Apify actor for extracting data about hotels from Booking.com.
Booking.com provides various types of accommodation all around the world. The user interface is quite friendly for a human user, however to get the data in a machine processable format is not a simple task, since there is no official Booking.com API. This is where this new Apify actor comes in handy.
This actor extracts hotel data from Booking.com, it can either extract directly from the hotel list page or navigate to the detail page to get more detailed information. The results can be ordered by any criteria supported by Booking.com.
Since Booking.com allows only 1000 search results, in case you need to download more,
you will need to utilize the useFilters attribute to tell the crawler to enqueue all the criteria
filtered pages. This will overcome the limit, but will significantly increase the crawling time.
Input is a JSON object with the following properties:
{
"search": SEARCH_QUERY,
"destType": DESTINATION_TYPE,
"simple": EXTRACT_FROM_LIST,
"useFilters": USE_CRITERIA_FILTERING,
"minScore": MINIMUM_HOTEL_RATING,
"maxPages": MAXIMUM_PAGINATION_PAGES,
"concurrency": MAXIMUM_CONCURRENT_PAGES,
"checkIn": CHECK_IN_DATE,
"checkOut": CHECK_OUT_DATE,
"rooms": NUMBER_OF_ROOMS,
"adults": NUMBER_OF_ADULTS,
"children": NUMBER_OF_CHILDREN,
"currency": PREFERRED_CURRENCY,
"language": PREFERRED_LANGUAGE,
"sortBy": BOOKING_SORT_TYPE,
"propertyType": PROPERTY_TYPE,
"minMaxPrice": MIN_MAX_PRICE_RANGE,
"proxyConfig": APIFY_PROXY_CONFIG,
"extendOutputFunction": EXTEND_OUTPUT_FUNCTION
}searchis the only required attribute. This is the Booking.com search query.destTypespecifies type of search, available values arecityandregion.simpledefines if the data should be extracted just from the list page, default isfalse.useFilterssets if the crawler should utilize criteria filters to overcome the limit for 1000 results.minScorespecifies the minimum allowed rating of the hotel to be included in results, default is8.4.maxPagessets maximum number of pagination pages to be crawled.checkIncheck-in date in the yyyy-mm-dd format.checkOutcheck-out date in the yyyy-mm-dd format.roomsnumber of rooms to be set for the search.adultsnumber of adults to be set for the search.childrennumber of children to be set for the search.currencypreferred currency code to be set on the site.languagepreferred language code to be set on the site.propertyTypetype of property to search, it will use filters, so cannot be combined withuseFilters. Must be one of the following:
[
"none",
"Hotels",
"Apartments",
"Hostels",
"Guest houses",
"Homestays",
"Bed and breakfasts",
"Holiday homes",
"Boats",
"Villas",
"Motels",
"Holiday parks",
"Campsites",
"Luxury tents"
]minMaxPricemin-max price range, it will use filters, so cannot be combined withuseFilters. Must be one of the following:
[
"none",
"0-50",
"50-100",
"100-150",
"150-200",
"200+"
]proxyConfigdefines Apify proxy configuration and default group is SHADER, it should respect this format:
"proxyConfig": {
"useApifyProxy": true,
"apifyProxyGroups": [
"SHADER"
]
}sortBysets a hotel attribute by which the results will be ordered, must be one of the following.
[
"upsort_bh", // Show homes first
"price", // Price (lowest first)
"closest_ski_lift_distance", // Distance to nearest ski lift
"class", // Stars [5->1]
"class_asc", // Stars [1->5]
"class_and_price", // Star rating and price
"distance_from_search" // Distance from city centre
"bayesian_review_score" // Top reviewed
]extendOutputFunctionFunction that takes a JQuery handle ($) as argument and returns data that will be merged with the default output, only whensimple= false.
Instead of search INPUT attribute, it is also possible to start the crawler with an array of startUrls.
