Merge pull request 'feat(site-publish): reconcile split-surface CORS' (#6) from feat/split-surface-cors into main

README.md

@@ -52,6 +52,7 @@ artifacts:
   - name: distributions
     type: static
     content_dir: dist
+    cors_origins: ['*']
     publish:
       bucket: baseline-dist
       credentials:
@@ -151,6 +152,12 @@ protected input from entering a public artifact through dereference. Split
 storage endpoints are pinned to Garage, and each website
 authority is derived from its bucket; a site cannot expose an arbitrary backend.
 
+`cors_origins` is reconciled as a bucket policy on every split publication. Values are either `*`
+or HTTPS origins; browser access is limited to `GET` and `HEAD`. Omitting the field removes stale
+CORS from that bucket. Protected artifacts cannot allow wildcard CORS. All immutable objects and
+all bucket CORS policies complete before mutable channels change; if any policy write fails, the
+policies already attempted are restored to their pre-publication values.
+
 Each split route gets a bucket-specific `<bucket>.web.sjc001.fritzlab.net`
 ExternalName Service annotated to disable pass-host-header and a separate Ingress. Route
 Ingresses share the hostname's certificate Secret. The access middleware and

scripts/deploy.py

@@ -8,6 +8,7 @@ import os
 import re
 import shutil
 import subprocess
+import tempfile
 from pathlib import Path
 from urllib.error import HTTPError, URLError
 from urllib.request import Request, urlopen
@@ -169,6 +170,98 @@ def publication_aws_env(artifact, credential_env_names=None):
     return aws_env
 
 
+def configure_cors(bucket, origins, endpoint, aws_env):
+    """Reconcile read-only browser access without exposing publication credentials."""
+    if origins is None:
+        return
+    config = None
+    if origins:
+        config = {
+            "CORSRules": [{
+                "AllowedOrigins": origins,
+                "AllowedMethods": ["GET", "HEAD"],
+                "AllowedHeaders": ["*"],
+                "ExposeHeaders": ["ETag"],
+                "MaxAgeSeconds": 3600,
+            }],
+        }
+    set_cors_configuration(bucket, config, endpoint, aws_env)
+
+
+def set_cors_configuration(bucket, config, endpoint, aws_env):
+    """Apply an exact bucket CORS configuration, or remove it when absent."""
+    if config is None:
+        run([
+            "aws", "--endpoint-url", endpoint, "s3api", "delete-bucket-cors",
+            "--bucket", bucket,
+        ], env=aws_env)
+        return
+    with tempfile.NamedTemporaryFile("w", suffix=".json", encoding="utf-8") as handle:
+        json.dump(config, handle)
+        handle.flush()
+        run([
+            "aws", "--endpoint-url", endpoint, "s3api", "put-bucket-cors",
+            "--bucket", bucket, "--cors-configuration", f"file://{handle.name}",
+        ], env=aws_env)
+
+
+def get_cors_configuration(bucket, endpoint, aws_env):
+    """Read the exact bucket CORS configuration for rollback."""
+    result = _aws_capture([
+        "aws", "--endpoint-url", endpoint, "s3api", "get-bucket-cors",
