# Changelog of Douyin Scraper — TikTok China API & Video Downloader (`sian.agency/douyin-scraper`) Actor

- **URL**: https://apify.com/sian.agency/douyin-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/sian.agency/douyin-scraper.md

## Changelog

All notable changes to **Douyin Scraper** will be documented in this file.

### \[2026-08-24]

#### 📊 A run report that shows what you got, not just that it finished

- **Results you can paste straight out** — every run now opens with your top videos, creators or comments as a one-click copy panel, ranked by what actually travelled. Drop it into a brief, a trend report or a sentiment model without touching the raw dataset.
- **An itemized bill on every run** — the report closes with exactly what you were charged for, priced at your own plan's rate. The headline number and the invoice are the same number.
- **One-click retry for empty pages** — anything Douyin did not answer comes back with a reason, a concrete next step, and a copy-paste input that repeats just that run.
- **Creator avatars render in the dataset** — profile pictures now show as images in the Douyin dataset view instead of a bare URL, so a creator shortlist is scannable at a glance.
- **Chinese text renders correctly in the report** — captions, comments and creator names in the run report now display as written, on every browser.

#### 🎁 Clearer free-tier terms, stated up front

- **Your tier and its limits are printed at the top of every run** — rows per run, pages per run, and what a paid plan changes. No surprises halfway through a job.
- **Free runs return up to 25 rows** with the full field set and the same data quality. Paid plans return everything Douyin has.
- **You are never charged for an empty page or an error** — billing is per row delivered, and the run report says so with the numbers to check it.

#### 💎 User Benefits

- Read the outcome of a run in five seconds instead of scrolling a dataset to find out whether it worked.
- Check what a job cost before it appears on your invoice, with per-row prices from your own plan.
- Recover a partial run in one click rather than rebuilding the input by hand.

#### 🎯 Use Cases

- **KOL Agency** scanning a fresh Douyin creator shortlist by avatar and follower count before anyone opens the dataset.
- **NLP Researcher** copying a run's whole comment thread as plain text straight into a sentiment pipeline.
- **Market Intelligence Analyst** reconciling a month of scheduled Douyin runs against the itemized charges on each one.

### \[2026-07-14]

#### 🔥 Sort a creator's catalog by popularity

- 🔥 **NEW: Sort User Videos by popularity.** The `User Videos` operation now accepts a `sortType` input — keep the default `Latest first` for chronological catalogs, or switch to `Most popular first` to pull a creator's all-time top videos ranked by engagement. Perfect for instant KOL "greatest hits" audits.
- 🛡️ **Better User Videos coverage.** If a creator's video list comes back empty from the default data source, the actor now automatically retries through an alternate source before giving up — fewer empty results on hard-to-reach accounts.

### \[2026-06-20]

#### 🧯 Fewer confusing failures on busy days

- 🧯 **Graceful gateway-error handling.** Temporary error pages served by Douyin are now retried and routed cleanly instead of surfacing as confusing parsing failures.

### \[2026-05-12]

#### 🎉 Douyin Scraper — Launch!

- **Seven Scrapers in One Actor** — extract Douyin (TikTok China) video data, profiles, user video catalogs, comments, sub-comments (reply threads), and run keyword searches across videos and creators — pick the operation, paste the input, one clean dataset per run.
- **Video Detail Extraction** — full Douyin video payload by ID (`aweme_id`): caption, play/like/share/comment/collect counts, playable video URLs, cover, dynamic cover, music metadata, creator info, duration, timestamps.
- **User Profile Scraping** — full Douyin creator profile by `secUid`: follower count, following count, total likes received, post count, bio, verification badge, IP location, full avatar set — ideal for KOL discovery.
- **User Video Catalog Mapping** — paginated history of any creator's posts (~20 videos/page) with engagement metrics for KOL audit & content-strategy reverse engineering.
- **Video Comment Mining** — paginated top-level comments with author, IP location, like count, reply count, hot-comment flag — feed your sentiment models with structured comment data.
- **Video Sub-Comment / Reply Thread Extraction** — paginated reply threads under any comment with `replyCount > 0`. **Unique feature** — no other Douyin actor on the Apify store exposes nested replies.
- **Search Videos by Keyword** — surface trending Douyin videos in any niche (Chinese · English · mixed queries supported), ~12 results/page.
- **Search Creators by Keyword** — discover Chinese-market KOLs by keyword with handle, followers, verification, bio — ~19 results/page.
- **BigInt-Safe 19-Digit IDs** — Douyin `aweme_id`, `uid`, and `cid` are preserved at full precision (most scrapers silently corrupt them and break cross-operation joins).
- **HTTPS-Normalized Media URLs** — every video URL, cover, avatar, and music track is delivered ready to embed.
- **Cursor Pagination + Safety Cap** — stream up to 50 pages per run with a single input parameter; auto-stop on last page.
- **Curated camelCase Aliases + Raw Fields** — clean `videoId` · `secUid` · `playCount` · `userName` · `caption` aliases for everyday users, plus every raw Douyin field spread alongside for power users.
- **HTML Run Report** — every run produces an HTML summary (success rate, errors, duration, inputs) saved to the key-value store.
- **Predefined Dataset View** — an Overview layout that opens the right table in the Apify Console with one click, whichever operation you ran.
- **Auto-Retry on Transient Failures** — a page that does not answer first time is retried automatically, with backoff, before it is ever reported as failed. You see the results, not the retries.

#### 💎 User Benefits

- **Save hours per KOL shortlist** — search Douyin creators by keyword and rank by followers in 5 minutes instead of 5 hours of manual scrolling.
- **No API key, no setup, no proxies** — paste a video ID, secUid, or keyword and run. Production-ready from the first click.
- **Pay-per-result pricing** — only charged for successful extractions; never billed for empty pages or errors.
- **Deeper sentiment data than any rival** — top-level + nested reply threads give you the full conversation tree, not just the noisy surface layer.
- **Bulletproof data shape** — flat structure with both camelCase aliases and raw fields means every downstream tool (BigQuery, Notion, Airtable, Slack) just works.
- **Cross-operation joins that actually work** — bigint-safe IDs let you pivot from search → video detail → comments → sub-comments without losing a single character of precision.
- **GDPR-friendly defaults** — extracts only publicly shared data; no private user info, no email or phone scraping.

#### 🎯 Use Cases

- **KOL Agency** building a Douyin shortlist for a brand campaign — runs Search Users + User Profile across 10 keywords in 30 minutes.
- **Market Intelligence Analyst** at a global FMCG brand monitoring China short-video trends before launching a new SKU — schedules daily Search Videos runs across product-category keywords.
- **Brand Manager** at a luxury label tracking counterfeit & gray-market mentions on Douyin — hourly Search Videos with brand keyword + Slack alerts on high-engagement matches.
- **NLP Researcher** building a Chinese-language sentiment dataset — pulls Video Comments + Sub-Comments across hundreds of viral videos in one weekend, getting both surface reactions and threaded debate.
- **Influencer-Marketing Agency** doing due diligence on a creator before signing — runs User Videos across 20 pages to audit posting cadence, engagement drift, and content quality.
- **Content Strategist** at a media agency tracking viral video patterns weekly — scheduled Search Videos run feeds a trend dashboard with view-velocity rankings.
- **Music-Industry Analyst** tracking which sounds are trending on Douyin — Video Detail rows expose `musicId`, `musicTitle`, `musicAuthor`, and `musicPlayUrl` for every scraped video.
