Yandex Video Scraper

Searches the public Yandex Video interface and returns normalized video cards from supported platforms such as YouTube, VK Video, Rutube, and Dzen.

Modes

• query: one search phrase.
• queries: up to 25 search phrases in one run.
• When multiple queries are supplied, output is interleaved round-robin and constrained by both maxItems and maxItemsPerQuery.
• Pagination uses Yandex's page-number behavior and stops at maxPages or the requested item limit.

Output

Every record includes a stable ID, position, query position, page, title, direct video URL, domain, platform, search context, source, and scrape timestamp. When Yandex displays them, records also include the Yandex preview URL, thumbnail and dimensions, duration and duration seconds, channel, displayed and numeric view counts, and published-date text.

The optional country input accepts a country code. When Apify Proxy is enabled and no proxy country is set, it selects that country for the proxy session. Video title HTML, embed URL, favicon, source dimensions, category, country, and Yandex region are included only when available from the result card or request context.

Optional fields are omitted rather than filled with nulls or empty strings. CAPTCHA, verification, error, and block pages are never stored as records. Transient Yandex request identifiers are removed from output URLs.

Proxy guidance

Direct access can work, but Yandex may challenge datacenter IPs. On Apify, use a Residential proxy and a suitable country, commonly RU. The GOOGLE_SERP proxy group cannot be used because it supports Google Search and Google Shopping only, not Yandex.

This Actor does not solve or bypass CAPTCHAs. A challenge fails the run so consumers never receive misleading partial data.

Local development

npm install
npm test
apify run --purge --input-file test/inputs/single.json
node validate-datasets.js

The Actor uses the Chrome Playwright image in Docker and the installed Chrome channel on Windows local runs. The default media optimization blocks video and font downloads but preserves page HTML, scripts, styles, and thumbnail URLs.

Example input

{
  "queries": ["machine learning", "космос"],
  "maxItems": 20,
  "maxItemsPerQuery": 10,
  "maxPages": 2,
  "language": "ru",
  "maxConcurrency": 2,
  "proxyConfiguration": {
    "useApifyProxy": true,
    "groups": ["RESIDENTIAL"],
    "countryCode": "RU"
  }
}