Author SHA1 Message Date
Evelyn Chen 069d1baaba feat(site-publish): add split-surface publishing
Test / contract (pull_request) Successful in 6s
Authored-By: OpenAI (GPT-5) <noreply@openai.com>
2026-08-29 21:59:13 +00:00
6 changed files with 36 additions and 808 deletions
+2 -25
View File
@@ -52,7 +52,6 @@ artifacts:
- name: distributions
type: static
content_dir: dist
cors_origins: ['*']
publish:
bucket: baseline-dist
credentials:
@@ -129,35 +128,13 @@ cache policy, content type, and bytes. That content address makes concurrent
writes identical even though Garage v2.2.0 has no conditional destination
write. An identical retry converges; a changed object, missing digest metadata,
wrong address, or nested policy under that immutable prefix fails publication.
Every immutable target across every artifact is validated and published before
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 and `site-publish-history.yaml` retain every
seen bucket's access, prior route, and cumulative bucket-relative immutable key
prefixes, including while an artifact is absent. Removed or renamed rules stay
recorded. When a move places a retired prefix inside the new sync scope, its
immutable subtrees are excluded; a current-file collision fails publication.
The bucket-keyed history rejects a protected-to-public transition even when the
artifact is renamed; publishing that artifact publicly requires a new bucket.
Legacy single-surface is public for this downgrade check. Removing or renaming
an immutable rule preserves its prior URLs; current mutable content at one of
those paths is rejected instead of replacing it.
Artifact input directories must be pairwise disjoint after filesystem
resolution. Publication stops before build or upload if one contains another or
escapes the repository. Symlinked roots, components, and descendants are also rejected, preventing
protected input from entering a public artifact through dereference. Split
storage endpoints are pinned to Garage, and each website
escapes the repository, preventing protected input from entering a public
artifact. 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
+30 -293
View File
@@ -8,13 +8,10 @@ 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
import yaml
from utils import (
NAMESPACE,
clone_apps,
@@ -152,8 +149,9 @@ def publish_immutable_rule(artifact, route, rule, html_dir, aws_env):
)
def publication_aws_env(artifact, credential_env_names=None):
"""Build the route-scoped AWS environment without leaking other credentials."""
def s3_sync(artifact, route, site_dir, credential_env_names=None):
endpoint = artifact["s3_endpoint"]
html_dir = site_dir / artifact["build_dir"]
access_key = env(artifact["credentials"]["access_key_env"])
secret_key = env(artifact["credentials"]["secret_key_env"])
aws_env = os.environ.copy()
@@ -167,155 +165,6 @@ def publication_aws_env(artifact, credential_env_names=None):
"AWS_SECRET_ACCESS_KEY": secret_key,
"AWS_DEFAULT_REGION": os.environ.get("AWS_DEFAULT_REGION", "sjc001"),
})
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"]
aws_env = publication_aws_env(artifact, credential_env_names)
for rule in artifact["cache_rules"]:
if _is_immutable(rule):
publish_immutable_rule(artifact, route, rule, html_dir, aws_env)
def immutable_key_prefixes(artifact, route):
"""Return immutable partitions as bucket-relative key prefixes."""
route_prefix = route["path"].strip("/")
return [
"/".join(part for part in (route_prefix, rule["path"]) if part)
for rule in artifact["cache_rules"] if _is_immutable(rule)
]
def retained_immutable_paths(artifact, route, previous_contract):
"""Carry all bucket history forward so later route moves cannot delete it."""
previous_paths = previous_contract["immutable_paths"] if previous_contract else []
return sorted(set(previous_paths) | set(immutable_key_prefixes(artifact, route)))
def retired_immutable_filters(artifact, route, html_dir, previous_contract):
"""Protect historical immutable keys that fall inside the current sync scope."""
