fix: address second review round

Signed-off-by: anilb <epipav@gmail.com>

Author: epipav

services/apps/packages_worker/package.json

@@ -29,8 +29,8 @@
     "dev:npm-worker": "CROWD_TEMPORAL_TASKQUEUE=npm-worker SERVICE=npm-worker LOG_LEVEL=trace nodemon --watch src --watch ../../libs --ext ts --exec tsx --inspect=0.0.0.0:9236 src/bin/npm-worker.ts",
     "dev:npm-worker:local": "set -a && . ../../../backend/.env.dist.local && . ../../../backend/.env.override.local && set +a && CROWD_TEMPORAL_TASKQUEUE=npm-worker SERVICE=npm-worker LOG_LEVEL=trace nodemon --watch src --watch ../../libs --ext ts --exec tsx --inspect=0.0.0.0:9236 src/bin/npm-worker.ts",
     "start:pypi-worker": "CROWD_TEMPORAL_TASKQUEUE=pypi-worker SERVICE=pypi-worker tsx src/bin/pypi-worker.ts",
-    "dev:pypi-worker": "CROWD_TEMPORAL_TASKQUEUE=pypi-worker CROWD_TEMPORAL_NAMESPACE=$CROWD_PACKAGES_TEMPORAL_NAMESPACE SERVICE=pypi-worker LOG_LEVEL=trace nodemon --watch src --watch ../../libs --ext ts --exec tsx --inspect=0.0.0.0:9242 src/bin/pypi-worker.ts",
-    "dev:pypi-worker:local": "set -a && . ../../../backend/.env.dist.local && . ../../../backend/.env.override.local && set +a && CROWD_TEMPORAL_TASKQUEUE=pypi-worker SERVICE=pypi-worker LOG_LEVEL=trace nodemon --watch src --watch ../../libs --ext ts --exec tsx --inspect=0.0.0.0:9242 src/bin/pypi-worker.ts",
+    "dev:pypi-worker": "CROWD_TEMPORAL_TASKQUEUE=pypi-worker SERVICE=pypi-worker LOG_LEVEL=trace nodemon --watch src --watch ../../libs --ext ts --exec tsx --inspect=0.0.0.0:9243 src/bin/pypi-worker.ts",
+    "dev:pypi-worker:local": "set -a && . ../../../backend/.env.dist.local && . ../../../backend/.env.override.local && set +a && CROWD_TEMPORAL_TASKQUEUE=pypi-worker SERVICE=pypi-worker LOG_LEVEL=trace nodemon --watch src --watch ../../libs --ext ts --exec tsx --inspect=0.0.0.0:9243 src/bin/pypi-worker.ts",
     "start:osv-worker": "CROWD_TEMPORAL_TASKQUEUE=osv-worker SERVICE=osv-worker tsx src/bin/osv-worker.ts",
     "dev:osv-worker": "CROWD_TEMPORAL_TASKQUEUE=osv-worker SERVICE=osv-worker LOG_LEVEL=trace nodemon --watch src --watch ../../libs --ext ts --exec tsx --inspect=0.0.0.0:9238 src/bin/osv-worker.ts",
     "dev:osv-worker:local": "set -a && . ../../../backend/.env.dist.local && . ../../../backend/.env.override.local && set +a && CROWD_TEMPORAL_TASKQUEUE=osv-worker SERVICE=osv-worker LOG_LEVEL=trace nodemon --watch src --watch ../../libs --ext ts --exec tsx --inspect=0.0.0.0:9238 src/bin/osv-worker.ts",

services/apps/packages_worker/src/pypi/downloads/__tests__/pypiDownloads.test.ts

@@ -57,7 +57,7 @@ describe('computeLast30dWindows', () => {
 
 // ── Criterion 3: daily trailing window + first-of-month helper ─────────────────────────────
 describe('defaultDailyRange', () => {
-  it('defaults to a 2-day self-healing window [today-2, today-1]', () => {
+  it('defaults to a 2-day trailing re-scan window [today-2, today-1]', () => {
     expect(defaultDailyRange('2026-06-30')).toEqual({
       startDate: '2026-06-28',
       endDate: '2026-06-29',

services/apps/packages_worker/src/pypi/downloads/ingestPypiDownloads.ts

@@ -168,8 +168,8 @@ export async function ingestPypiDownloadsLast30d(opts: { fromDate?: string }): P
   }
 }
 
-// Daily downloads for the critical pypi subset. Scheduled daily with no range → the last 2-day
-// self-healing window. Pass an explicit startDate/endDate to backfill an arbitrary range.
+// Daily downloads for the critical pypi subset. Scheduled daily with no range → the 2-day
+// trailing re-scan window. Pass an explicit startDate/endDate to backfill an arbitrary range.
 export async function ingestPypiDownloadsDaily(opts: {
   startDate?: string
   endDate?: string

services/apps/packages_worker/src/pypi/downloads/pypiDownloads.ts

@@ -35,7 +35,7 @@ export async function schedulePypiDownloads30d(): Promise<void> {
 }
 
 // Daily downloads for the critical pypi subset. Runs daily at 06:00 UTC; no args → the workflow
-// scans the self-healing 2-day trailing window.
+// scans the 2-day trailing re-scan window.
 export async function schedulePypiDownloadsDaily(): Promise<void> {
   const { temporal } = svc
   if (!temporal) throw new Error('Temporal client not initialized')

services/apps/packages_worker/src/pypi/downloads/pypiDownloadsDates.ts

@@ -58,7 +58,7 @@ export function computeLast30dWindows(
 // re-scanned once on the next run and corrected, while keeping the daily BigQuery scan small.
 const DAILY_TRAILING_DAYS = 2
 
-// Default daily trailing window: [today - 2, today - 1] inclusive (self-healing).
+// Default daily trailing re-scan window: [today - 2, today - 1] inclusive.
 export function defaultDailyRange(today: string): { startDate: string; endDate: string } {
   return { startDate: addDaysUTC(today, -DAILY_TRAILING_DAYS), endDate: addDaysUTC(today, -1) }
 }

services/apps/packages_worker/src/pypi/upsertProject.ts

@@ -26,8 +26,8 @@ export async function upsertProject(
   project: PyPiProject,
   purl: string,
 ): Promise<{ purl: string; changedFields: string[] }> {
+  stripNullBytesDeep(project)
   const info = project.info
-  stripNullBytesDeep(info)
 
   const name = info.name
   const status = info.yanked ? 'yanked' : 'active'

services/libs/data-access-layer/src/packages/maintainers.ts

@@ -1,6 +1,6 @@
 import { QueryExecutor } from '../queryExecutor'
 
-export interface NpmMaintainerInput {
+export interface PackageMaintainerInput {
   username: string
   displayName: string | null
   email: string | null
@@ -10,7 +10,7 @@ export interface NpmMaintainerInput {
 export async function upsertPackageMaintainers(
   qx: QueryExecutor,
   packageId: string,
-  maintainers: NpmMaintainerInput[],
+  maintainers: PackageMaintainerInput[],
   ecosystem = 'npm',
 ): Promise<string[]> {
   const changed = new Set<string>()