+        "--bucket", bucket, "--output", "json",
+    ], aws_env)
+    if result.returncode == 0:
+        try:
+            config = json.loads(result.stdout)
+        except json.JSONDecodeError as error:
+            raise RuntimeError(f"get-bucket-cors returned invalid JSON for {bucket}") from error
+        if not isinstance(config, dict) or not isinstance(config.get("CORSRules"), list):
+            raise RuntimeError(f"get-bucket-cors returned an invalid policy for {bucket}")
+        return config
+    error = f"{result.stdout}\n{result.stderr}"
+    if "NoSuchCORSConfiguration" in error:
+        return None
+    raise RuntimeError(f"get-bucket-cors failed for {bucket}: {error.strip()}")
+
+
+def reconcile_artifact_cors(artifacts, credential_env_names=None):
+    """Reconcile all policies, restoring the prior set if any write fails."""
+    snapshots = []
+    for artifact in artifacts:
+        if artifact["cors_origins"] is None:
+            continue
+        aws_env = publication_aws_env(artifact, credential_env_names)
+        snapshots.append((
+            artifact,
+            aws_env,
+            get_cors_configuration(artifact["bucket"], artifact["s3_endpoint"], aws_env),
+        ))
+
+    attempted = []
+    try:
+        for artifact, aws_env, previous in snapshots:
+            attempted.append((artifact, aws_env, previous))
+            configure_cors(
+                artifact["bucket"], artifact["cors_origins"], artifact["s3_endpoint"], aws_env,
+            )
+    except Exception as error:
+        rollback_errors = []
+        for artifact, aws_env, previous in reversed(attempted):
+            try:
+                set_cors_configuration(
+                    artifact["bucket"], previous, artifact["s3_endpoint"], aws_env,
+                )
+            except Exception as rollback_error:
+                rollback_errors.append(f"{artifact['bucket']}: {rollback_error}")
+        if rollback_errors:
+            raise RuntimeError(
+                f"CORS reconciliation failed ({error}); rollback also failed for "
+                f"{'; '.join(rollback_errors)}"
+            ) from error
+        raise
+
+
 def publish_route_immutables(artifact, route, site_dir, credential_env_names=None):
     """Publish one route's immutable partitions during the global preflight."""
     html_dir = site_dir / artifact["build_dir"]
@@ -460,6 +553,10 @@ def deploy_static(site_name, site_dir, action_dir, token, cfg):
         publish_route_immutables(
             artifact_by_name[route["artifact"]], route, site_dir, credential_env_names,
         )
+    # Reconcile every browser-read policy before publishing mutable content.
+    # A CORS failure therefore cannot leave a new channel pointing at a release
+    # whose cross-origin assets browsers cannot consume.
+    reconcile_artifact_cors(cfg["artifacts"], credential_env_names)
     for route in cfg["routes"]:
         s3_sync(
             artifact_by_name[route["artifact"]], route, site_dir, credential_env_names,

