时间戳

首页 > 解决方案 > 来自 R API 调用的重复行

r - 来自 R API 调用的重复行

问题描述

对于 NPPES API 的每个请求，我都会返回两行，我很难理解为什么。

我可以使用 dplyr 选择不同的，但我宁愿更好地理解为什么我得到重复的行而不是使用解决方法。您的帮助将不胜感激。

我的代码如下所示：

pacman::p_load(tidyverse,httr,jsonlite,purrr)

path <- "https://npiregistry.cms.hhs.gov/api/?"

# CREATE A FUNCTION TO QUERY THE NPPES NPI REGISTRY
getNPI <- function(object) {
  request <- httr::GET(url = path,
                       query = list(version = "2.0",
                         number = object))

  df <- content(request, as = "text", encoding = "UTF-8") %>%
    jsonlite::fromJSON(., flatten = TRUE) %>%
    data.frame() %>% 
    tidyr::unnest(c(results.addresses,results.taxonomies), names_repair = "unique")

  df_col_names <- names(df)
  cols_to_add <- setdiff(c("result_count",
                           "results.enumeration_type",
                           "results.number",
                           "results.last_updated_epoch",
                           "results.created_epoch",
                           "results.other_names",
                           "country_code",
                           "country_name",
                           "address_purpose",
                           "address_type",
                           "address_1",
                           "address_2",
                           "city",
                           "state",
                           "postal_code",
                           "telephone_number",
                           "code",
                           "desc",
                           "primary",
                           "state1",
                           "license",
                           "results.identifiers",
                           "results.basic.first_name",
                           "results.basic.last_name",
                           "results.basic.middle_name",
                           "results.basic.credential",
                           "results.basic.sole_proprietor",
                           "results.basic.gender",
                           "results.basic.enumeration_date",
                           "results.basic.last_updated",
                           "results.basic.status",
                           "results.basic.name"), df_col_names)
  if (length(cols_to_add) > 0) {
      for(i in cols_to_add){
        df[,i] <- "UNKNOWN"
      }
   }

  # df_col_names <- names(df$results.taxonomies[[1]])
  # cols_to_add <- setdiff(c("desc","state"), df_col_names)
  #     if (length(cols_to_add) > 0) {
  #     for(i in cols_to_add){
  #       df$results.taxonomies[[1]][,i] <- "UNKNOWN"
  #     }
  #  }

  df %>%
    select(results.number,
           results.basic.name,
           results.enumeration_type,
           results.basic.gender,
           results.basic.credential,
           desc) %>%
    rename(Provider_NPI = results.number,
           Provider_Name = results.basic.name,
           Provider_Gender = results.basic.gender,
           Provider_Credentials = results.basic.credential,
           Provider_Taxonomy = desc) %>%
    mutate(
      Provider_Type = case_when(
        results.enumeration_type == "NPI-1" ~ 'Individual Provider',
        results.enumeration_type == "NPI-2" ~ 'Organizational Provider'
      )
    ) %>%
    select(-results.enumeration_type)
}

providerIDs <- c(
                 #'1477765634',
                 '1376815795'
                 )

test <- lapply(providerIDs, getNPI)

final_df <- do.call("rbind",test)

标签： rjsondplyr

解决方案

如果我们一步一步地这样做，“data.frame”的“results.taxonomies”、“results.addresses”分别有 1 行和 2 行

out <- content(request, as = "text", encoding = "UTF-8") %>%
     jsonlite::fromJSON(., flatten = TRUE) %>%
    data.frame()
str(out)
#...
# $ results.addresses             :List of 1
#  ..$ :'data.frame':   2 obs. of  10 variables:  # 2 row
#  .. ..$ country_code    : chr  "US" "US"
#  .. ..$ country_name    : chr  "United States" "United States"
#  .. ..$ address_purpose : chr  "LOCATION" "MAILING"
#  .. ..$ address_type    : chr  "DOM" "DOM"
#  .. ..$ address_1       : chr  "1122 BROADWAY" "1122 BROADWAY"
#  .. ..$ address_2       : chr  "" ""
#  .. ..$ city            : chr  "WOODMERE" "WOODMERE"
#  .. ..$ state           : chr  "NY" "NY"
#  .. ..$ postal_code     : chr  "115981242" "115981242"
#  .. ..$ telephone_number: chr  "516-295-3838" "516-295-3838"
# $ results.taxonomies            :List of 1
#  ..$ :'data.frame':   1 obs. of  5 variables:   # one row
#  .. ..$ code   : chr "363A00000X"
#  .. ..$ desc   : chr "Physician Assistant"
#  .. ..$ primary: logi TRUE
#  .. ..$ state  : chr "NY"
#  .. ..$ license: chr "013595-1"
# ...

在unnest两个数据集上，它产生两行

out1 <- out %>% 
            tidyr::unnest(c(results.addresses,results.taxonomies), names_repair = "unique")
dim(out1)
#[1]  2 32

但是，列中存在一些差异，例如

out1 %>% 
    select(address_purpose)
# A tibble: 2 x 1
#  address_purpose
#   <chr>          
#1 LOCATION       
#2 MAILING

推荐阅读