2GIS Business Scraper

One 2GIS organisation read in full from its id — including the three fields a search row can never carry: the phone numbers, the websites and the email addresses. Alongside them the row holds the street address with its city, map coordinates, every category, opening hours, the chain the branch belongs to with its branch count, the branch's own rating beside the chain's, and 2GIS's full structured attribute groups.

This Actor exists because of one measured fact: 2GIS's own search request omits contact fields entirely. The same organisation returns no contacts at all from a search and four phone numbers, a website and two email addresses from here. It is a property of which lookup you make, not of the business, so no amount of paging a search will ever produce a phone number.

Accepted input

placeIds is required and takes one 2GIS branch id per line — exactly the id every 2GIS Search Scraper row publishes, which is how the two compose.

Three forms all reduce to the same value and can be mixed in one list: the bare id (70000001007179642), a 2GIS place link (https://2gis.ru/moscow/firm/70000001007179642), and the decorated id from a raw 2GIS search response (70000001007179642_ku4dsp…). That last one is worth knowing about: 2GIS's own search keys a row as the branch id plus a 130-character per-request context blob, and the blob works nowhere else on the site — not here, not on reviews, not on the public place URL.

locale is optional and takes one of the city's own locales, written as ru_RU or en_AE. Leave it empty for the city's default. A locale a city does not offer is refused by 2GIS outright rather than falling back.

maxItems defaults to 100 and caps how many businesses the run reads. One business is one request and one row here, so it is a cap on the list rather than on a page of results.

{
  "placeIds": [
    "70000001007179642",
    "https://2gis.ru/moscow/firm/70000001006123456"
  ],
  "locale": "ru_RU",
  "maxItems": 100
}

Response fields

{
  "requested_id": "70000001007179642",
  "id": "70000001007179642",
  "name": "Му-Му, кафе",
  "type": "branch",
  "url": "https://2gis.ru/moscow/firm/70000001007179642",
  "phones": ["+74957818781", "+74959999999"],
  "websites": ["https://mu-mu.ru"],
  "emails": ["info@mu-mu.ru"],
  "contacts": {
    "phones": [
      {
        "value": "+74957818781",
        "text": "+7 (495) 781‒87‒81",
        "comment": "служба доставки",
        "tracking_url": null
      }
    ],
    "websites": [
      {
        "value": "https://mu-mu.ru",
        "text": "mu-mu.ru",
        "tracking_url": "http://link.2gis.ru/1.2/97BA0CE6/…"
      }
    ],
    "emails": []
  },
  "street_address": "Манежная площадь, 1 ст2",
  "address_comment": "-2 этаж; Средний уровень",
  "building_name": "Охотный ряд",
  "postcode": "125009",
  "country": "Россия",
  "region": "Москва",
  "address_city": "Москва",
  "district": "Тверской",
  "latitude": 55.755301,
  "longitude": 37.614586,
  "region_id": "32",
  "city_alias": "moscow",
  "primary_rubric": "Кафе",
  "rubrics": [{ "id": "161", "name": "Кафе", "kind": "primary" }],
  "schedule": { "Mon": { "working_hours": [{ "from": "09:00", "to": "22:00" }] } },
  "attribute_groups": [{ "name": "Средний счёт" }],
  "chain_id": "4504136498368318",
  "chain_name": "Му-Му, кафе",
  "chain_branch_count": 18,
  "rating": 4.8,
  "review_count": 1090,
  "org_rating": 4.3,
  "org_review_count": 7817,
  "is_reviewable": true
}

Fields absent from a record are returned as null rather than omitted, so every row has the same shape.

Contacts are published twice, on purpose. phones, websites and emails are plain lists of the real values — the dialable number, the actual website, the mailbox — which is what a spreadsheet wants. contacts keeps everything beside them: 2GIS's own formatting of each number, the comment naming which department it belongs to, and its click-tracking redirector.

That redirector is why the two exist. 2GIS itself puts its link.2gis.ru tracking link where the website should be, and the real address in a separate field — the inverse of what it does for phones and emails. Republishing that verbatim would hand you a tracking link as a business's website. So the two are swapped back, websites always carries the real site, and the redirector is kept under tracking_url inside contacts rather than thrown away.

street_address is the street line only — the city is not in it. That is 2GIS's own shape, and the city is lifted out of its administrative division list into address_city.

rating and org_rating are both published because a branch of a chain frequently disagrees with its parent. review_count is the branch's rating count, and it is larger than the number of written reviews 2GIS Reviews Scraper will return, because 2GIS counts star-only ratings here and returns written reviews there.

attribute_groups is 2GIS's own structured attribute list — average cheque, cuisine, amenities — grouped as it groups them. It is republished whole rather than renamed, because there is no measured field list to rename it into, and renaming fields nobody has seen populated publishes a permanently-null column and calls it a schema.

Behaviour on partial results

A branch that no longer resolves is skipped with a line in the run log and the rest of the list still runs — twenty ids taken from a search last week will not all still be listed, and losing nineteen good rows to one closed restaurant is the wrong trade.

An entry that is not a branch id is skipped before any request is made, with a message naming the shape.

A business that publishes no contacts returns a successful row with empty lists rather than nulls, so the column can be counted without special-casing. The run log states how many phone numbers, websites and email addresses arrived for each one, because that is the reason the request was spent.

Frequently asked questions

Why is this a separate Actor rather than an option on the search? Because it is a separate cost, by a factor of fifty. A search request returns fifty organisations and this one returns one, so folding it into the search as a checkbox would hide a fifty-fold price difference behind a tick box. It is priced on its own cost driver instead.

Where do I get branch ids? From 2GIS Search Scraper — every row publishes exactly the value this takes, as id. A 2GIS place link works too if you have one in a browser.

I pasted the id from a 2GIS API response and it was rejected. That is the decorated search id: the branch id plus a 130-character per-request context blob. It is accepted here and reduced automatically, so if it was rejected the value was something else — check for a truncated paste.

Why does the website column not match what I see in 2GIS's own JSON? Because 2GIS puts its click-tracking redirector in the value position for websites and the real address elsewhere, the inverse of what it does for phones. The two are swapped back here; the redirector is still available under contacts.websites[].tracking_url.

Can I look a business up by name instead of by id? Not here. 2GIS Search Scraper is the lookup by name, and its rows carry the ids this takes.

Is a 2GIS account or API key required? No. No account, session cookie or key of any kind is supplied to the Actor or needed by it. Open the url on any row and you see the same record.

Related Actors

2GIS Search Scraper is where the ids come from: business search across 206 cities, with a flag on every row saying whether 2GIS matched the query at all.

2GIS Reviews Scraper takes the same ids and returns the written reviews, each naming which source it came from — 2GIS aggregates Flamp, Booking and others under one listing.

2GIS Covered Cities lists the city slugs and the locales each city accepts, which is what the locale input here needs.