if not previous_contract:
return []
current_prefix = route["path"].strip("/")
filters = []
current_immutable = set(immutable_key_prefixes(artifact, route))
for immutable_path in previous_contract["immutable_paths"]:
if immutable_path in current_immutable:
continue
if current_prefix:
marker = f"{current_prefix}/"
if not immutable_path.startswith(marker):
continue
retired_path = immutable_path[len(marker):]
else:
retired_path = immutable_path
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_contract=None):
endpoint = artifact["s3_endpoint"]
html_dir = site_dir / artifact["build_dir"]
aws_env = publication_aws_env(artifact, credential_env_names)
bucket = artifact["bucket"]
object_prefix = route["path"].strip("/")
destination = f"s3://{bucket}/{object_prefix + '/' if object_prefix else ''}"
@@ -327,26 +176,33 @@ def s3_sync(artifact, route, site_dir, credential_env_names=None, previous_contr
exclude_args = [arg for pattern in artifact["excludes"] for arg in ("--exclude", pattern)]
if artifact["excludes"]:
print(f"Excluding patterns: {artifact['excludes']}")
# Validate and publish every append-only target before a mutable channel can
# point at it. Partial immutable success is safe; partial mutable success is
# not.
for rule in artifact["cache_rules"]:
if _is_immutable(rule):
publish_immutable_rule(artifact, route, rule, html_dir, aws_env)
print(f"Syncing artifact {artifact['name']}{destination} via {endpoint}")
# Upload with the final cache policy before cleanup. Sync and deletion are
# scoped to the same current route prefix and cache partition. A route move
# Sync and deletion are scoped to the current route prefix. A route move
# leaves its old bucket partition intact but unreachable after the old
# Ingress disappears, while stale mutable keys on the serving prefix are
# deleted. Immutable subtrees are structurally excluded. `cp --recursive`
# refreshes metadata atomically per object before `sync --delete` removes
# stale keys without ever exposing new bytes under a provisional policy.
# deleted. Immutable subtrees are structurally excluded. Partitioned
# `cp --recursive` calls then re-upload each file once to refresh metadata
# (cache-control, content-type) on objects sync skipped as unchanged.
# A no-op deploy therefore transfers the artifact bytes once.
# AWS CLI guesses Content-Type from file extension on local→S3 uploads,
# 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, *retired_filters, *exclude_args], env=aws_env)
sync_excludes = [*exclude_args,
*(arg for path in immutable_paths for arg in ("--exclude", f"{path}/*"))]
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)
*sync_excludes], env=aws_env)
print("Re-stamping metadata on all objects...")
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}/*")]
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)
for rule in artifact["cache_rules"]:
if not rule["path"]:
continue
@@ -361,9 +217,6 @@ def s3_sync(artifact, route, site_dir, credential_env_names=None, previous_contr
run(["aws", "--endpoint-url", endpoint, "s3", "cp", f"{html_dir}/", destination,
"--recursive", "--only-show-errors", "--cache-control", rule["cache_control"],
"--exclude", "*", "--include", include, *child_filters, *exclude_args], env=aws_env)
run(["aws", "--endpoint-url", endpoint, "s3", "sync", f"{html_dir}/", destination,
"--delete", "--only-show-errors", "--cache-control", rule["cache_control"],
"--exclude", "*", "--include", include, *child_filters, *exclude_args], env=aws_env)
def garage_admin(method, path, token, body=None):
@@ -412,32 +265,16 @@ def ensure_bucket_aliases(site_name, aliases, admin_token):
raise
def render_site_manifests(
site_name, action_dir, app_dir, manifests_dir, cfg, previous_contracts=None,
):
def render_site_manifests(site_name, action_dir, app_dir, manifests_dir, cfg):
"""Always re-render manifests from current site.yaml. Templates own
domain + aliases, so changes propagate without manual edits."""