scripts/utils.py

@@ -1,5 +1,6 @@
 """Shared utilities for the site-publish action."""
 
+import ipaddress
 import os
 import re
 import shutil
@@ -104,6 +105,55 @@ def _strings(value, label):
     return values
 
 
+def _cors_origins(value, label):
+    origins = _list(value, label)
+    if any(not isinstance(item, str) or not item for item in origins):
+        raise ConfigError(f"{label} must be a list of non-empty strings")
+    canonical = []
+    for origin in origins:
+        if origin == "*":
+            canonical.append(origin)
+            continue
+        parsed = urlparse(origin)
+        try:
+            port = parsed.port
+        except ValueError:
+            raise ConfigError(f"{label} must contain '*' or canonical HTTPS origins") from None
+        if parsed.scheme != "https" or not parsed.hostname or parsed.path or parsed.params or (
+            parsed.query or parsed.fragment or parsed.username or parsed.password
+        ):
+            raise ConfigError(f"{label} must contain '*' or canonical HTTPS origins")
+        try:
+            address = ipaddress.ip_address(parsed.hostname)
+        except ValueError:
+            try:
+                hostname = parsed.hostname.encode("idna").decode("ascii")
+            except UnicodeError:
+                raise ConfigError(
+                    f"{label} must contain '*' or canonical HTTPS origins"
+                ) from None
+            try:
+                _hostname(hostname, f"{label} hostname")
+            except ConfigError:
+                raise ConfigError(
+                    f"{label} must contain '*' or canonical HTTPS origins"
+                ) from None
+        else:
+            if getattr(address, "scope_id", None) is not None:
+                raise ConfigError(
+                    f"{label} must contain '*' or canonical HTTPS origins"
+                )
+            hostname = f"[{address.compressed}]" if address.version == 6 else address.compressed
+        canonical.append(f"https://{hostname}{f':{port}' if port not in (None, 443) else ''}")
+    if len(canonical) != len(set(canonical)):
+        raise ConfigError(f"{label} must not contain duplicate canonical origins")
+    if "*" in canonical and len(canonical) != 1:
+        raise ConfigError(f"{label} wildcard must be the only origin")
+    if origins != canonical:
+        raise ConfigError(f"{label} must contain '*' or canonical HTTPS origins")
+    return origins
+
+
 def _hostname(value, label):
     if not isinstance(value, str) or len(value) > 253 or value.endswith("."):
         raise ConfigError(f"{label} must be a lowercase DNS hostname without a trailing dot")
@@ -215,6 +265,7 @@ def _legacy_config(raw, site_name):
         "website_authority": "garage-s3.storage.svc.k8s.sjc001.fritzlab.net",
         "credentials": {"access_key_env": "AWS_ACCESS_KEY_ID", "secret_key_env": "AWS_SECRET_ACCESS_KEY"},
         "cache_rules": [{"path": "", "cache_control": DEFAULT_CACHE_CONTROL}],
+        "cors_origins": None,
     }
     return {
         "version": 1,
@@ -234,7 +285,11 @@ def _legacy_config(raw, site_name):
 def _artifact(item, index):
     label = f"artifacts[{index}]"
     item = _mapping(item, label)
-    _known_keys(item, {"name", "type", "content_dir", "tidy", "excludes", "publish", "cache"}, label)
+    _known_keys(
+        item,
+        {"name", "type", "content_dir", "tidy", "excludes", "publish", "cache", "cors_origins"},
+        label,
+    )
     name = item.get("name")
     if not isinstance(name, str) or not NAME_RE.fullmatch(name):
         raise ConfigError(f"{label}.name must be a DNS label")
@@ -303,6 +358,10 @@ def _artifact(item, index):
         "website_authority": authority,
         "credentials": normalized_credentials,
         "cache_rules": cache_rules,
+        "cors_origins": (
+            _cors_origins(item["cors_origins"], f"{label}.cors_origins")
+            if "cors_origins" in item else []
+        ),
     }
 
 
@@ -389,6 +448,10 @@ def _validate_multi(cfg):
                 raise ConfigError(f"protected route {route['name']} cannot use shared-cache max-age")
             if route["access"] == "public" and "private" in directives:
                 raise ConfigError(f"public route {route['name']} cannot use private cache policy")
+        if route["access"] == "protected" and "*" in artifact["cors_origins"]:
+            raise ConfigError(
+                f"protected route {route['name']} cannot allow wildcard CORS"
+            )
     root = next(route for route in routes if route["path"] == "/")
     if any(route["access"] == "protected" for route in routes) and root["access"] == "public":
         raise ConfigError("a public '/' catch-all would expose unmatched protected content")

tests/fixtures/split-site.yaml

@@ -15,6 +15,7 @@ artifacts:
   - name: distributions
     type: static
     content_dir: dist
+    cors_origins: ['*']
     publish:
       bucket: baseline-dist
       credentials:

