【问题标题】:httr POST request to API returns 400 error对 API 的 httr POST 请求返回 400 错误
【发布时间】:2016-08-01 14:01:57
【问题描述】:

我正在使用 R 中的 httr 包来尝试查询 postcode.io API (http://postcodes.io/docs)。

我可以按照以下说明成功查询单个邮政编码: sample4 <- GET("api.postcodes.io/postcodes/EN14RF")

当我尝试查询多个邮政编码时,我变得有点不知所措。 postcode.io 说明建议

POST https://api.postcodes.io/postcodes?q=[postcode]

其中指定了包含邮政编码数组的 JSON 对象。我有一个包含邮政编码的 R 向量,我试图将其转换为 JSON 对象:

a <- toJSON(a)

我的 R 向量“a”在哪里:

structure(c(4L, 5L, 3L, 6L, 1L, 2L), .Label = c("Bn14 9aw", "CR0 4BE", "E5 8HB", "EN1 4RF", "G42 8QN", "SA1 3UL"), class = "factor")

现在,当我尝试使用以下代码行查询 API 时:

sample4 <- POST("https://api.postcodes.io/postcodes?q=[postcode]", body = list(postcode = add1JSON))

我收到一个错误:"Invalid JSON submitted. You need to submit a JSON object with an array of postcodes or geolocation objects"

我感觉这是因为我提供的不是数组而是未命名的列表,例如我的 JSON 对象应如下所示:

{"postcodes":"[ \"EN14RF\", \"G428QN\", \"E58HB\", \"SA13UL\", \"Bn149aw\", \"CR04BE\" ]"}

不是这个:"[ \"EN14RF\", \"G428QN\", \"E58HB\", \"SA13UL\", \"Bn149aw\", \"CR04BE\" ]"

谁能帮我解决这个问题?我感觉这与我的toJSON 通话有关,但在论坛或 API 开发人员页面上找不到类似的示例:(

非常感谢

马蒂

【问题讨论】:

    标签: json r api httr


    【解决方案1】:

    他们的 API 文档写得很糟糕。这似乎是这个特定 API 调用的意图:

    library(httr)
    library(jsonlite)
    library(dplyr)
    
    post_codes <- c("Bn14 9aw", "CR0 4BE", "E5 8HB", "EN1 4RF", "G42 8QN", "SA1 3UL")
    
    res <- POST("https://api.postcodes.io/postcodes",
                body=list(postcodes=post_codes),
                encode="json")
    
