Olx India Scraper

OLX India Public Classifieds Scraper

This Actor extracts publicly visible OLX India classifieds from olx.in search, category, or listing pages. It returns titles, INR-aware prices, locations, categories, conditions, seller fields, descriptions, images, dates, and canonical listing URLs when those fields are present in public HTML or structured data.

Input

• startUrls: HTTPS OLX India URLs; only olx.in and its public subdomains are accepted.
• maxItems: maximum unique listings, from 1 to 100.
• maxPages: safety cap on public result pages, from 1 to 10.
• requestTimeoutSecs: per-page timeout, from 5 to 60 seconds.
• maxBytes: response-size limit.
• proxyConfiguration: optional standard Apify Proxy settings. Direct requests remain the default; successful navigation reuses one session and one bounded retry rotates the session after a transient/access-boundary response.
• browserFallback: when direct HTML is blocked or contains no parseable listings, render the same public URL in an anonymous Chrome session (enabled by default).
• browserMode: auto (desktop then mobile once), desktop, or mobile.
• requestDelayMs / maxRequestRetries: bounded pacing and retry controls.

The zero-input route is the current locale/category search https://www.olx.in/en-in/mobile-phones_c1453/q-iphone?filter=make_eq_iphone; legacy /items searches, locale routes, category routes, and explicit /item/...-iid-... pages remain accepted. The Actor follows public redirects only when the final URL remains on OLX India. It first uses Apify's impit client with a current desktop Chrome identity and India-consistent language profile. If that response is blocked or parses no listings, it uses a bounded persistent anonymous Chrome context (desktop, then mobile in auto mode) and feeds only the rendered public HTML to the same parser. It does not log in, access private endpoints, submit credentials, solve CAPTCHAs, or bypass challenges. Proxy credentials are passed only to the request transport and are never written to dataset rows or summaries. Failures are logged and summarized outside the dataset. a genuine empty search completes as NO_RESULTS, while a request failure with no listings fails the run.

Output

Successful rows contain business fields plus source provenance (sourceUrl, pageUrl, pageNumber, itemPosition, and scrapedAt). OUTPUT and OUTPUT_SUMMARY contain listing counts, request-failure counts/codes, and browser attempt/session-rotation counts. The dataset contains only actual listing rows. If usable listings were stored before a later failure, the run completes as PARTIAL.

You can download the dataset in various formats such as JSON, HTML, CSV, or Excel. Respect OLX terms, request limits, applicable law, and privacy obligations.

The repository test suite uses checked-in OLX listing fixtures for deterministic parser checks; they are not Actor inputs.