tests/test_contract.py

@@ -113,6 +113,69 @@ class ConfigContractTests(unittest.TestCase):
         self.assertEqual(["/dist", "/"], [route["path"] for route in cfg["routes"]])
         credentials = {artifact["name"]: artifact["credentials"] for artifact in cfg["artifacts"]}
         self.assertNotEqual(credentials["distributions"], credentials["portal"])
+        distributions = next(
+            artifact for artifact in cfg["artifacts"] if artifact["name"] == "distributions"
+        )
+        self.assertEqual(["*"], distributions["cors_origins"])
+
+    def test_cors_origins_are_https_origins_or_wildcard(self):
+        self.assert_invalid(
+            lambda raw: raw["artifacts"][1].__setitem__(
+                "cors_origins", ["http://consumer.example"]
+            ),
+            "must contain '\\*' or canonical HTTPS origins",
+        )
+        for origin in (
+            "https://consumer example",
+            "https://Consumer.example",
+            "https://consumer.example:443",
+            "https://consumer.example/",
+            "https://consumer.example:invalid",
+            "https://[fe80::1%eth0]",
+            "https://[fe80::1%25eth0]",
+        ):
+            with self.subTest(origin=origin):
+                self.assert_invalid(
+                    lambda raw, origin=origin: raw["artifacts"][1].__setitem__(
+                        "cors_origins", [origin]
+                    ),
+                    "must contain '\\*' or canonical HTTPS origins",
+                )
+        self.assert_invalid(
+            lambda raw: raw["artifacts"][1].__setitem__(
+                "cors_origins", ["https://consumer.example", "https://consumer.example:443"]
+            ),
+            "must not contain duplicate canonical origins",
+        )
+        self.assert_invalid(
+            lambda raw: raw["artifacts"][1].__setitem__(
+                "cors_origins", ["*", "https://consumer.example"]
+            ),
+            "wildcard must be the only origin",
+        )
+        raw = copy.deepcopy(self.raw)
+        raw["artifacts"][1]["cors_origins"] = ["https://[2602:817:3000::1]:8443"]
+        cfg = normalize_site_config(raw, "baseline.fritzlab.net")
+        self.assertEqual(
+            ["https://[2602:817:3000::1]:8443"],
+            next(a for a in cfg["artifacts"] if a["name"] == "distributions")["cors_origins"],
+        )
+
+    def test_declared_cors_origins_must_be_a_list(self):
+        for value in (None, False, 0, "", {}):
+            with self.subTest(value=value):
+                self.assert_invalid(
+                    lambda raw, value=value: raw["artifacts"][1].__setitem__(
+                        "cors_origins", value
+                    ),
+                    "cors_origins must be a list",
+                )
+
+    def test_protected_route_rejects_wildcard_cors(self):
+        self.assert_invalid(
+            lambda raw: raw["artifacts"][0].__setitem__("cors_origins", ["*"]),
+            "protected route portal cannot allow wildcard CORS",
+        )
 
     def test_equivalent_route_paths_are_ambiguous(self):
         self.assert_invalid(lambda raw: raw["routes"].append({
@@ -131,6 +194,7 @@ class ConfigContractTests(unittest.TestCase):
             raw["routes"][1]["access"] = {
                 "mode": "protected", "middleware": "authentik-forwardauth"
             }
+            raw["artifacts"][1]["cors_origins"] = []
             raw["artifacts"][0]["cache"]["rules"][0]["cache_control"] = (
                 "public, max-age=0, must-revalidate"
             )
@@ -341,6 +405,75 @@ class BuildTests(unittest.TestCase):
 