    
    status_code(res)
    ## [1] 200
    
    content(res, as="text") %>%
      fromJSON(flatten=TRUE) %>%
      glimpse()
    ## List of 2
    ##  $ status: int 200
    ##  $ result:'data.frame':  6 obs. of  29 variables:
    ##   ..$ query                            : chr [1:6] "Bn14 9aw" "E5 8HB" "CR0 4BE" "EN1 4RF" ...
    ##   ..$ result.postcode                  : chr [1:6] "BN14 9AW" "E5 8HB" "CR0 4BE" "EN1 4RF" ...
    ##   ..$ result.quality                   : int [1:6] 1 1 1 1 1 1
    ##   ..$ result.eastings                  : int [1:6] 514948 534934 531978 534957 264583 258092
    ##   ..$ result.northings                 : int [1:6] 104386 185332 164963 199610 192273 662417
    ##   ..$ result.country                   : chr [1:6] "England" "England" "England" "England" ...
    ##   ..$ result.nhs_ha                    : chr [1:6] "South East Coast" "London" "London" "London" ...
    ##   ..$ result.longitude                 : num [1:6] -0.3693 -0.0553 -0.1055 -0.0494 -3.9572 ...
    ##   ..$ result.latitude                  : num [1:6] 50.8 51.6 51.4 51.7 51.6 ...
    ##   ..$ result.parliamentary_constituency: chr [1:6] "East Worthing and Shoreham" "Hackney South and Shoreditch" "Croydon South" "Enfield North" ...
    ##   ..$ result.european_electoral_region : chr [1:6] "South East" "London" "London" "London" ...
    ##   ..$ result.primary_care_trust        : chr [1:6] "West Sussex" "City and Hackney Teaching" "Croydon" "Enfield" ...
    ##   ..$ result.region                    : chr [1:6] "South East" "London" "London" "London" ...
    ##   ..$ result.lsoa                      : chr [1:6] "Worthing 008C" "Hackney 017B" "Croydon 024D" "Enfield 002E" ...
    ##   ..$ result.msoa                      : chr [1:6] "Worthing 008" "Hackney 017" "Croydon 024" "Enfield 002" ...
    ##   ..$ result.incode                    : chr [1:6] "9AW" "8HB" "4BE" "4RF" ...
    ##   ..$ result.outcode                   : chr [1:6] "BN14" "E5" "CR0" "EN1" ...
    ##   ..$ result.admin_district            : chr [1:6] "Worthing" "Hackney" "Croydon" "Enfield" ...
    ##   ..$ result.parish                    : chr [1:6] "Worthing, unparished area" "Hackney, unparished area" "Croydon, unparished area" "Enfield, unparished area" ...
    ##   ..$ result.admin_county              : chr [1:6] "West Sussex" NA NA NA ...
    ##   ..$ result.admin_ward                : chr [1:6] "Gaisford" "Homerton" "Waddon" "Turkey Street" ...
    ##   ..$ result.ccg                       : chr [1:6] NA "NHS City and Hackney" "NHS Croydon" "NHS Enfield" ...
    ##   ..$ result.nuts                      : chr [1:6] "West Sussex (South West)" "Hackney and Newham" "Croydon" "Enfield" ...
    ##   ..$ result.codes.admin_district      : chr [1:6] "E07000229" "E09000012" "E09000008" "E09000010" ...
    ##   ..$ result.codes.admin_county        : chr [1:6] "E10000032" "E99999999" "E99999999" "E99999999" ...
    ##   ..$ result.codes.admin_ward          : chr [1:6] "E05007698" "E05009376" "E05000167" "E05000211" ...
    ##   ..$ result.codes.parish              : chr [1:6] "E43000150" "E43000202" "E43000198" "E43000200" ...
    ##   ..$ result.codes.ccg                 : chr [1:6] "E38000213" "E38000035" "E38000040" "E38000057" ...
    ##   ..$ result.codes.nuts                : chr [1:6] "UKJ27" "UKI41" "UKI62" "UKI54" ...
    

    【讨论】:

    • 对于编写不佳的 API 文档,我深表歉意。我已经修改了它们 - 特别是提到 POST 请求应该具有内容类型application/json
    【解决方案2】:

    根据 httr 包的documentation,您需要将encode 参数设置为json,如下所示:

    url <- "https://api.postcodes.io/postcodes"
    body <- list(a = 1, b = 2, c = 3)
    
    # JSON encoded
    r <- POST(url, body = body, encode = "json")
    

    要查看实际请求,您可以使用交互式控制台中的verbose() 参数:

    POST(url, body = body, encode = "json", verbose())
    

    【讨论】:

      【解决方案3】:

      你可以把任何 JSON 数组变成可以像这样通过 httr 的 R 格式:

      body <- fromJSON("{"postcodes":"[ \"EN14RF\", \"G428QN\", \"E58HB\", \"SA13UL\", \"Bn149aw\", \"CR04BE\" ]"}")
      

      然后发帖:

      sample4 <- POST("https://api.postcodes.io/postcodes?q=[postcode]", 
      body = body, 
      encode = "json")
      

      【讨论】:

        猜你喜欢
        • 1970-01-01
        • 1970-01-01
        • 1970-01-01
        • 2015-06-02
        • 1970-01-01
        • 1970-01-01
        • 1970-01-01
        • 1970-01-01
        • 1970-01-01
        相关资源
        最近更新 更多