manifests_dir.mkdir(parents=True, exist_ok=True)
artifact_by_name = {artifact["name"]: artifact for artifact in cfg["artifacts"]}
routes = []
previous_contracts = previous_contracts or {}
next_contracts = {bucket: dict(contract)
for bucket, contract in previous_contracts.items()}
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']}"
previous = previous_contracts.get(artifact["bucket"])
immutable_paths = retained_immutable_paths(artifact, route, previous)
next_contracts[artifact["bucket"]] = {
"path": route["path"],
"access": "public" if route["access"] == "legacy" else route["access"],
"artifact": route["artifact"],
"immutable_paths": immutable_paths,
}
routes.append({
**route, "resource_name": resource_name, "artifact_config": artifact,
"immutable_paths_json": json.dumps(immutable_paths, separators=(",", ":")),
})
routes.append({**route, "resource_name": resource_name, "artifact_config": artifact})
template_vars = {
"site": site_name,
"site_k8s": k8s_name(site_name),
@@ -448,89 +285,6 @@ def render_site_manifests(
"routes": routes,
}
render_templates(action_dir, template_vars, app_dir, manifests_dir)
if not cfg["compatibility"] or previous_contracts:
(app_dir / "site-publish-history.yaml").write_text(yaml.safe_dump(
{"version": 1, "buckets": next_contracts}, sort_keys=True,
))
def _validate_route_contract(bucket, contract, path):
expected = {"path", "access", "artifact", "immutable_paths"}
if (not isinstance(bucket, str) or not bucket or not isinstance(contract, dict)
or set(contract) != expected
or contract.get("access") not in {"public", "protected"}
or not isinstance(contract.get("path"), str)
or not contract["path"].startswith("/")
or not isinstance(contract.get("artifact"), str) or not contract["artifact"]
or not isinstance(contract.get("immutable_paths"), list)
or any(not isinstance(item, str) or not item or item.startswith("/")
or any(part in {"", ".", ".."} for part in item.split("/"))
for item in contract["immutable_paths"])):
raise RuntimeError(f"invalid site-publish route history in {path}")
return {
"path": contract["path"], "access": contract["access"],
"artifact": contract["artifact"],
"immutable_paths": sorted(set(contract["immutable_paths"])),
}
def previous_route_contracts(app_dir):
"""Read bucket-keyed route history from generated Ingresses."""
history_path = app_dir / "site-publish-history.yaml"
if history_path.exists():
document = yaml.safe_load(history_path.read_text())
if (not isinstance(document, dict) or set(document) != {"version", "buckets"}
or document["version"] != 1 or not isinstance(document["buckets"], dict)):
raise RuntimeError(f"invalid site-publish route history in {history_path}")
return {
bucket: _validate_route_contract(bucket, contract, history_path)
for bucket, contract in document["buckets"].items()
}
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] = _validate_route_contract(bucket, {
"path": route_path, "access": access, "artifact": artifact,
"immutable_paths": immutable_paths,
}, path)
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):
@@ -541,33 +295,16 @@ 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.
for route in cfg["routes"]:
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,
previous_contracts.get(artifact_by_name[route["artifact"]]["bucket"]),
)
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"))
render_site_manifests(
site_name, action_dir, app_dir, manifests_dir, cfg, previous_contracts,
)
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)
+2 -82
View File
@@ -1,6 +1,5 @@
"""Shared utilities for the site-publish action."""
import ipaddress
import os
import re
import shutil
@@ -105,55 +104,6 @@ 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")
@@ -265,7 +215,6 @@ 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,
@@ -285,11 +234,7 @@ 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", "cors_origins"},
label,
)
_known_keys(item, {"name", "type", "content_dir", "tidy", "excludes", "publish", "cache"}, 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")
@@ -358,10 +303,6 @@ 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 []
),
}
@@ -448,10 +389,6 @@ 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")
@@ -490,31 +427,14 @@ def normalize_site_config(raw, site_name):
def validate_artifact_inputs(site_dir, cfg):
"""Reject source containment before a public or protected build starts."""
