Presearch Videos Scraper

This Actor searches Presearch Videos and publishes normalized public video records.

Input

• query (required): video-search keywords.
• maxItems: maximum records to publish, from 1 to 500.
• maxRequestRetries: retries per failed request, from 0 to 5 (default: 2).
• safeSearch: enables Presearch's explicit-content filter; enabled by default.
• proxyConfiguration: Apify Proxy or custom proxy configuration.

The complete input contract, including types, defaults, and bounds, is in .actor/input_schema.json.

Dataset records

Records include query, sequential position, title, URL, platform, channel name/URL, description, source duration and parsed seconds, ISO publication time plus source time, thumbnail/embed URLs, view count, category, tags, live/short flags, and scrape timestamp. Optional fields are omitted when the target does not provide them. URLs are validated and duplicate videos are removed.

The complete dataset field schema is in .actor/dataset_schema.json.

The Actor captures the structured /results response before navigation, validates status and content type, maps only known public video fields, and uses rendered-card extraction as a bounded fallback. Challenge pages, malformed payloads, and empty result sets fail instead of publishing placeholders.

Target availability

As of 2026-10-25 the whole Presearch web property is down: every route (/, /search, /videos, /images, /robots.txt, www.presearch.com/…, presearch.io/) answers HTTP/1.1 301 with Location: https://account.presearch.com/, and that origin answers HTTP/1.1 530 with a 17-byte text/plain body error code: 1016 (Cloudflare origin DNS error). No Presearch-owned host serves a 200 page, so there is no embedded payload to parse. Without a browser user agent the edge answers HTTP 403 with a Cloudflare challenge (title "Just a moment...").

The Actor therefore fails closed with the true cause instead of reporting an empty result set. validateDocumentResponse classifies a Cloudflare challenge (cf-mitigated: challenge or a challenge page title) before the status branch, walks Playwright's request().redirectedFrom() chain for every hop, and reports the status, page title and a bounded body excerpt, e.g.:

Presearch returned HTTP 530 for https://account.presearch.com/ (redirect chain: https://presearch.com/videos?q=climate%20change -> https://account.presearch.com/) [status 530, title "Origin DNS error | account.presearch.com | Cloudflare", body "Error 1016 …"]; the target origin is unavailable, so this request will not be retried.

Nothing is clicked, injected or solved, no credential is used and no third-party video host is substituted. Revisit when the 301 to account.presearch.com stops and that origin resolves.

How to use

Enter video-search keywords and limits in the Apify Input tab, start the Actor, and review records in the Output dataset. Current availability depends on Presearch restoring its public search service.