# Changelog of Hugging Face Scraper — Models, Datasets & Spaces (`khadinakbar/huggingface-all-in-one-scraper`) Actor

- **URL**: https://apify.com/khadinakbar/huggingface-all-in-one-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/khadinakbar/huggingface-all-in-one-scraper.md

### 1.2 (2026-05-29)

- Fix: profile counts (modelsCount/datasetsCount/spacesCount) were 1 because HF API does not return x-total-count. Now fetched via limit=1000 and array length.

### 1.1 (2026-05-29)

- Reliability hardening after brutal test battery (100% success rate target).
- Fix: gated boolean false from HF API caused schema validation failures on detailed mode for datasets, spaces, models. Now normalized via normalizeGated() helper.
- Fix: dropped-record counts no longer increment on failed pushes (move counts after pushData success).
- Fix: per-record try/catch — single bad portfolio entry no longer fails the whole user/org target.
- Fix: empty/whitespace targets now surface as explicit warnings instead of silent drops.
- Fix: URL paths with /tree, /blob, /commits, /discussions, /settings suffix now strip correctly.
- Add: user/org records now include accurate modelsCount, datasetsCount, spacesCount via x-total-count header.
- Add: sanitizeRecord() strips undefined fields before push.
- Add: dataset_schema gated accepts boolean defensively.
- Add: concise summary fields truncate at 600 chars to keep token budget.

### 1.0 (2026-05-28)

- Initial release: URL auto-detect across models, datasets, spaces, papers, users, orgs, collections.
- Deep model card extraction with lineage (base model, adapters, quantized children), eval results, dataset citations.
- Public HF API + raw README fetch. No auth required. Optional HF_TOKEN env var for 5K req/h.
- MCP-first input/output, concise/detailed response modes.