if cfg["compatibility"]:
return
root = Path(site_dir).resolve()
sources = []
for artifact in cfg["artifacts"]:
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()
source = (root / artifact["content_dir"]).resolve()
if source != root and root not in source.parents:
raise ConfigError(
f"artifact {artifact['name']} content_dir resolves outside the repository"
)
if source.exists():
symlink = next((path for path in source.rglob("*") if path.is_symlink()), None)
if symlink is not None:
raise ConfigError(
f"artifact {artifact['name']} build input contains symlink: "
f"{symlink.relative_to(root)}"
)
sources.append((artifact["name"], source))
for index, (name, source) in enumerate(sources):
for other_name, other_source in sources[index + 1:]:
-7
View File
@@ -4,13 +4,6 @@ metadata:
name: {{ route.resource_name }}
namespace: {{ namespace }}
annotations:
{%- if not compatibility %}
site-publish.fritzlab.net/artifact: {{ route.artifact | tojson }}
site-publish.fritzlab.net/access: {{ route.access | tojson }}
site-publish.fritzlab.net/bucket: {{ route.artifact_config.bucket | tojson }}
site-publish.fritzlab.net/immutable-paths: {{ route.immutable_paths_json | tojson }}
site-publish.fritzlab.net/route-path: {{ route.path | tojson }}
{%- endif %}
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
-1
View File
@@ -15,7 +15,6 @@ artifacts:
- name: distributions
type: static
content_dir: dist
cors_origins: ['*']
publish:
bucket: baseline-dist
credentials:
+2 -400
View File
@@ -113,69 +113,6 @@ 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({
@@ -194,7 +131,6 @@ 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"
)
@@ -283,27 +219,6 @@ class ConfigContractTests(unittest.TestCase):
with self.assertRaisesRegex(ConfigError, "build inputs overlap after resolution"):
validate_artifact_inputs(root, cfg)
def test_descendant_symlink_cannot_cross_artifact_boundary(self):
cfg = normalize_site_config(self.raw, "baseline.fritzlab.net")
with tempfile.TemporaryDirectory() as tmp:
root = Path(tmp)
(root / "dist").mkdir()
(root / "portal" / "build").mkdir(parents=True)
(root / "portal" / "build" / "private.txt").write_text("private")
(root / "dist" / "portal-link").symlink_to(root / "portal" / "build")
with self.assertRaisesRegex(ConfigError, "build input contains symlink"):
validate_artifact_inputs(root, cfg)
def test_artifact_root_symlink_is_rejected_before_resolution(self):
cfg = normalize_site_config(self.raw, "baseline.fritzlab.net")
with tempfile.TemporaryDirectory() as tmp:
root = Path(tmp)
(root / "dist-real").mkdir()
(root / "dist").symlink_to(root / "dist-real")
(root / "portal" / "build").mkdir(parents=True)
with self.assertRaisesRegex(ConfigError, "content_dir contains symlink component"):
validate_artifact_inputs(root, cfg)
class GenerationTests(unittest.TestCase):
def render(self, raw):
@@ -327,7 +242,6 @@ class GenerationTests(unittest.TestCase):
"app.yaml", "manifests/certificate.yaml", "manifests/ingress-distributions.yaml",
"manifests/ingress-portal.yaml", "manifests/kustomization.yaml",
"manifests/service-distributions.yaml", "manifests/service-portal.yaml",
"site-publish-history.yaml",
}, set(files))
for content in files.values():
self.assertIsNotNone(yaml.safe_load(content))
@@ -338,17 +252,6 @@ class GenerationTests(unittest.TestCase):
self.assertNotIn("passhostheader", files["manifests/ingress-portal.yaml"])
self.assertIn("baseline-dist.web.sjc001.fritzlab.net", files["manifests/service-distributions.yaml"])
def test_yaml_ambiguous_artifact_name_stays_a_string_annotation(self):
raw = fixture("split-site.yaml")
raw["artifacts"][0]["name"] = "yes"
raw["routes"][0]["artifact"] = "yes"
tmp, _, files = self.render(raw)
self.addCleanup(tmp.cleanup)
ingress = yaml.safe_load(files["manifests/ingress-portal.yaml"])
self.assertEqual(
"yes", ingress["metadata"]["annotations"]["site-publish.fritzlab.net/artifact"],
)
def test_generation_is_deterministic_when_input_lists_are_reversed(self):
raw = fixture("split-site.yaml")
first_tmp, _, first = self.render(raw)
@@ -372,14 +275,12 @@ class GenerationTests(unittest.TestCase):
