Google Place with R

Question

My purpose is to use R to query google api. I have a list of addresses and names (belong to shops, restaurants, etc.) and for each of them I need to store:

"latitude", "longitude", "business type"

My idea is to use google place api:

-Use the Text Search Requests to search the address to store latitude e longitude

https://maps.googleapis.com/maps/api/place/textsearch/xml?query=address&key=AddYourOwnKeyHere

-Then use the Radar Search Requests to search the name nearby the latitude e longitude. So I get the ID_place

https://maps.googleapis.com/maps/api/place/radarsearch/json?location=latitude,longitude&radius=1&keyword=name&key=AddYourOwnKeyHere

-Thanks to the ID_place, I query the Place Details Requests to get detail about business type (es. "types" : [ "food" ] )

https://maps.googleapis.com/maps/api/place/details/json?placeid=ID_place&key=AddYourOwnKeyHere

I'm very new to use google api and maybe is not the most efficient way to do it. But I need to give some consideration: -Sometimes the address is not complete (sometimes I have the civic number missing) -Not always the name is correct (sometimes I have abbreviations)

I need to understand some things:

• -If this is the best way to do it
• -How query these Google api with R
• -How handle the json output

Indopacific

Answer 1

To access the Google Places API in R you can use my googleway package, and in particular google_places().

This also handles the JSON output for you.

library(googleway)

key <- 'your_api_key_goes_here'

df_places <- google_places(search_string = "cafe", 
                           location = c(-37.81827, 144.9671),   ## melbourne, AU
                           key = key)

df_places$results$name

# [1] "Time Out Fed Square"    "Dukes Coffee Roasters"  "Alice Nivens"           "Little Cupcakes"        "Lindt Chocolate Cafe"   "Cafe Andiamo"          
# [7] "The Journal Cafe"       "RMB Cafe Bar"           "Cafe Issus"             "ACMI Cafe & Bar"        "Ponyfish Island"        "Aix Cafe"              
# [13] "Seedling Cafe"          "Eliana Lulu"            "B3 Cafe"                "Lindt Chocolate Cafe"   "Switch Board Cafe"      "Urban Express Cafe"    
# [19] "Cento Mani"             "1932 Cafe & Restaurant"

If you want more details about a given place you can use google_place_details() to search for one of the place_ids returned from the previous query

df_details <- google_place_details(place_id = df_places$results[1, "place_id"],
                                   key = key)

df_details$result
# [1] "restaurant"        "cafe"              "bar"               "food"              "point_of_interest" "establishment"

df_details$result$reviews
#      aspects            author_name                                    author_url language rating
# 1 0, overall Fredrich Oliver-bently https://plus.google.com/114792371400821038660       en      2
# 2 0, overall           Jenn Besonia https://plus.google.com/110502657363574676299       en      2
# 3 0, overall             Sewa G S R https://plus.google.com/118332347441422887680       en      1
# 4 0, overall             M Mathumbu https://plus.google.com/104636428392041496439       en      2
# 5 2, overall                 Bo Cui https://plus.google.com/104475569220729624507       en      4

Miranda

Answer 2

You could use library "httr" to get the result from google API:

res<-GET( "https://maps.googleapis.com/maps/api/place/radarsearch/json?location=35.325153,-80.946239&radius=5000&name=YOURNAME&key=YOURKEY")
jsonAnsw<-content(res,"text")

then you could use library "jsonlite" to handle the result:

 myDataframe<- jsonlite::fromJSON(content(res,"text"))

Overfill

What does the name=YOURNAME part stands for? API name? – Pus

Also important: Notice: Radar Search is deprecated as of June 30, 2017. This feature will be turned down on June 30, 2018, and will no longer be available after that date. – Pus

Answer 3

You can use ggmap package with something like :

library('ggmap')
dataToMap = rbind(lon=longitude, lat=latitude)
geo.df = NULL
for(i in 1:length(dataToMap[,1]){
    location = geocode(dataToMap[1,],override_limit=TRUE,messaging=FALSE)
     geo.df = rbind(lon=location$lon[1], lat=location$lat[1])
    Sys.sleep(1.5)
}

Honeydew

though, it seems geocode() fromggmap uses the Google Map Geocode API instead of Google Maps Places API see here – Tersanctus

Answer 4

You can use newplaces which is a R wrapper package to the Google Places API (New API released in the end of 2023).