 
 class PublishingTests(unittest.TestCase):
+    def test_cors_policy_is_reconciled_as_read_only_browser_access(self):
+        captured = {}
+
+        def capture(command, **_kwargs):
+            config_path = command[command.index("--cors-configuration") + 1]
+            captured["command"] = command
+            captured["config"] = json.loads(Path(config_path.removeprefix("file://")).read_text())
+
+        with patch.object(deploy, "run", side_effect=capture):
+            deploy.configure_cors(
+                "baseline-dist", ["*"], "http://garage-s3.storage.svc:3900", {}
+            )
+
+        self.assertIn("put-bucket-cors", captured["command"])
+        self.assertEqual(["GET", "HEAD"], captured["config"]["CORSRules"][0]["AllowedMethods"])
+        self.assertEqual(["*"], captured["config"]["CORSRules"][0]["AllowedOrigins"])
+
+    def test_empty_cors_policy_removes_stale_bucket_cors(self):
+        with patch.object(deploy, "run") as request:
+            deploy.configure_cors(
+                "baseline-catalogue", [], "http://garage-s3.storage.svc:3900", {}
+            )
+        self.assertIn("delete-bucket-cors", request.call_args.args[0])
+
+    def test_cors_reconciliation_restores_prior_policies_on_failure(self):
+        artifacts = [
+            {
+                "bucket": "first", "cors_origins": ["https://new.example"],
+                "s3_endpoint": "http://garage-s3.storage.svc:3900", "credentials": {},
+            },
+            {
+                "bucket": "second", "cors_origins": [],
+                "s3_endpoint": "http://garage-s3.storage.svc:3900", "credentials": {},
+            },
+        ]
+        prior = [
+            {"CORSRules": [{"AllowedOrigins": ["https://old.example"]}]},
+            None,
+        ]
+        writes = []
+
+        def write(bucket, config, *_args):
+            writes.append((bucket, config))
+            if bucket == "second" and len(writes) == 2:
+                raise RuntimeError("write failed")
+
+        with patch.object(deploy, "publication_aws_env", return_value={}), patch.object(
+            deploy, "get_cors_configuration", side_effect=prior,
+        ) as read, patch.object(deploy, "set_cors_configuration", side_effect=write), \
+                self.assertRaisesRegex(RuntimeError, "write failed"):
+            deploy.reconcile_artifact_cors(artifacts)
+
+        self.assertEqual(2, read.call_count)
+        self.assertEqual(
+            [
+                ("first", {"CORSRules": [{
+                    "AllowedOrigins": ["https://new.example"],
+                    "AllowedMethods": ["GET", "HEAD"],
+                    "AllowedHeaders": ["*"],
+                    "ExposeHeaders": ["ETag"],
+                    "MaxAgeSeconds": 3600,
+                }]}),
+                ("second", None),
+                ("second", None),
+                ("first", prior[0]),
+            ],
+            writes,
+        )
+
     def test_apps_clone_never_places_token_in_argv_or_log(self):
         calls = []
         secret = "clone-secret-must-not-appear"
@@ -645,15 +778,42 @@ class PublishingTests(unittest.TestCase):
                     patch.object(deploy, "clone_apps", return_value=root / "apps"), patch.object(
                         deploy, "publish_route_immutables",
                         side_effect=[None, RuntimeError("immutable failed")],
-                    ) as immutable_publish, patch.object(deploy, "s3_sync") as mutable_sync, \
+                    ) as immutable_publish, patch.object(
+                        deploy, "reconcile_artifact_cors"
+                    ) as cors_reconcile, patch.object(deploy, "s3_sync") as mutable_sync, \
                     self.assertRaisesRegex(
                 RuntimeError, "immutable failed"
             ):
                 deploy.deploy_static("baseline", root, root, "token", cfg)
 
             self.assertEqual(2, immutable_publish.call_count)
+            cors_reconcile.assert_not_called()
             mutable_sync.assert_not_called()
 
+    def test_all_cors_policies_complete_before_mutable_publication(self):
+        cfg = normalize_site_config(fixture("split-site.yaml"), "baseline.fritzlab.net")
+        events = []
+        with tempfile.TemporaryDirectory() as tmp:
+            root = Path(tmp)
+            apps = root / "apps"
+            apps.mkdir()
+            with patch.object(deploy, "validate_publication_environment"), patch.object(
+                deploy, "validate_artifact_output"
+            ), patch.object(
+                deploy, "publish_route_immutables", side_effect=lambda *_args: events.append("immutable")
+            ), patch.object(
+                deploy, "reconcile_artifact_cors", side_effect=lambda *_args: events.append("cors")
+            ), patch.object(
+                deploy, "s3_sync", side_effect=lambda *_args: events.append("mutable")
+            ), patch.object(deploy, "clone_apps", return_value=apps), patch.object(
+                deploy, "render_site_manifests"
+            ), patch.object(deploy, "commit_and_push"):
+                deploy.deploy_static("baseline", root, root, "token", cfg)
+
+        self.assertEqual(
+            ["immutable", "immutable", "cors", "mutable", "mutable"], events
+        )
+
 
     def test_absent_artifact_is_detected_before_publish(self):
         cfg = normalize_site_config(fixture("split-site.yaml"), "baseline.fritzlab.net")