self.assertFalse(stale.exists())
def test_legacy_names_and_garage_s3_target_are_preserved(self):
tmp, app_dir, files = self.render(fixture("legacy-site.yaml"))
tmp, _, files = self.render(fixture("legacy-site.yaml"))
self.addCleanup(tmp.cleanup)
self.assertIn("manifests/service.yaml", files)
self.assertIn("manifests/ingress.yaml", files)
self.assertIn("garage-s3.storage.svc.k8s.sjc001.fritzlab.net", files["manifests/service.yaml"])
self.assertNotIn("passhostheader", files["manifests/ingress.yaml"])
self.assertNotIn("site-publish.fritzlab.net", files["manifests/ingress.yaml"])
self.assertEqual({}, deploy.previous_route_contracts(app_dir))
class BuildTests(unittest.TestCase):
@@ -405,75 +306,6 @@ 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"
@@ -563,7 +395,6 @@ class PublishingTests(unittest.TestCase):
}, clear=False), patch.object(deploy, "run", side_effect=capture), \
patch.object(deploy, "publish_immutable_rule", side_effect=publish_immutable) as immutable_publish, \
redirect_stdout(output):
deploy.publish_route_immutables(artifact, route, root)
deploy.s3_sync(artifact, route, root)
self.assertTrue(all(secret not in " ".join(command) for command, _ in commands))
@@ -574,8 +405,7 @@ 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.assertNotIn("--delete", rendered[0])
self.assertIn("--delete", rendered[1])
self.assertIn("--delete", rendered[0])
self.assertTrue(all("s3://baseline-dist/dist/" in command for command in rendered[1:]))
self.assertTrue(any("channels/" in command and
"public, max-age=0, must-revalidate" in command
@@ -586,234 +416,6 @@ 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_contract={
"path": "/foo", "access": "public", "artifact": "distributions",
"immutable_paths": ["foo/releases"],
})
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_protected_bucket_cannot_become_public_across_deployments(self):
cfg = normalize_site_config(fixture("split-site.yaml"), "baseline.fritzlab.net")
previous = {
"baseline-dist": {
"path": "/dist", "access": "protected", "artifact": "old-name",
"immutable_paths": ["dist/releases"],
}
}
with self.assertRaisesRegex(RuntimeError, "cannot become public while reusing protected"):
deploy.validate_route_migrations(cfg, previous)
renamed = copy.deepcopy(cfg)
artifact = next(item for item in renamed["artifacts"] if item["name"] == "distributions")
artifact["name"] = "downloads"
route = next(item for item in renamed["routes"] if item["artifact"] == "distributions")
route["artifact"] = "downloads"
with self.assertRaisesRegex(RuntimeError, "cannot become public while reusing protected"):
deploy.validate_route_migrations(renamed, previous)
previous = {"retired-protected-bucket": previous["baseline-dist"]}
deploy.validate_route_migrations(cfg, previous)
def test_protected_split_bucket_cannot_become_legacy_public(self):
cfg = normalize_site_config(fixture("legacy-site.yaml"), "baseline.fritzlab.net")
previous = {
"baseline.fritzlab.net": {
"path": "/portal", "access": "protected", "artifact": "portal",
"immutable_paths": [],
}
}
with self.assertRaisesRegex(RuntimeError, "cannot become public while reusing protected"):
deploy.validate_route_migrations(cfg, previous)
def test_removed_immutable_rule_preserves_prior_keys(self):
cfg = normalize_site_config(fixture("split-site.yaml"), "baseline.fritzlab.net")
artifact = next(item for item in cfg["artifacts"] if item["name"] == "distributions")
artifact["cache_rules"] = [
rule for rule in artifact["cache_rules"] if rule["path"] != "releases"
]
route = next(item for item in cfg["routes"] if item["artifact"] == "distributions")
with tempfile.TemporaryDirectory() as tmp:
html = Path(tmp)
filters = deploy.retired_immutable_filters(artifact, route, html, {
"path": "/dist", "access": "public", "artifact": "distributions",
"immutable_paths": ["dist/releases"],
})
self.assertEqual(["--exclude", "releases/*"], filters)
(html / "releases").mkdir()
(html / "releases" / "replacement.js").write_text("mutable")
with self.assertRaisesRegex(RuntimeError, "collides with retired immutable"):
deploy.retired_immutable_filters(artifact, route, html, {
