我需要处理这个数据框，其中包含数据框内两列嵌套且不等排的层，我需要广泛扩展这些列中可能的内容。
jsonblob1 = c("{\"basestep_id\":\"BSa7pt5o7xx8dxaderflht3o\",\"step_act\":\"SET_NEW\",\"step_count\":0,\"custm\":false,\"edited\":false,\"name\":null,\"text\":\"Lorem ipsum dolor sit.\",\"crits\":[\"Lorem ipsum dolor sit.\",\"Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019.\",\"Lorem ipsum dolor sit.\",\"Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019 Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit..\\u2019\",\"Lorem ipsum dolor sit.\"],\"tasks\":[]}","{\"basestep_id\":\"BSa7pt5o7xx8dxaderflht3o\",\"step_act\":\"SET_NEW\",\"step_count\":0,\"custm\":false,\"edited\":false,\"name\":null,\"text\":\"Lorem ipsum dolor sit.\",\"crits\":[\"Lorem ipsum dolor sit.\",\"Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019.\",\"Lorem ipsum dolor sit.\",\"Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019 Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit..\\u2019\",\"Lorem ipsum dolor sit.\"],\"tasks\":[]}")
jsonblob2 = c("{\"basestep_id\":\"BSa7pt5o7xx8dxaderflht3o\",\"step_act\":\"SET_NEW\",\"step_count\":0,\"custm\":false,\"edited\":false,\"name\":null,\"text\":\"Lorem ipsum dolor sit.\",\"crits\":[\"Lorem ipsum dolor sit.\",\"Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019.\",\"Lorem ipsum dolor sit.\",\"Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019 Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit..\\u2019\",\"Lorem ipsum dolor sit.\"],\"tasks\":[]}","{\"basestep_id\":\"BSa7pt5o7xx8dxaderflht3o\",\"step_act\":\"SET_NEW\",\"step_count\":0,\"custm\":false,\"edited\":false,\"name\":null,\"text\":\"Lorem ipsum dolor sit.\",\"crits\":[\"Lorem ipsum dolor sit.\",\"Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019.\",\"Lorem ipsum dolor sit.\",\"Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019 Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit..\\u2019\",\"Lorem ipsum dolor sit.\"],\"tasks\":[\"Lorem ipsum dolor sit.\",\"Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019.\",\"Lorem ipsum dolor sit.\",\"Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit.\\u2019 Lorem ipsum dolor sit. \\u2018Lorem ipsum dolor sit..\\u2019\",\"Lorem ipsum dolor sit.\"]}")
otherc1 = c(1,2)
otherc2 = c(2,5)
df = data.frame(jsonblob1,jsonblob2,otherc1,otherc2)
到目前为止，我已经尝试过这段代码，我之前曾用它来扩展其他多层 JSON
df |>
mutate(
jsonblob1 = map(jsonblob1, ~ jsonlite::fromJSON(.x) |>
unnest_wider(everything()) |>
unnest_wider(everything(), names_sep = "_") |>
mutate(across(everything(), as.character)))) |>
mutate(
jsonblob2 = map(jsonblob2, ~ jsonlite::fromJSON(.x) |>
unnest_wider(everything()) |>
unnest_wider(everything(), names_sep = "_") |>
mutate(across(everything(), as.character))))
unnest(jsonblob1,jsonblob2) |>
type_convert()
但它返回此错误，我无法修复它。
Error in `mutate()`:
ℹ In argument: `jsonblob1 = map(...)`.
Caused by error in `map()`:
ℹ In index: 1.
Caused by error in `unnest_wider()`:
! `data` must be a data frame, not a list.
Run `rlang::last_trace()` to see where the error occurred.
为了确保目标明确，我想尽可能将包含单行的所有图层扩展到同一行，以及那些具有多行的元素（例如示例中的
critsjsonblobjsonblob2我很感谢您的帮助。
也许首先考虑转向更长的时间。
library(purrr)
library(dplyr)
library(tidyr)
as_tibble(df)
#> # A tibble: 2 × 4
#> jsonblob1 jsonblob2 otherc1 otherc2
#> <chr> <chr> <dbl> <dbl>
#> 1 "{\"basestep_id\":\"BSa7pt5o7xx8dxaderflht3o\",\"st… "{\"base… 1 2
#> 2 "{\"basestep_id\":\"BSa7pt5o7xx8dxaderflht3o\",\"st… "{\"base… 2 5
flat <- df |>
pivot_longer(starts_with("json"), names_to = "blob_id", values_to = "json") |>
mutate(json = map(json, jsonlite::parse_json)) |>
unnest_wider(json) |>
unnest_longer(crits) |>
unnest_longer(tasks, keep_empty = TRUE)
flat
#> # A tibble: 40 × 12
#> otherc1 otherc2 blob_id basestep_id step_act step_count custm edited name
#> <dbl> <dbl> <chr> <chr> <chr> <int> <lgl> <lgl> <lgl>
#> 1 1 2 jsonblob1 BSa7pt5o7xx… SET_NEW 0 FALSE FALSE NA
#> 2 1 2 jsonblob1 BSa7pt5o7xx… SET_NEW 0 FALSE FALSE NA
#> 3 1 2 jsonblob1 BSa7pt5o7xx… SET_NEW 0 FALSE FALSE NA
#> 4 1 2 jsonblob1 BSa7pt5o7xx… SET_NEW 0 FALSE FALSE NA
#> 5 1 2 jsonblob1 BSa7pt5o7xx… SET_NEW 0 FALSE FALSE NA
#> 6 1 2 jsonblob2 BSa7pt5o7xx… SET_NEW 0 FALSE FALSE NA
#> 7 1 2 jsonblob2 BSa7pt5o7xx… SET_NEW 0 FALSE FALSE NA
#> 8 1 2 jsonblob2 BSa7pt5o7xx… SET_NEW 0 FALSE FALSE NA
#> 9 1 2 jsonblob2 BSa7pt5o7xx… SET_NEW 0 FALSE FALSE NA
#> 10 1 2 jsonblob2 BSa7pt5o7xx… SET_NEW 0 FALSE FALSE NA
#> # ℹ 30 more rows
#> # ℹ 3 more variables: text <chr>, crits <chr>, tasks <chr>
glimpse(flat)
#> Rows: 40
#> Columns: 12
#> $ otherc1 <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2…
#> $ otherc2 <dbl> 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5…
#> $ blob_id <chr> "jsonblob1", "jsonblob1", "jsonblob1", "jsonblob1", "jsonb…
#> $ basestep_id <chr> "BSa7pt5o7xx8dxaderflht3o", "BSa7pt5o7xx8dxaderflht3o", "B…
#> $ step_act <chr> "SET_NEW", "SET_NEW", "SET_NEW", "SET_NEW", "SET_NEW", "SE…
#> $ step_count <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
#> $ custm <lgl> FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FA…
#> $ edited <lgl> FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FA…
#> $ name <lgl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA…
#> $ text <chr> "Lorem ipsum dolor sit.", "Lorem ipsum dolor sit.", "Lorem…
#> $ crits <chr> "Lorem ipsum dolor sit.", "Lorem ipsum dolor sit. ‘Lorem i…
#> $ tasks <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA…
创建于 2024-05-01，使用 reprex v2.1.0