fix(preprocess): harden S3 and real Compose jobs

harness/tht/adapters/evidence/s3.py

@@ -1,8 +1,7 @@
 """Bounded S3-compatible Evidence source using the supported boto3 client."""
 
 import hashlib
-import ipaddress
-import socket
+import re
 from datetime import UTC, datetime
 from urllib.parse import quote, urlsplit
 
@@ -15,40 +14,44 @@ class S3EvidenceSource:
     def __init__(self, *, bucket: str, prefix: str = "", endpoint_url: str | None = None,
                  region: str | None = None, access_key: str | None = None,
                  secret_key: str | None = None, session_token: str | None = None,
+                 trusted_endpoint: bool = False,
                  allow_private_endpoint: bool = False, allow_insecure_endpoint: bool = False,
                  max_bytes: int = 10 * 1024 * 1024, max_objects: int = 10_000,
                  max_pages: int = 100, page_size: int = 1000, client=None) -> None:
-        if not bucket or any(value < 1 for value in (max_bytes, max_objects, max_pages, page_size)):
+        if (not re.fullmatch(r"(?=.{3,63}$)(?!-)(?!.*\.\.)(?!.*\.-)(?!.*-\.)"
+                            r"[a-z0-9](?:[a-z0-9.-]*[a-z0-9])?", bucket)
+                or any(value < 1 for value in (max_bytes, max_objects, max_pages, page_size))):
             raise ValueError("S3 evidence limits and bucket must be non-empty and positive")
         if endpoint_url:
             parsed = urlsplit(endpoint_url)
             if parsed.username or parsed.password:
                 raise ValueError("S3 endpoint must not contain credentials")
-            if parsed.scheme != "https" and not allow_insecure_endpoint:
+            if parsed.scheme not in {"http", "https"}:
+                raise ValueError("S3 endpoint scheme must be exactly https or explicitly allowed http")
+            if parsed.scheme == "http" and not allow_insecure_endpoint:
                 raise ValueError("S3 endpoint must use HTTPS unless explicitly allowed")
             if not parsed.hostname:
                 raise ValueError("S3 endpoint must include a hostname")
-            if not allow_private_endpoint:
-                try:
-                    addresses = {ipaddress.ip_address(row[4][0].split("%", 1)[0]) for row in
-                                 socket.getaddrinfo(parsed.hostname, parsed.port or 443,
-                                                    type=socket.SOCK_STREAM)}
-                except (OSError, ValueError) as exc:
-                    raise ValueError("S3 endpoint resolution failed") from exc
-                if not addresses or any(not address.is_global for address in addresses):
-                    raise ValueError("S3 private endpoint requires explicit opt-in")
+            if parsed.path not in {"", "/"} or parsed.query or parsed.fragment:
+                raise ValueError("S3 custom endpoint must be an origin root without query/fragment")
+            if not trusted_endpoint:
+                raise ValueError("S3 custom endpoint requires explicit trusted_endpoint opt-in")
+            if parsed.hostname in {"localhost", "127.0.0.1", "::1"} and not allow_private_endpoint:
+                raise ValueError("S3 private endpoint requires explicit opt-in")
         self.bucket, self.prefix = bucket, prefix.lstrip("/")
         self.max_bytes, self.max_objects = max_bytes, max_objects
         self.max_pages, self.page_size = max_pages, min(page_size, 1000)
         if client is None:
             try:
                 import boto3
+                from botocore.config import Config as BotoConfig
             except ImportError as exc:  # pragma: no cover - deployment optional dependency
                 raise RuntimeError("Install tht[s3] to use S3 Evidence") from exc
             client = boto3.client("s3", endpoint_url=endpoint_url, region_name=region,
                                   aws_access_key_id=access_key,
                                   aws_secret_access_key=secret_key,
-                                  aws_session_token=session_token, verify=True)
+                                  aws_session_token=session_token, verify=True,
+                                  config=BotoConfig(s3={"addressing_style": "path"}))
         self._client = client
         self._items: dict[str, tuple[str, str | None, str | None]] = {}
 
@@ -72,12 +75,16 @@ class S3EvidenceSource:
                     count += 1
                     if count > self.max_objects:
                         raise self._error("object_limit")
-                    key, version, etag = row["Key"], row.get("VersionId"), row.get("ETag")
+                    key, etag = row.get("Key"), row.get("ETag")
+                    if (not isinstance(key, str) or not key.startswith(self.prefix)
+                            or len(key.encode()) > 1024 or any(ord(char) < 32 for char in key)):
+                        raise self._error("invalid_key")
+                    if not isinstance(etag, str) or not etag or len(etag) > 1024:
+                        raise self._error("missing_validator")
                     uri = f"s3://{self.bucket}/{quote(key, safe='/')}"
-                    stable = version or hashlib.sha256((etag or "").encode()).hexdigest()
-                    fingerprint = f"s3-version:{stable}" if version else f"etag:{stable}"
+                    fingerprint = f"etag:{hashlib.sha256(etag.encode()).hexdigest()}"
                     source_id = "s3:" + hashlib.sha256(uri.encode()).hexdigest()
-                    self._items[source_id] = (key, version, etag)
+                    self._items[source_id] = (key, None, etag)
                     modified = row.get("LastModified")
                     if modified is not None:
                         modified = modified.astimezone(UTC)

harness/tht/adapters/factory.py

@@ -127,6 +127,7 @@ def build_evidence_sources(cfg: Config):
                     endpoint_url=resource.endpoint_url, region=resource.region,
                     access_key=secret(resource.access_key), secret_key=secret(resource.secret_key),
                     session_token=secret(resource.session_token),
+                    trusted_endpoint=resource.trusted_endpoint,
                     allow_private_endpoint=resource.allow_private_endpoint,
                     allow_insecure_endpoint=resource.allow_insecure_endpoint,
                     max_bytes=resource.max_bytes, max_objects=resource.max_objects,

harness/tht/config.py

@@ -204,6 +204,7 @@ class S3EvidenceSourceConfig(BaseModel):
     access_key: SecretStr | None = None
     secret_key: SecretStr | None = None
     session_token: SecretStr | None = None
+    trusted_endpoint: bool = False
     allow_private_endpoint: bool = False
     allow_insecure_endpoint: bool = False
     max_bytes: int = Field(default=10 * 1024 * 1024, gt=0)