Facebook

Posts from public Facebook pages with their reactions, plus page details. scraper_type facebook and facebook_profile.

facebook collects the posts of public Facebook pages. One result is one row. For page details only, use facebook_profile.

Run it

Python

import os

from gofetch import GoFetchClient

client = GoFetchClient(api_key=os.environ["GOFETCH_API_KEY"])

run = client.actor("facebook").call(run_input={
    "directUrls": ["https://www.facebook.com/metmuseum"],
    "limit": 20,
})

for row in client.dataset(run["defaultDatasetId"]).iterate_items():
    if "postId" in row:
        print(row["time"], row["url"], row["likes"], row["comments"], row["shares"])

facebook_posts is another name for facebook.

Inputs

Field	Type	Default	What it does
directUrls	list of URLs		Page links: https://www.facebook.com/<page>, /profile.php?id=<id>, /people/<name>/<id> or /p/<name>-<id>. A link to a page’s post collects the page. pageUrls, urls and startUrls are accepted as other names.
limit	integer	50	Posts per page, from 0 to 1,000,000. 0 collects every post.
onlyPostsNewerThan	date (YYYY-MM-DD)		Only posts published on or after this date (from 00:00 UTC). Not in the future.
maxTotalChargeUsd	number		A spending cap, split across the pages as a per-page limit that never exceeds your limit. See Billing.

Groups, events, Watch, Marketplace and other links that are not pages are skipped; the job fails only if no link is usable.

A date filter within the last 29 days makes the job a recent crawl, which is billed at its own rate; see Billing. A recent crawl reads the recent part of each page’s timeline, and reports in completeness any stretch of dates it could not cover.

Output

The job returns one profile_info row per page (the fields of page details) and one row per post. Both are billed as results. A real post row from a job for The Met:

[
  {
    "postId": "1553172153522100",
    "url": "https://www.facebook.com/metmuseum/posts/pfbid0QyLKm7CvMnijUF5WqimuPEqufW2fMvoh4asVzqon72vfwnsRV2EN7uEQnJachTYql",
    "facebookUrl": "https://www.facebook.com/metmuseum",
    "pageName": "The Metropolitan Museum of Art, New York | New York NY",
    "pageVerified": true,
    "text": "🎃 Lights out at The Met! 🌙 Join us for The Met Pajama Party, a Halloween-themed evening for the whole family. Wear your costume or your coziest pajamas and enjoy art activities, a Halloween-inspired menu, and live entertainment in the Temple of Dendur and The Charles Engelhard Court.\n\n📅 Wednesday, October 28, 5-7 PM\n📍 The Met Fifth Avenue\n\nGet your tickets and see event details: https://met.org/4B4xP8W",
    "time": "2026-10-14T15:41:04.000Z",
    "likes": 108,
    "comments": 2,
    "shares": 8,
    "isVideo": false,
    "viewsCount": 0,
    "media": [
      {
        "__typename": "Photo",
        "__isMedia": null,
        "id": "100064879894574",
        "image": {
          "uri": "https://lookaside.fbsbx.com/lookaside/crawler/media/?media_id=1553172086855440"
        },
        "thumbnail": "https://lookaside.fbsbx.com/lookaside/crawler/media/?media_id=1553172086855440"
      },
      {
        "__typename": "Photo",
        "__isMedia": null,
        "id": "1553172086855440",
        "image": {
          "uri": "https://lookaside.fbsbx.com/lookaside/crawler/media/?media_id=1553172086855440"
        },
        "thumbnail": "https://lookaside.fbsbx.com/lookaside/crawler/media/?media_id=1553172086855440"
      },
      {
        "__typename": "Photo",
        "__isMedia": null,
        "id": "1552215593617756",
        "image": {
          "uri": "https://lookaside.fbsbx.com/lookaside/crawler/media/?media_id=1552215593617756"
        },
        "thumbnail": "https://lookaside.fbsbx.com/lookaside/crawler/media/?media_id=1552215593617756"
      },
      {
        "__typename": "Photo",
        "__isMedia": null,
        "id": "1552215650284417",
        "image": {
          "uri": "https://lookaside.fbsbx.com/lookaside/crawler/media/?media_id=1552215703617745"
        },
        "thumbnail": "https://lookaside.fbsbx.com/lookaside/crawler/media/?media_id=1552215703617745"
      }
    ],
    "attachments": [],
    "error": null,
    "errorDescription": null
  }
]

Field	Type	What it holds
postId	text	Facebook’s id for the post.
url	URL	Link to the post; videos link to https://www.facebook.com/reel/<id>/.
facebookUrl	URL	The page link exactly as you sent it. Use it to match rows to your input.
pageName, pageVerified	text, true / false / null	The page and its verified badge.
text	text	The post’s text; "" when there is none.
time	date-time or null	When it was posted (ISO 8601, UTC).
likes	integer	All reactions together, not only likes; 0 when unknown.
comments, shares	integer	Comment and share counts.
isVideo, viewsCount	true / false, integer	A video post and its views; 0 when unknown.
media	list	Facebook’s media objects for the post, passed through as Facebook describes them.
attachments	list	Always empty.
error, errorDescription	null	Always null on a post row.

