如何根据特定值过滤列表并取消嵌套？

Question

我正在尝试为我的 Github 存储库创建一个包含所有目录 URL 的数据框。这是我写的：

library(httr)
  req <- GET("https://api.github.com/repos/thedivtagguy/daily-data/git/trees/master")

在 urls 中，我希望只存储 type: tree 中那些项目的 URL。这是 req 的样子：

{
  "sha": "e5acaf1fd8973e010922ffc4366af68359de8456",
  "url": "https://api.github.com/repos/thedivtagguy/daily-data/git/trees/e5acaf1fd8973e010922ffc4366af68359de8456",
  "tree": [
    {
      "path": ".gitignore",
      "mode": "100644",
      "type": "blob",
      "sha": "aaf221658979cc888d499702eea62beafa52d4e4",
      "size": 631,
      "url": "https://api.github.com/repos/thedivtagguy/daily-data/git/blobs/aaf221658979cc888d499702eea62beafa52d4e4"
    },
    {
      "path": "README.md",
      "mode": "100644",
      "type": "blob",
      "sha": "2b0515e164c8a76d877324984f01e4bde6725410",
      "size": 1119,
      "url": "https://api.github.com/repos/thedivtagguy/daily-data/git/blobs/2b0515e164c8a76d877324984f01e4bde6725410"
    },
    {
      "path": "_config.yml",
      "mode": "100644",
      "type": "blob",
      "sha": "2f7efbeab578c8042531ea7908ee8ffd7589fe46",
      "size": 27,
      "url": "https://api.github.com/repos/thedivtagguy/daily-data/git/blobs/2f7efbeab578c8042531ea7908ee8ffd7589fe46"
    },
    {
      "path": "byrne_pebbles_8.png",
      "mode": "100644",
      "type": "blob",
      "sha": "95d9552636489a812d164e63776892c91c859785",
      "size": 10862,
      "url": "https://api.github.com/repos/thedivtagguy/daily-data/git/blobs/95d9552636489a812d164e63776892c91c859785"
    },
    {
      "path": "dd01_kharifAndRabiCrops",
      "mode": "040000",
      "type": "tree",
      "sha": "b0863850cf04b73f76a8ed1f60558c6d340d142a",
      "url": "https://api.github.com/repos/thedivtagguy/daily-data/git/trees/b0863850cf04b73f76a8ed1f60558c6d340d142a"
    }

这是我写的：

  reponse <- content(req)$tree
  urls <- response %>%
    map( ~ ., ~ filter(.x, type == "tree")) %>%
    unnest(url)

但这不起作用，我收到此错误：

Error in UseMethod("unnest") : 
  no applicable method for 'unnest' applied to an object of class "list"

如何过滤以便我只能存储类型为 tree 的项目的 URL？我知道如何在 base R 中做到这一点，但我更喜欢整洁的方法。

Answer 1

我们可以遍历 'response' 嵌套的 list，并通过与 bind_cols 绑定将命名的内部列表转换为 tibble - 如果我们使用 map 循环，它将 return list of tibble，通过添加后缀 _dfr，它将 lists of tibbles 绑定到单个 tibble 输出.然后，在 'type' 列

上执行 filtering

library(purrr)
library(dplyr)
map_dfr(response, bind_cols) %>% 
    filter(type == 'tree')

-输出

# A tibble: 8 × 6
  path                    mode   type  sha                                       size url                                   
  <chr>                   <chr>  <chr> <chr>                                    <int> <chr>                                 
1 dd01_kharifAndRabiCrops 040000 tree  b0863850cf04b73f76a8ed1f60558c6d340d142a    NA https://api.github.com/repos/thedivta…
2 dd02_commonSenseMedia   040000 tree  93461ea8897a3dcd026a88e62a284ba4a835a219    NA https://api.github.com/repos/thedivta…
3 dd03_cropYieldsD3       040000 tree  63b5de06defae6e8524e40d7942ba61ed2a599fc    NA https://api.github.com/repos/thedivta…
4 dd04_digitsPi           040000 tree  0f7e99e051a41fef6cde2706e6ded7538c3b6f8f    NA https://api.github.com/repos/thedivta…
5 dd05_indiaR             040000 tree  2bd552a7f9918602df043d54130c3d0b33d27294    NA https://api.github.com/repos/thedivta…
6 dd06_ttDrWho            040000 tree  7a0e0461228e0b39ba202032e09b46ba49e7eab6    NA https://api.github.com/repos/thedivta…
7 dd07_ggWaves            040000 tree  06def68fecdf576d45efa16ba6d98bba27edcc8f    NA https://api.github.com/repos/thedivta…
8 resources               040000 tree  ed4cb9ff46f19764292dea8e655dd4a500389dda    NA https://api.github.com/repos/thedivta…

如何根据特定值过滤列表并取消嵌套？

How do I filter a list based on a certain value and unnest it?

r

dplyr

purrr