@gscdump/lakehouse 7.11.0 → 7.11.1

package/README.md

@@ -206,6 +206,8 @@ The caller owns the target and excludes concurrent writers through settlement.
 
 Supply the saved metadata location, SHA-256, table UUID, head snapshot ID, and complete snapshot ID inventory.
 The operation checks the target, saved metadata bytes, and identity before registration.
+Metadata can use plain JSON or gzip. The byte limit applies to stored and expanded bytes.
+The SHA-256 always authenticates the stored object bytes before decompression.
 It registers with `overwrite: false` and verifies the returned identity.
 Existing tables produce `target-exists`, including tables with matching saved identities.
 Malformed bytes or mismatched identities produce `backup-invalid` before registration.

package/dist/table-restore.d.mts

@@ -10,7 +10,7 @@ export interface RestoreIcebergTableOptions {
   snapshotIds: readonly string[];
   /** The caller excludes writers and owns the target through settlement. */
   confirmAuthority: () => Promise<boolean>;
-  /** Defaults to 16 MiB. The maximum is 64 MiB. */
+  /** Bounds stored and expanded metadata. Defaults to 16 MiB. The maximum is 64 MiB. */
   maxMetadataBytes?: number;
   cache?: CatalogCache;
 }

package/dist/table-restore.mjs

@@ -16,6 +16,35 @@ function matchesBackup(loaded, options) {
 	const head = rawHead == null || String(rawHead) === "-1" ? null : String(rawHead);
 	return loaded.metadataLocation === options.metadataLocation && metadata["table-uuid"] === options.tableUuid && head === options.headSnapshotId && ids.length === expected.size && new Set(ids).size === expected.size && ids.every((id) => expected.has(id));
 }
+async function expandMetadata(bytes, maxBytes) {
+	if (bytes[0] !== 31 || bytes[1] !== 139) return bytes;
+	const reader = new Blob([bytes]).stream().pipeThrough(new DecompressionStream("gzip")).getReader();
+	try {
+		const chunks = [];
+		let total = 0;
+		for (;;) {
+			const chunk = await reader.read();
+			if (chunk.done) break;
+			total += chunk.value.byteLength;
+			if (total > maxBytes) {
+				await reader.cancel();
+				return null;
+			}
+			chunks.push(chunk.value);
+		}
+		const expanded = new Uint8Array(total);
+		let offset = 0;
+		for (const chunk of chunks) {
+			expanded.set(chunk, offset);
+			offset += chunk.byteLength;
+		}
+		return expanded;
+	} catch {
+		return null;
+	} finally {
+		reader.releaseLock();
+	}
+}
 async function restoreIcebergTable(conn, table, options) {
 	if (!validOptions(table, options)) return {
 		_tag: "Err",
@@ -51,9 +80,14 @@ async function restoreIcebergTable(conn, table, options) {
 		_tag: "Err",
 		reason: "backup-invalid"
 	};
+	const expanded = await expandMetadata(bytes, maxBytes);
+	if (!expanded) return {
+		_tag: "Err",
+		reason: "backup-invalid"
+	};
 	let metadata;
 	try {
-		metadata = parseIcebergJson(new TextDecoder("utf-8", { fatal: true }).decode(bytes));
+		metadata = parseIcebergJson(new TextDecoder("utf-8", { fatal: true }).decode(expanded));
 	} catch {
 		return {
 			_tag: "Err",

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@gscdump/lakehouse",
   "type": "module",
-  "version": "7.11.0",
+  "version": "7.11.1",
   "description": "Dataset-agnostic Iceberg lakehouse layer + dataset registry for R2 Data Catalog producers (ADR-0021).",
   "author": {
     "name": "Harlan Wilton",