 from utils import (
     NAMESPACE,
     clone_apps,
@@ -176,7 +178,37 @@ def publish_route_immutables(artifact, route, site_dir, credential_env_names=Non
             publish_immutable_rule(artifact, route, rule, html_dir, aws_env)
 
 
-def s3_sync(artifact, route, site_dir, credential_env_names=None):
+def retired_immutable_filters(artifact, route, html_dir, previous_contract):
+    """Protect immutable keys only when an old route falls inside the new scope."""
+    if not previous_contract:
+        return []
+    current_prefix = route["path"].strip("/")
+    previous_prefix = previous_contract["path"].strip("/")
+    if current_prefix:
+        marker = f"{current_prefix}/"
+        if previous_prefix == current_prefix:
+            previous_prefix = ""
+        elif not previous_prefix.startswith(marker):
+            return []
+        else:
+            previous_prefix = previous_prefix[len(marker):]
+    filters = []
+    current_immutable = {
+        rule["path"] for rule in artifact["cache_rules"] if _is_immutable(rule)
+    }
+    for immutable_path in previous_contract["immutable_paths"]:
+        retired_path = "/".join(part for part in (previous_prefix, immutable_path) if part)
+        collision = html_dir / retired_path
+        if (immutable_path not in current_immutable and 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_contract=None):
     endpoint = artifact["s3_endpoint"]
     html_dir = site_dir / artifact["build_dir"]
     aws_env = publication_aws_env(artifact, credential_env_names)
@@ -204,12 +236,13 @@ def s3_sync(artifact, route, site_dir, credential_env_names=None):
     # 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_contract)
     run(["aws", "--endpoint-url", endpoint, "s3", "cp", f"{html_dir}/", destination,
          "--recursive", "--only-show-errors", "--cache-control", default_cache,
-         *default_filters, *exclude_args], env=aws_env)
+         *default_filters, *retired_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, *exclude_args], env=aws_env)
+         *default_filters, *retired_filters, *exclude_args], env=aws_env)
     for rule in artifact["cache_rules"]:
         if not rule["path"]:
             continue
@@ -284,7 +317,13 @@ def render_site_manifests(site_name, action_dir, app_dir, manifests_dir, cfg):
     for route in cfg["routes"]:
         artifact = artifact_by_name[route["artifact"]]
         resource_name = k8s_name(site_name) if cfg["compatibility"] else f"{k8s_name(site_name)}-{route['name']}"
-        routes.append({**route, "resource_name": resource_name, "artifact_config": artifact})
+        immutable_paths = [
+            rule["path"] for rule in artifact["cache_rules"] if _is_immutable(rule)
+        ]
+        routes.append({
+            **route, "resource_name": resource_name, "artifact_config": artifact,
+            "immutable_paths_json": json.dumps(immutable_paths, separators=(",", ":")),
+        })
     template_vars = {
         "site": site_name,
         "site_k8s": k8s_name(site_name),
@@ -297,6 +336,55 @@ 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_contracts(app_dir):
+    """Read bucket-keyed route history from generated Ingresses."""
+    contracts = {}
+    manifests = app_dir / "manifests"
+    if not manifests.exists():
+        return contracts
+    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")
+        access = annotations.get("site-publish.fritzlab.net/access")
+        bucket = annotations.get("site-publish.fritzlab.net/bucket")
+        immutable_paths_json = annotations.get("site-publish.fritzlab.net/immutable-paths")
+        route_path = annotations.get("site-publish.fritzlab.net/route-path")
+        values = (artifact, access, bucket, immutable_paths_json, route_path)
+        if all(value is None for value in values):
+            continue
+        if (not all(isinstance(value, str) for value in values)
+                or access not in {"public", "protected"} or not route_path.startswith("/")):
+            raise RuntimeError(f"invalid site-publish route history in {path}")
+        try:
+            immutable_paths = json.loads(immutable_paths_json)
+        except json.JSONDecodeError as exc:
+            raise RuntimeError(f"invalid site-publish route history in {path}") from exc
+        if not isinstance(immutable_paths, list) or any(not isinstance(item, str) for item in immutable_paths):
+            raise RuntimeError(f"invalid site-publish route history in {path}")
+        if bucket in contracts:
+            raise RuntimeError(f"duplicate site-publish route history for bucket {bucket}")
+        contracts[bucket] = {
+            "path": route_path, "access": access, "artifact": artifact,
+            "immutable_paths": immutable_paths,
+        }
+    return contracts
+
+
+def validate_route_migrations(cfg, previous_contracts):
+    artifacts = {artifact["name"]: artifact for artifact in cfg["artifacts"]}
+    for route in cfg["routes"]:
+        artifact = artifacts[route["artifact"]]
+        previous = previous_contracts.get(artifact["bucket"])
+        if (previous and previous["access"] == "protected"
+                and route["access"] in {"public", "legacy"}
+        ):
+            raise RuntimeError(
+                f"artifact {route['artifact']} cannot become public while reusing protected "
+                f"bucket {artifact['bucket']}"
+            )
+
+
 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 = {
@@ -305,6 +393,11 @@ 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_contracts = previous_route_contracts(app_dir)
+    validate_route_migrations(cfg, previous_contracts)
     # 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.
@@ -313,14 +406,13 @@ 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)
+        s3_sync(
+            artifact_by_name[route["artifact"]], route, site_dir, credential_env_names,
+            previous_contracts.get(artifact_by_name[route["artifact"]]["bucket"]),
+        )
     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)

scripts/utils.py

@@ -432,7 +432,15 @@ def validate_artifact_inputs(site_dir, cfg):
     root = Path(site_dir).resolve()
     sources = []
     for artifact in cfg["artifacts"]:
-        source = (root / artifact["content_dir"]).resolve()
+        declared = root
+        for component in Path(artifact["content_dir"]).parts:
+            declared /= component
+            if declared.is_symlink():
+                raise ConfigError(
+                    f"artifact {artifact['name']} content_dir contains symlink component: "
+                    f"{declared.relative_to(root)}"
+                )
+        source = declared.resolve()
         if source != root and root not in source.parents:
             raise ConfigError(
                 f"artifact {artifact['name']} content_dir resolves outside the repository"