In such case all the other attributes modifying the URLs will still be applied, it is therefore suggested to
use simple urls and set all the other options using INPUT attributes instead of leaving them in the URL to
avoid URL parameter clashing.
In case the startUrl is a hotel detail page, it will be scraped. In case it is a hotel list page, the result
will depend on the simple attribute. If it's true, the page will be scraped, otherwise all the links to
detail pages will be added to the queue and scraped afterwards.
The startUrls attribute should cotain an array of URLs as follows:
{
"startUrls": [
"https://www.booking.com/hotel/fr/ariane-montparnasse.en-gb.html",
"https://www.booking.com/hotel/fr/heliosopera.en-gb.html",
"https://www.booking.com/hotel/fr/ritz-paris-paris.en-gb.html",
...
],
"simple": false,
"minScore": 8.4,
...
}In case of using the simple INPUT attribute, an example output for a single hotel can look like this:
{
"url": "https://www.booking.com/hotel/cz/elia-ky-kra-snohorska-c-apartments-prague.en-gb.html",
"name": "Centrum Apartments Old Town",
"rating": 10,
"reviews": 7,
"stars": 4,
"price": 86,
"currency": "€",
"roomType": "Deluxe Three-Bedroom Apartment with Terrace",
"persons": 4,
"address": "Prague 01, Prague",
"location": {
"lat": "14.4199419021606",
"lng": "50.0903216331068"
}
}Otherwise the output will be much more comprehensive, especially the rooms array, which will however
contain data only if the checkIn and checkOut INPUT attributes are set.
{
"url": "https://www.booking.com/hotel/cz/elia-ky-kra-snohorska-c-apartments-prague.en-gb.html",
"name": "Centrum Apartments Old Town",
"type": "Apartment",
"description": "Situated in the centre of Prague in a historical building near the Pařížská street, 500 metres from the Old Town Square, the Pragueaparts Old town E offers...",
"stars": "4",
"rating": 10,
"reviews": 7,
"breakfast": null,
"checkInFrom": "15:00",
"checkInTo": "00:00",
"location": {
"lat": "50.0903216",
"lng": "14.4199419"
},
"address": {
"full": "Elišky Krásnohorské 2, Prague, 11000, Czech Republic",
"postalCode": "11000",
"street": "Elišky Krásnohorské 2",
"country": "Czech Republic",
"region": ""
},
"image": "https://t-ec.bstatic.com/images/hotel/max1024x768/183/183313960.jpg",
"rooms": [
{
"available": true,
"roomType": "Deluxe Three-Bedroom Apartment with Terrace",
"bedType": " Bedroom 1: 1 extra-large double bed Bedroom 2: 2 single beds Bedroom 3: 3 single beds and 1 sofa bed ",
"persons": 1,
"price": 85.54,
"currency": "€",
"features": [
"80 m²",
"City view",
"Terrace",
"Flat-screen TV",
"Air conditioning",
"Private bathroom",
...
],
"conditions": [
"Non-refundable"
]
},
...
]
}Keep in mind that it is much more efficient to run one longer scrape (at least one minute) than more shorter ones because of the startup time.
The average consumption is 1 Compute unit for 75 actor pages scraped
-
The actor will not work without proxy, i.e. if you try running it without setting a proxy, it will fail with a message explaining exactly that. There can be a slight difference in price depending on the proxy you use.
-
Booking.com will only display maximum of 1000 results, if you need to circumvent this limitation, you can utilize the
useFiltersINPUT attribute. However, in such case it will not be possible to use any limiting filters in start URLs, because the scraper will override those. -
If you need to get data about specific rooms, the crawler needs to be started with
checkInandcheckOutINPUT attributes (Booking.com only shows room info for specific dates). -
Booking.com sometimes returns some suggested hotels that are outside of the expected city/region as a recommendation. The ator will return all of them in the crawling results so you may recognize more results than your search.
Thank you for trying my actor. I will be very glad for a feedback that you can send to my email dtrungtin@gmail.com. If you find any bug, please create an issue on the Github page.