In your example you can run the following:

# Install package from github:
remotes::install_github("aTnT/newplaces")

library(newplaces)

# Set your API key:
set_api_key(key = "YOUR_MAPS_API_KEY")

# Get restaurants within a 500 m radius around a location:
myPlaces <- text_search(textQuery = "restaurants", location = "bias", circle_center_latitude = 40.75797, circle_center_longitude = -73.98554, circle_radius = 500)

You can access a tibble dataframe with the places at myPlaces$places:

myPlaces$places

# A tbl_json: 20 x 77 tibble with a "JSON" attribute
   ..JSON document.id name  id    nationalPhoneNumber internationalPhoneNu…¹ formattedAddress rating googleMapsUri
   <chr>        <int> <chr> <chr> <chr>               <chr>                  <chr>             <dbl> <chr>        
 1 "{\"n…           1 plac… ChIJ… [phone]      [phone]        200 W 44th St, …    4.5 https://maps…
 2 "{\"n…           2 plac… ChIJ… [phone]      [phone]        1515 Broadway @…    4.4 https://maps…
 3 "{\"n…           3 plac… ChIJ… <NA>                <NA>                   229 W 43rd St, …    4.8 https://maps…
 4 "{\"n…           4 plac… ChIJ… [phone]      [phone]        1501 Broadway, …    4.4 https://maps…
 5 "{\"n…           5 plac… ChIJ… [phone]      [phone]        1567 Broadway, …    4   https://maps…
 6 "{\"n…           6 plac… ChIJ… [phone]      [phone]        1440 Broadway, …    4.3 https://maps…
 7 "{\"n…           7 plac… ChIJ… [phone]      [phone]        2 Times Sq, New…    4.1 https://maps…
 8 "{\"n…           8 plac… ChIJ… [phone]      [phone]        326 W 46th St, …    4.4 https://maps…
 9 "{\"n…           9 plac… ChIJ… [phone]      [phone]        620 8th Ave, Ne…    4.2 https://maps…
10 "{\"n…          10 plac… ChIJ… [phone]      [phone]        136 W 46th St, …    4.6 https://maps…
11 "{\"n…          11 plac… ChIJ… [phone]      [phone]        575 7th Ave, Ne…    4.5 https://maps…
12 "{\"n…          12 plac… ChIJ… <NA>                <NA>                   135 W 50th St, …    3.9 https://maps…
13 "{\"n…          13 plac… ChIJ… [phone]      [phone]        132 W 43rd St, …    4.4 https://maps…
14 "{\"n…          14 plac… ChIJ… [phone]      [phone]        691 8th Ave, Ne…    4.4 https://maps…
15 "{\"n…          15 plac… ChIJ… [phone]      [phone]        67 W 44th St, N…    4.2 https://maps…
16 "{\"n…          16 plac… ChIJ… [phone]      [phone]        121 W 45th St, …    4.4 https://maps…
17 "{\"n…          17 plac… ChIJ… [phone]      [phone]        36 W 48th St, N…    4.2 https://maps…
18 "{\"n…          18 plac… ChIJ… [phone]      [phone]        155 W 43rd St, …    4.2 https://maps…
19 "{\"n…          19 plac… ChIJ… [phone]      [phone]        151 W 46th St, …    4.3 https://maps…
20 "{\"n…          20 plac… ChIJ… [phone]      [phone]        254 W 47th St, …    4.4 https://maps…
# ℹ abbreviated name: ¹​internationalPhoneNumber
# ℹ 68 more variables: websiteUri <chr>, utcOffsetMinutes <dbl>, adrFormatAddress <chr>, businessStatus <chr>,
#   priceLevel <chr>, userRatingCount <dbl>, iconMaskBaseUri <chr>, iconBackgroundColor <chr>, takeout <lgl>,
#   delivery <lgl>, dineIn <lgl>, curbsidePickup <lgl>, reservable <lgl>, servesBreakfast <lgl>,
#   servesLunch <lgl>, servesDinner <lgl>, servesBeer <lgl>, servesWine <lgl>, servesBrunch <lgl>,
#   servesVegetarianFood <lgl>, primaryType <chr>, shortFormattedAddress <chr>, outdoorSeating <lgl>,
#   liveMusic <lgl>, menuForChildren <lgl>, servesCocktails <lgl>, servesDessert <lgl>, servesCoffee <lgl>, …

You could also use the function argument includedType the specify a business type if your textQuery specifies something else.

More info at https://github.com/aTnT/newplaces

Burnell