"path": "/dist", "access": "public", "artifact": "distributions",
"immutable_paths": ["dist/releases"],
})
def test_removed_immutable_rule_remains_in_next_manifest_history(self):
cfg = normalize_site_config(fixture("split-site.yaml"), "baseline.fritzlab.net")
artifact = next(item for item in cfg["artifacts"] if item["name"] == "distributions")
artifact["cache_rules"] = [
rule for rule in artifact["cache_rules"] if rule["path"] != "releases"
]
previous = {
"baseline-dist": {
"path": "/dist", "access": "public", "artifact": "distributions",
"immutable_paths": ["dist/releases"],
}
}
with tempfile.TemporaryDirectory() as tmp:
app_dir = Path(tmp) / "app"
manifests = app_dir / "manifests"
deploy.render_site_manifests(
"baseline.fritzlab.net", ROOT, app_dir, manifests, cfg, previous,
)
first = deploy.previous_route_contracts(app_dir)
self.assertEqual(["dist/releases"], first["baseline-dist"]["immutable_paths"])
deploy.render_site_manifests(
"baseline.fritzlab.net", ROOT, app_dir, manifests, cfg, first,
)
second = deploy.previous_route_contracts(app_dir)
self.assertEqual(first, second)
def test_protected_split_bucket_survives_absence_and_blocks_legacy(self):
raw = fixture("split-site.yaml")
raw["artifacts"] = [
item for item in raw["artifacts"] if item["name"] == "distributions"
]
raw["routes"] = [
item for item in raw["routes"] if item["artifact"] == "distributions"
]
raw["routes"][0]["path"] = "/"
cfg = normalize_site_config(raw, "baseline.fritzlab.net")
previous = {
"baseline.fritzlab.net": {
"path": "/", "access": "protected", "artifact": "portal",
"immutable_paths": [],
}
}
with tempfile.TemporaryDirectory() as tmp:
app_dir = Path(tmp) / "app"
deploy.render_site_manifests(
"baseline.fritzlab.net", ROOT, app_dir, app_dir / "manifests", cfg, previous,
)
retained = deploy.previous_route_contracts(app_dir)
self.assertEqual(
previous["baseline.fritzlab.net"], retained["baseline.fritzlab.net"],
)
legacy = normalize_site_config(
fixture("legacy-site.yaml"), "baseline.fritzlab.net",
)
with self.assertRaisesRegex(RuntimeError, "cannot become public while reusing protected"):
deploy.validate_route_migrations(legacy, retained)
def test_malformed_persistent_history_fails_closed(self):
with tempfile.TemporaryDirectory() as tmp:
app_dir = Path(tmp)
(app_dir / "site-publish-history.yaml").write_text(
"version: 1\nbuckets:\n bucket:\n path: /\n"
" access: public\n artifact: site\n"
" immutable_paths: [../releases]\n"
)
with self.assertRaisesRegex(RuntimeError, "invalid site-publish route history"):
deploy.previous_route_contracts(app_dir)
def test_retired_immutable_history_survives_a_route_move(self):
cfg = normalize_site_config(fixture("split-site.yaml"), "baseline.fritzlab.net")
artifact = next(item for item in cfg["artifacts"] if item["name"] == "distributions")
artifact["cache_rules"] = [
rule for rule in artifact["cache_rules"] if rule["path"] != "releases"
]
route = next(item for item in cfg["routes"] if item["artifact"] == "distributions")
route["path"] = "/"
previous = {
"baseline-dist": {
"path": "/foo", "access": "public", "artifact": "distributions",
"immutable_paths": ["foo/releases"],
}
}
with tempfile.TemporaryDirectory() as tmp:
root = Path(tmp)
app_dir = root / "app"
manifests = app_dir / "manifests"
app_dir.mkdir()
deploy.render_site_manifests(
"baseline.fritzlab.net", ROOT, app_dir, manifests, cfg, previous,
)
following = deploy.previous_route_contracts(app_dir)
self.assertEqual(["foo/releases"], following["baseline-dist"]["immutable_paths"])
html = root / "html"
html.mkdir()
first_filters = deploy.retired_immutable_filters(
artifact, route, html, previous["baseline-dist"],
)
following_filters = deploy.retired_immutable_filters(
artifact, route, html, following["baseline-dist"],
)
self.assertEqual(["--exclude", "foo/releases/*"], first_filters)
self.assertEqual(first_filters, following_filters)
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(
deploy, "publish_route_immutables",
side_effect=[None, RuntimeError("immutable failed")],
) 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")