# Changelog of YouTube Shorts Competitor & Viral Trend Radar (`nexascout/youtube-shorts-trend-radar`) Actor

- **URL**: https://apify.com/nexascout/youtube-shorts-trend-radar/changelog.md
- **Full Actor documentation**: https://apify.com/nexascout/youtube-shorts-trend-radar.md

## 0.1.2 — first-run output with unknown publication dates

- Runtime and input-schema defaults now use maxAgeDays: 0.
- Starter and niche example inputs also disable the age filter.
- Explicit positive age filters stay strict: both old videos and unknown dates are excluded.
- OUTPUT now includes appliedFilters and filterCounts (UNKNOWN\_PUBLICATION\_DATE, OLDER\_THAN\_MAX\_AGE, BELOW\_MIN\_VIEWS, NOT\_SUPPORTED\_SHORT).
- History identity and billing behavior remain unchanged. First-run rows can be BASELINE even when publication dates are null.
- 48 automated tests passed, including default first-run output and explicit filter behavior.

Existing saved inputs/tasks are not silently rewritten. Replace maxAgeDays: 90 with 0, or use examples/cloud-retest.json. Publication date, age and lifetime-average view pace remain null when the date is unavailable. Measured growth between observations does not need the publication date.

The installed 0.1.1 collector already returned three rows with maxAgeDays: 0; 0.1.2 makes the setting default and adds filter diagnostics. Do not confuse configuration verification on the installed version with deployment of this package.

## 0.1.1 — public metadata independent of playback

Two cloud tests of package 0.1.0 / Apify build 0.1.1 on 2026-09-21 discovered three Shorts but rejected all three because the player returned LOGIN\_REQUIRED. The tests used direct access and US Residential Proxy respectively. Both produced zero rows and saved no video history.

The old parser required player status OK before checking any other metadata. This release removes that unnecessary coupling: when the player is unavailable, it can read exact views and other metadata already exposed in the anonymous public watch page.

- Validates the current video identity against page metadata before extracting.
- Reads the main video's full view label, never rounded shelf labels or recommended-video counts.
- Preserves missing dates/counters as unavailable; output filtering remains unchanged.
- Adds metadataSource, viewsSource, playerStatus and a warning when public-only metadata is used.
- Adds diagnostic details when neither metadata path is usable.
- Makes no extra HTTP requests and does not require a YouTube login.
- Preserves the existing history identity and event-billing behavior.

45 automated tests passed. The LOGIN\_REQUIRED fallback tests use a simulated player denial with public metadata extracted from a real page. Actual cloud responses were not saved by the old version, so their public metadata availability is not yet confirmed. The new build still needs one bounded cloud test after deployment.

Deploy in the same Apify account using deploy.sh; do not publish yet. Use examples/cloud-retest.json with 512 MB and 300 seconds. Check OUTPUT.collectorVersion is 0.1.1 to confirm the new code ran. Package version and Apify build number are separate.