Join GitHub today
GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software together.
Sign upgeocode(): Toggle geographic columns returned #58
Assignees
Labels
Comments
|
Thank you for reporting. The reason for this behavior of library(hereR)
set_verbose(TRUE)
addr_1 <- "boston, massachusetts"
addr_2 <- "massachusetts"
url_1 <- geocode(addr_1, url_only = TRUE)
url_2 <- geocode(addr_2, url_only = TRUE)
json_1 <- hereR:::.get_content(url_1)
#> Sending 1 request(s) to: 'https://geocoder.ls.hereapi.com/6.2/geocode.json?...'
#> Received 1 response(s) with total size: 1.2 Kb
json_2 <- hereR:::.get_content(url_2)
#> Sending 1 request(s) to: 'https://geocoder.ls.hereapi.com/6.2/geocode.json?...'
#> Received 1 response(s) with total size: 1.1 KbSee the First responsejsonlite::prettify(json_1)
#> {
#> "Response": {
#> "MetaInfo": {
#> "Timestamp": "2020-08-26T18:04:05.229+0000"
#> },
#> "View": [
#> {
#> "_type": "SearchResultsViewType",
#> "ViewId": 0,
#> "Result": [
#> {
#> "Relevance": 1.0,
#> "MatchLevel": "city",
#> "MatchQuality": {
#> "State": 1.0,
#> "City": 1.0
#> },
#> "Location": {
#> "LocationId": "NT_0iuf-KHj2X8GfdEYL.GQXD",
#> "LocationType": "point",
#> "DisplayPosition": {
#> "Latitude": 42.35866,
#> "Longitude": -71.05675
#> },
#> "NavigationPosition": [
#> {
#> "Latitude": 42.35866,
#> "Longitude": -71.05675
#> }
#> ],
#> "MapView": {
#> "TopLeft": {
#> "Latitude": 42.39726,
#> "Longitude": -71.19121
#> },
#> "BottomRight": {
#> "Latitude": 42.22786,
#> "Longitude": -70.80852
#> }
#> },
#> "Address": {
#> "Label": "Boston, MA, United States",
#> "Country": "USA",
#> "State": "MA",
#> "County": "Suffolk",
#> "City": "Boston",
#> "PostalCode": "02109",
#> "AdditionalData": [
#> {
#> "value": "United States",
#> "key": "CountryName"
#> },
#> {
#> "value": "Massachusetts",
#> "key": "StateName"
#> },
#> {
#> "value": "Suffolk",
#> "key": "CountyName"
#> },
#> {
#> "value": "N",
#> "key": "PostalCodeType"
#> }
#> ]
#> }
#> }
#> }
#> ]
#> }
#> ]
#> }
#> }
#> Second responsejsonlite::prettify(json_2)
#> {
#> "Response": {
#> "MetaInfo": {
#> "Timestamp": "2020-08-26T18:04:05.243+0000"
#> },
#> "View": [
#> {
#> "_type": "SearchResultsViewType",
#> "ViewId": 0,
#> "Result": [
#> {
#> "Relevance": 1.0,
#> "MatchLevel": "state",
#> "MatchQuality": {
#> "State": 1.0
#> },
#> "Location": {
#> "LocationId": "NT_xsuPM8XX3.ynKq8rCjhCcB",
#> "LocationType": "point",
#> "DisplayPosition": {
#> "Latitude": 42.35866,
#> "Longitude": -71.05675
#> },
#> "NavigationPosition": [
#> {
#> "Latitude": 42.35866,
#> "Longitude": -71.05675
#> }
#> ],
#> "MapView": {
#> "TopLeft": {
#> "Latitude": 42.88677,
#> "Longitude": -73.50814
#> },
#> "BottomRight": {
#> "Latitude": 41.13234,
#> "Longitude": -69.92823
#> }
#> },
#> "Address": {
#> "Label": "MA, United States",
#> "Country": "USA",
#> "State": "MA",
#> "AdditionalData": [
#> {
#> "value": "United States",
#> "key": "CountryName"
#> },
#> {
#> "value": "Massachusetts",
#> "key": "StateName"
#> }
#> ]
#> }
#> }
#> }
#> ]
#> }
#> ]
#> }
#> }
#> A quick fix for your automated workflow could be data.table::rbindlist(
list(
geocode(addr_1, sf = FALSE),
geocode(addr_2, sf = FALSE)
),
fill = TRUE
)
#> Sending 1 request(s) to: 'https://geocoder.ls.hereapi.com/6.2/geocode.json?...'
#> Received 1 response(s) with total size: 1.2 Kb
#> Sending 1 request(s) to: 'https://geocoder.ls.hereapi.com/6.2/geocode.json?...'
#> Received 1 response(s) with total size: 1.1 Kb
#> id address postalCode city county state country type
#> 1: 1 Boston, MA, United States 02109 Boston Suffolk MA USA point
#> 2: 1 MA, United States <NA> <NA> <NA> MA USA point
#> lng lat
#> 1: -71.05675 42.35866
#> 2: -71.05675 42.35866I will consider adding consistent columns in the return value of Session info |
|
Thank you for such a detailed response! I came across |
|
The columns are now consistently the same, regardless of the level of the input address. library(hereR)
set_verbose(TRUE)
geocode("boston, massachusetts", sf = FALSE)
#> Sending 1 request(s) to: 'https://geocode.search.hereapi.com/v1/geocode?...'
#> Received 1 response(s) with total size: 896 bytes
#> id address type street house_number postal_code
#> 1 1 Boston, MA, United States locality <NA> <NA> 02109
#> district city county state country lng_access lat_access
#> 1 <NA> Boston Suffolk Massachusetts United States NA NA
#> lng_position lat_position
#> 1 -71.05674 42.35866
geocode("massachusetts", sf = FALSE)
#> Sending 1 request(s) to: 'https://geocode.search.hereapi.com/v1/geocode?...'
#> Received 1 response(s) with total size: 832 bytes
#> id address type street house_number postal_code
#> 1 1 MA, United States administrativeArea <NA> <NA> <NA>
#> district city county state country lng_access lat_access
#> 1 <NA> <NA> <NA> Massachusetts United States NA NA
#> lng_position lat_position
#> 1 -71.05674 42.35866To check, install the development version from GitHub: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I've noticed that depending on the specificity of the address sent to
hereR::geocode(), different columns are returned in the resulting sf/data frame object.Here's a reproducible example (assuming you have the geocoding API set up):
Note that because
vec2could not be geocoded more precisely than a US state, additional columns (city, county, postalCode) don't show up.I can understand why this behavior would be desired: it prevents the user from getting back a needlessly-large data frame with a huge number of NA values. But I'm trying to write an automated workflow where I'll be geocoding addresses at varying levels of specificity. I need to know ahead of time what columns will be present in my output data frame, so that I can write functions to operate on it without first checking whether the county/state/city columns exist.
Is there currently a way to toggle the inclusion of additional columns in
geocode()? If not, would you consider adding this feature? Maybe an argumentshow_all_colsthat defaults to FALSE, or something of that nature.