@gscdump/engine 1.4.6 → 1.4.7

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/dist/entities.mjs CHANGED
@@ -374,6 +374,7 @@ function createInspectionStore(opts) {
374
374
  for (const row of baseRows) consider(row);
375
375
  let eventsFolded = 0;
376
376
  const consumed = [];
377
+ const eventRows = [];
377
378
  const eventFiles = await mapEntityIo(eventKeys.sort(), async (key) => {
378
379
  const bytes = await readOptional(ds, key);
379
380
  if (!bytes) return void 0;
@@ -386,11 +387,11 @@ function createInspectionStore(opts) {
386
387
  if (!file) continue;
387
388
  const { key, rows } = file;
388
389
  consumed.push(key);
389
- for (const row of rows) {
390
- consider(row);
391
- eventsFolded++;
392
- }
390
+ eventRows.push(...rows);
391
+ eventsFolded += rows.length;
393
392
  }
393
+ eventRows.sort((a, b) => String(a.urlHash).localeCompare(String(b.urlHash)) || String(a.inspectedAt ?? "").localeCompare(String(b.inspectedAt ?? "")));
394
+ for (const row of eventRows) consider(row);
394
395
  const merged = [];
395
396
  for (const [h, row] of latest) {
396
397
  const fc = earliestChecked.get(h);
@@ -401,8 +402,8 @@ function createInspectionStore(opts) {
401
402
  columns: INSPECTION_EVENT_COLUMNS,
402
403
  sortKey: ["urlHash"]
403
404
  });
404
- await ds.write(baseKey, bytes);
405
405
  const transitionsWritten = opts?.transitions ? await appendTransitions(ds, ctx, transitions) : 0;
406
+ await ds.write(baseKey, bytes);
406
407
  if (consumed.length > 0) await ds.delete(consumed);
407
408
  return {
408
409
  baseRowCount: merged.length,
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "@gscdump/engine",
3
3
  "type": "module",
4
- "version": "1.4.6",
4
+ "version": "1.4.7",
5
5
  "description": "Append-only Parquet/DuckDB storage engine + planner + adapters for the gscdump pipeline. Node + edge runtimes; opt-in heavy peers.",
6
6
  "author": {
7
7
  "name": "Harlan Wilton",
@@ -181,9 +181,9 @@
181
181
  "dependencies": {
182
182
  "drizzle-orm": "1.0.0-rc.3",
183
183
  "proper-lockfile": "^4.1.2",
184
- "@gscdump/contracts": "^1.4.6",
185
- "gscdump": "^1.4.6",
186
- "@gscdump/lakehouse": "^1.4.6"
184
+ "@gscdump/contracts": "^1.4.7",
185
+ "@gscdump/lakehouse": "^1.4.7",
186
+ "gscdump": "^1.4.7"
187
187
  },
188
188
  "devDependencies": {
189
189
  "@duckdb/duckdb-wasm": "1.33.1-dev57.0",