What logged-out means here

You get: posts from public pages, with their text, reaction, comment and share counts, as a signed-out visitor sees them.

Not available: pages behind a login wall or with no public timeline, groups, events, and anything only a signed-in user can see. A page that returned no posts is listed in scraper_metadata.targets_failed:

reason	Means	Worth retrying
unavailable	The page shows no public timeline to signed-out visitors, or asks to log in	No
not_found	The page does not exist or is not available	No
blocked	Facebook refused the request	No
extraction_failed	A temporary failure	Yes

Completeness

Key	What it says
profiles_scraped, profiles_failed	Pages processed, and pages that ended with an error and no posts. A page that returned an error counts in both.
completeness.complete	false when any page was not collected over the whole range you asked for.
completeness.pages_total, pages_incomplete, pages_failed, pages_rate_limited	Pages by outcome.
completeness.gaps	Stretches of dates a page could not cover, each with page, from and to.
completeness.reason	Why the job is incomplete, when it is.
targets_failed	Pages that returned nothing: target, reason, detail (for reading, not parsing) and retriable.

Page details

facebook_profile returns one row per page with its details. It takes the same page links, and bills one result per page that answered, including a page that answers it is not available. limit and date filters do not apply.

Python

import os

from gofetch import GoFetchClient

client = GoFetchClient(api_key=os.environ["GOFETCH_API_KEY"])

run = client.actor("facebook_profile").call(run_input={
    "directUrls": ["https://www.facebook.com/NASA", "https://www.facebook.com/metmuseum"],
})

for page in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(page.get("pageName"), page.get("followers"), page.get("categories"))

A real row for NASA’s page (the links to stored images are masked here):

[
  {
    "facebookUrl": "https://www.facebook.com/NASA",
    "facebookId": "100044561550831",
    "pageName": "NASA - National Aeronautics and Space Administration",
    "title": "NASA - National Aeronautics and Space Administration",
    "likes": 28714724,
    "followers": 28714724,
    "verified": true,
    "categories": [
      "User",
      "Government organization"
    ],
    "info": [
      "NASA - National Aeronautics and Space Administration. 28,714,724 followers",
      "124,545 talking about this. Explore the universe and discover our home planet. \nThere's space for everybody. ✨"
    ],
    "profilePictureUrl": "<masked: https://go-fetch.io/api/v1/jobs/{id}/media/{name}?token=…>",
    "coverPhotoUrl": "<masked: https://go-fetch.io/api/v1/jobs/{id}/media/{name}?token=…>",
    "profilePhoto": "https://www.facebook.com/photo/?fbid=423033359192046&set=a.416661003162615",
    "type": "profile_info"
  }
]

Field	Type	What it holds
type	text	profile_info.
facebookUrl	URL	The page link as you sent it.
facebookId	text or null	The page’s id.
pageName, title	text	The page’s name.
followers	integer	Followers; 0 when not shown.
likes	integer	Equal to followers when Facebook shows the count in the page’s summary, which it labels “followers” or “likes”; otherwise 0.
verified	true / false / null	The verified badge.
categories	list of text	The kind of page, then its category, such as ["User", "Government organization"].
info	list of text	The page’s public description, split into parts.
profilePictureUrl, coverPhotoUrl	text or null	Absolute links to copies we stored, https://go-fetch.io/api/v1/jobs/{id}/media/{name}?token=…. No API key needed: GET or HEAD on the link returns a 302 to a download link for that method, valid for 10 minutes. The link itself is valid for 5 days, and each read of the results returns fresh ones. Without the token, the path still works with your X-API-Key header (counts toward the key’s hourly rate limit); a bad or expired token is a 404. The copies go when the job’s results are deleted. Facebook’s own links if storing failed.
profilePhoto	URL or null	Link to the profile photo’s Facebook page.

A page that answers it is not available returns {"url", "error": "not_available", "errorDescription"}. Pages that do not answer at all are listed in targets_failed, not returned as rows. For this type, completeness reports no pages and can be ignored.