Dailymotion Scraper — Videos, Profiles & Search API

Extract public Dailymotion video data from direct video URLs or IDs, creator profiles, category channels and keyword search. Export titles, descriptions, publish times, authors, source-labelled view/like/comment counters, tags and thumbnail URLs to an Apify dataset for research, reporting or content discovery.

The Actor uses Dailymotion's native public HTTP API. No Dailymotion login, cookies, browser, proxy, paid external API or downstream Actor is required. An Apify account/token is needed to run it through Apify.

Quick start

For one public video:

{
  "videoUrls": ["https://www.dailymotion.com/video/x3rdtfy"],
  "maxItems": 1
}

For a mixed batch:

{
  "startUrls": [
    { "url": "https://dai.ly/x3rdtfy" },
    { "url": "https://www.dailymotion.com/NASA" },
    { "url": "https://www.dailymotion.com/search/space/videos" }
  ],
  "channels": ["travel"],
  "resultsPerInput": 5,
  "maxItems": 15,
  "maxConcurrency": 4,
  "failOnPartialFailure": false
}

These were working public targets during implementation tests; content and availability can change.

What each source means

• Direct videos: videoUrls accepts public x-prefixed IDs, /video/{id} links, /embed/video/{id} links and dai.ly/{id} links. A direct input requests one video. The returned ID must exactly match the requested ID.
• Creator profiles/channels: profiles accepts usernames, user IDs or profile URLs, such as NASA or https://www.dailymotion.com/NASA. It returns available public videos, not a standalone comprehensive profile report. A small identity request resolves the stable owner ID; every returned video must match that ID.
• Category channels: channels accepts Dailymotion category IDs such as travel, news and music, or /channel/{category} URLs. These are site-wide categories, not creators. Use profiles for a creator's channel.
• Keywords: searchKeywords runs native Dailymotion video search. Ranking is source-defined and may vary with time or region.
• Hashtags: hashtags: ["space"] runs a broad keyword search for #space, not an exact hashtag feed. Results may match ordinary keywords without containing that hashtag. This limitation is explicit in each row's provenance.querySemantics. Output hashtags are extracted from actual video text or explicitly #-prefixed native tags, not copied from your input.
• Mixed URLs: startUrls recognizes the routes above, /user/{username}, /username/videos, /search/{query}/videos and /search?q={query}. It accepts strings or { "url": "..." } objects. Unsupported hosts, homepage/playlist/login routes, malformed URLs and empty targets are ignored with warnings. At least one usable target must remain.

Duplicate seeds are collapsed where identifiable. Videos are deduplicated globally by native ID before saving. If sources overlap, the first successfully saved source supplies attribution; concurrent runs do not guarantee input-order attribution. A matching video already saved by another source counts as found but is not saved or billed again.

Pricing and budget controls

$0.0005 per saved video — $0.50 per 1,000 results — flat across all pricing tiers. No startup fee. The published Apify Pricing tab is authoritative.

Running users separately pay Apify platform usage, including compute, transfer and storage. The result price is not an all-inclusive platform-usage price. Retries and empty/failed runs can still incur platform usage.

One unique saved default-dataset video row is one result. Empty/error inputs, rejected records, duplicate videos and RUN_SUMMARY do not create result charges. Already-saved rows remain available and billable if strict partial-failure handling later makes the run fail.

The Apify SDK automatically charges the synthetic apify-default-dataset-item event on Actor.pushData(rows). The Actor does not manually charge that event, double-charge rows or charge per HTTP request. Remaining result-charge capacity is checked through the SDK's charging manager, alongside Apify's maximum-paid-dataset-items setting. Dataset writes are serialized; concurrent requests reserve available result capacity before fetching.

Developer-set Free-plan cap

When Apify explicitly reports an unpaid platform account (APIFY_IS_AT_HOME=1 and APIFY_USER_IS_PAYING=0), a global cap of 50 saved rows per run applies. This is the Actor developer's policy, not a universal Apify platform limit. Paid-plan and unknown-status runs have no developer 50-row cap. The per-result price still applies; a Free plan does not mean zero-priced results.

Input reference

Every target array defaults to []. Each array is limited to 200 entries, and the combined total is limited to 500 entries. Search text is limited to 300 characters. Provide at least one supported, non-empty target.

Field	Default	Meaning
startUrls	[]	Mixed supported Dailymotion URLs, as strings or { "url": "..." }.
videoUrls	[]	Public direct video URLs or IDs; at most one matching row per direct source.
profiles	[]	Creator usernames, user IDs or profile URLs.
channels	[]	Category IDs/URLs, not creator usernames.
searchKeywords	[]	Native video-search queries.
hashtags	[]	Tag names with/without #, using broad keyword-search semantics.
resultsPerInput	50	0–1,000 distinct valid videos examined per feed/query before global deduplication. 0 means up to the safety cap of 1,000, not unlimited history.
maxItems	100	Global unique saved-row limit, 1–10,000; charge, platform and Free-plan caps can reduce it.
failOnPartialFailure	false	Make an attempted empty/error/partially failed source fail the whole run when true.
maxConcurrency	4	1–16 lightweight HTTP sources. Output reservations reduce effective concurrency for small budgets.
requestTimeoutSecs	25	Per-request timeout including the body, 5–120 seconds.
maxRetries	2	0–3 retries after transient transport/timeouts or HTTP/API 408, 425, 429 and 5xx failures only. Permanent 400/401/403/404 responses are not retried.
maxPagesPerInput	10	1–10 native pages per feed/query, at most 100 videos per page. Page size stays constant within each source.
maxRequests	200	1–1,000 actual outbound requests globally, including identity requests, retries and redirects.
maxRunTimeSecs	300	Collection runtime bound, 10–1,800 seconds; an earlier Apify timeout takes precedence with shutdown headroom.

