Dongchedi Scraper
Pricing
from $3.00 / 1,000 results
Dongchedi Scraper
Scrape Dongchedi (dongchedi.com) - ByteDance's auto platform. Hot cars by category, recent new-car launches with prices and images, car-name lookup, and hot-brand directory - all parsed from the public homepage. No login required.
Pricing
from $3.00 / 1,000 results
Rating
0.0
(0)
Developer
Crawler Bros
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
a day ago
Last modified
Categories
Share
Scrape Dongchedi (dongchedi.com) — ByteDance's automotive platform. Hot cars by category, recent new-car launches with prices/status/dates/images, car-name lookup, and the hot-brand directory — all parsed from the public homepage. No login, HTTP-only.
What this actor does
- Six modes:
hotCars(default),newCars,byCarModel,byBrand,news(homepage headlines),videos(homepage video list) - Hot cars — the homepage's 轿车 / SUV / 其他 hot lists (27 cars each, with series ids and canonical series URLs)
- New-car launches — the 近期重磅新车 block parsed from the structured
__NEXT_DATA__JSON (HTML regex fallback): name, price range (万元), status (预售/已上市), release tags (改款/全新车系/新增车型/小改款), release date (ISO), brand name + id, energy type (fuel/ev/hybrid/phev/erev), article title + URL, and a verified-accessible image - byCarModel — look up a car name across the homepage lists (
凯美瑞,Model Y,奥迪A6L…) - byBrand — hot-brand directory (33 brands, was 16) with ids, urls and logos
- news — today's headlines (up to ~55): article id, title, article URL, video flag, hot flag
- videos — the homepage video feed (30): title, video URL, duration, watch count, column, publish date, cover
- Rehosted video covers: the homepage serves video covers as signed CDN URLs (
p*-dcd-sign.byteimg.comwithx-expires) that 403 about 7 days after the scrape. The actor downloads each cover and rehosts it into its Apify Key-Value Store, socoverUrlis a permanent, anonymous, hotlink-safe URL. The original signed URL is kept for traceability ascoverUrlOriginal(it expires — treat it as a short-lived cache hint). If a cover download fails the field is omitted entirely (never an expiringcoverUrl). - Auto-escalation: on 403/429 the actor lazily engages the free Apify AUTO datacenter proxy and retries with backoff
- Typed error records for invalid inputs
- Empty fields are omitted (
strip_nullsbefore every push)
Output fields
Car records (recordType: "car"): carName, seriesId, seriesUrl, category (轿车 / SUV / 其他)
New-car records (recordType: "newCar"): carName, priceMin, priceMax (10k CNY), status, releaseTag(+releaseTags), releaseDate, seriesId, seriesUrl, brandId, brandName, energyType, articleTitle, articleUrl, imageUrl
Brand records (recordType: "brand"): brandName, brandId, brandUrl, logoUrl
News records (recordType: "news"): articleId, title, articleUrl, isVideo, hot
Video records (recordType: "video"): videoId, title, videoUrl, coverUrl (permanent, rehosted in the actor's KV store), coverUrlOriginal (upstream signed URL — expires ~7 days after scraping), videoDuration, watchCount, columnName, publishDate
Error records (recordType: "error"): input, message
All data records carry recordType, scrapedAt, sourceUrl; error records are diagnostics (input, message) without a sourceUrl.
Input
| Field | Type | Default | Description |
|---|---|---|---|
mode | select | hotCars | hotCars / newCars / byCarModel / byBrand / news / videos |
carCategory | select | all | all / 轿车 / SUV / 其他 |
carModel | text | 凯美瑞 | Car name to look up (mode=byCarModel) |
brandId | select | 3 | 33 brands: 奔驰, 宝马, 大众, 奥迪, 小米汽车, 特斯拉, 本田, 丰田, 吉利汽车, 坦克, 路虎, 比亚迪, 红旗, 理想汽车, 雷克萨斯, 凯迪拉克, 保时捷, 蔚来, 极氪, 小鹏汽车, 现代, 沃尔沃, 福特, 日产, 广汽传祺, 别克, 长安, 腾势, 奇瑞, 北京越野, 方程豹, 哈弗, 零跑汽车 |
minPrice / maxPrice | number (0–1000) | – | Price filter (10k CNY, mode=newCars) |
containsKeyword | text | – | Substring filter on car/brand/news/video names |
maxItems | int (1–500) | 50 | Hard cap |
Examples
{ "mode": "hotCars", "carCategory": "SUV" }{ "mode": "newCars", "minPrice": 10, "maxPrice": 25 }{ "mode": "byCarModel", "carModel": "Model Y" }{ "mode": "byBrand", "brandId": "535", "maxItems": 10 }{ "mode": "news", "maxItems": 20 }{ "mode": "videos", "containsKeyword": "实测", "maxItems": 10 }
Data source
The dongchedi.com homepage is server-side rendered and requires no login. Parsing prefers the structured __NEXT_DATA__ JSON embedded in the page (stable schema: newCarData, popularModels.hot_brand, popularModels.*.series, todayNews, homeOriginal.video_list) with the HTML regex parsers as fallback. Car detail pages (/auto/series/<id>, /auto/library/*, brand libraries) redirect to a login wall, so this actor's coverage is the homepage surface: hot lists, new-car launches, brand directory, news and videos. This is documented as a limitation — see below.
Limitations
- Car series detail pages, the car library (
/auto/library/*) and brand library pages require a Dongchedi login — they redirect to/login-required. Only the homepage is public, sobyCarModelmatches against the homepage lists and new-car block (the 30 most prominent models). - Prices exist only for new-car launches, not for the generic hot lists.
- The homepage hot lists rotate; the car set is Dongchedi's editorial pick, not the full catalog.
- Some new-car entries have no price yet (
暂无报价) — those records simply omitpriceMin/priceMax. - News and videos are homepage-only feeds (no pagination);
hotmarks the curated head articles.
Use cases
- Daily new-car launch monitoring with pricing
- Market-mapping of the top sedans / SUVs in China
- Brand directory building with logos
- Reference data for automotive content pipelines
FAQ
What does priceMin/priceMax mean? The new-car block shows a range in 万元 (10k CNY), e.g. 19.79-19.79 → both 19.79. Single values are emitted as both bounds.
Why can't I get full specs? Spec tables live on the login-gated series pages. This actor covers the entire publicly accessible surface of dongchedi.com.
What are series ids? Dongchedi's internal car-series ids used in canonical URLs: https://www.dongchedi.com/auto/series/<id>.