Counts are upper bounds, not promised result counts. Empty or repeated pages stop pagination. The native catalog, page, request, runtime and charge limits mean a profile or search is not guaranteed to return complete history. Invalid records and cross-source duplicates can reduce delivered counts. Inputs after an exhausted output budget can be skipped; no new API requests are started after the limit is observed, although already-in-flight requests may complete.

Output fields and numeric semantics

One dataset row is one real, explicitly public and published native video record. A numeric play count is not required when the API omits it.

Field	Meaning
platform	Always dailymotion.
videoId, webVideoUrl	Native ID and canonical public video link.
title, description, text	Plain text; text combines title/description without duplicating an identical description.
createTimeISO	Native created_time converted to UTC ISO, or null; never inferred from the ID.
duration	Source-reported duration in seconds, or null. An explicit source zero is retained, including on live-event metadata.
author	Available stable ID, username (uniqueId), display name (nickname), URL, avatar and explicitly supplied verification boolean. Missing values are omitted, not defaulted to zero/false.
stats.playCount	Native views_total, or null.
stats.diggCount	Native likes_total, or null. The TikTok-compatible field name means Dailymotion likes here.
stats.commentCount	Native comments_total, or null.
stats.shareCount, stats.collectCount	null: reliable public share/save counters are not supplied by this implementation.
statsPrecision	native-api if at least one valid counter is supplied; unavailable if all requested counters are missing/invalid.
hashtags, mentions	Actual textual #tags/@mentions; mentions are not verified account identities.
tags	Native video tags, which can be phrases and are not necessarily hashtags.
coverUrl	Available public Dailymotion-CDN thumbnail URL; not a stored image file and may expire.
sourceType, sourceInput	direct, profile, channel, search or hashtag, plus normalized source attribution without URL tracking parameters.
scrapedAt, provenance	Collection timestamp, public native endpoint, counter-field mapping, stable profile owner ID when applicable, and search semantics.
contentType	Native mode, such as vod or live, when available.

native-api identifies source-reported unrounded values, not independently audited accuracy, real-time freshness or proof that every counter is available. Explicit numeric zeroes are retained; missing, negative, invalid or unsafe counters become null. Generic native tags are never blindly relabelled as hashtags.

Music and follower totals are omitted because this implementation has no reliable supported source for them. No fabricated soundtrack, verification or follower values are added.

Export the dataset as JSON, CSV or Excel from Apify. API clients can read their run's dataset through GET https://api.apify.com/v2/datasets/{datasetId}/items with the real dataset ID and appropriate Apify authentication.

Status and failure rules

RUN_SUMMARY in the default Key-Value Store is a non-result report with per-input statuses, saved-row count, HTTP request/retry totals, stop reasons and warnings.

Situation	Run outcome
No usable input	FAILED, no data rows.
All usable inputs produce zero saved rows	FAILED in either strictness mode.
Valid data plus an empty/deleted/missing/blocked source	Platform success with RUN_SUMMARY.status: SUCCEEDED_WITH_WARNINGS by default.
Same mixed batch with failOnPartialFailure: true	FAILED; already-saved rows remain available and billable.
Sources intentionally skipped at an output/charge/run budget	Reported as skipped, not source failures; zero saved rows still fails.
Unrecoverable dataset write error	FAILED; ambiguous writes are not retried to avoid duplicate rows or charges.

SUCCEEDED_WITH_WARNINGS is the summary's descriptive status; Apify's platform status is SUCCEEDED for such runs. Empty results are not proof of deletion: the source may be private, region-restricted, unavailable, empty or temporarily blocked. Partial failures after an earlier successful page preserve already-saved data.

Efficiency, safety and limitations

• Feed/search pages already contain output metadata: no detail request per video. Profiles add one small stable-identity request. Memory holds at most one page per worker plus a bounded ID set; recommended allocation is 256 MB, with 512 MB available for headroom.
• The Actor fetches only allowlisted HTTPS native API routes. User-supplied URLs are parsed into known IDs/queries, never fetched as arbitrary destinations. Credentials/nonstandard ports/lookalike hosts are rejected; every redirect is revalidated. Response bodies are capped at 4 MiB.
• Transient retries use bounded backoff and bounded Retry-After; all attempts consume the request budget. There is no browser or residential-proxy fallback. If Dailymotion blocks or changes its anonymous API, some/all inputs can fail.
• Metadata availability does not guarantee current playback, including archived live events. No login, age/region/private-content bypass or DRM bypass is provided.
• This Actor does not download cover files or full videos, provide raw MP4 URLs, or extract audio/music. Sending shouldDownloadVideos: true or shouldDownloadCovers: true fails explicitly instead of silently promising a download. Public thumbnail references are returned when available.

Development and verification

Requires Node.js 22+.

npm ci
npm test
npm audit --omit=dev
npm run smoke
npm run smoke:sdk

npm test is deterministic and offline; it does not call Apify. npm run smoke makes small real anonymous native-API requests and writes datasets/summaries under .smoke/, without invoking an Apify Actor. npm run smoke:sdk tests the actual SDK entrypoint using isolated local storage, a clean environment without inherited Apify credentials, zero-data failure, mixed warnings and simulated local PPE capacity. Local PPE simulation is not evidence of a paid user's real billing.

npm start runs the entrypoint with the Apify SDK. For local input, use the SDK's local default Key-Value Store INPUT.json; leave APIFY_IS_AT_HOME unset locally. Do not reuse a real cloud run/storage environment for local tests.

Collect and use public data in accordance with applicable law and platform terms.