将嵌套数据框展开到父级 [英] Expand nested dataframe into parent

问题描述

我有一个数据帧嵌套在我从 Mongo 获取的数据帧中.每行的行数匹配，因此在查看时它看起来像一个典型的数据帧.我的问题是，如何将嵌套的数据框扩展到父级中，以便我可以运行 dplyr 选择?看下面的布局

I have a dataframe nested within a dataframe that I'm getting from Mongo. The number of rows match in each so that when viewed it looks like a typical dataframe. My question, how do I expand the nested dataframe into the parent so that I can run dplyr selects? See the layout below

'data.frame':   10 obs. of  2 variables:
 $ _id         : int  1551 1033 1061 1262 1032 1896 1080 1099 1679 1690
 $ personalInfo:'data.frame':   10 obs. of  2 variables:
  ..$ FirstName     :List of 10
  .. ..$ : chr "Jack"
  .. ..$ : chr "Yogesh"
  .. ..$ : chr "Steven"
  .. ..$ : chr "Richard"
  .. ..$ : chr "Thomas"
  .. ..$ : chr "Craig"
  .. ..$ : chr "David"
  .. ..$ : chr "Aman"
  .. ..$ : chr "Frank"
  .. ..$ : chr "Robert"
  ..$ MiddleName    :List of 10
  .. ..$ : chr "B"
  .. ..$ : NULL
  .. ..$ : chr "J"
  .. ..$ : chr "I"
  .. ..$ : chr "E"
  .. ..$ : chr "A"
  .. ..$ : chr "R"
  .. ..$ : NULL
  .. ..$ : chr "J"
  .. ..$ : chr "E"

根据建议，以下是您重新创建数据的方法

As per suggestion, here's how you recreate the data

id <- c(1551, 1033, 1061, 1262, 1032, 1896, 1080, 1099, 1679, 1690)
fname <- list("Jack","Yogesh","Steven","Richard","Thomas","Craig","David","Aman","Frank","Robert")
mname <- list("B",NULL,"J","I","E","A","R",NULL,"J","E")

sub <- as.data.frame(cbind(fname, mname))
master <- as.data.frame(id)
master$personalInfo <- sub

推荐答案

虽然@akrun 的答案可能更实用，也可能是整理数据的方式，但我认为此输出更接近于您所描述的内容.

While @akrun's answer is probably more practical and probably the way to tidy your data, I think this output is closer to what you describe.

我创建了一个新环境，我在其中放置了 data.frame 的内容，在那里我将有问题的列的内容取消列出到所述环境中，最后我将其全部包装回 <代码>数据框.

I create a new environment where I put the data.frame's content, there I unlist to the said environment the content of your problematic column, and finally I wrap it all back into a data.frame.

我对 cbind 使用了一个奇怪的 hack，因为 as.data.frame 对列表列很烦人.但是使用 tibble::as_tibble 效果很好.

I use a strange hack with cbind as as.data.frame is annoying with list columns. Using tibble::as_tibble works fine however.

new_env <- new.env()
list2env(master,new_env)
list2env(new_env$personalInfo,new_env)
rm(personalInfo,envir = new_env)
res <- as.data.frame(do.call(cbind,as.list(new_env))) # or as_tibble(as.list(new_env))
rm(new_env)
res
#      fname   id mname
# 1     Jack 1551     B
# 2   Yogesh 1033  NULL
# 3   Steven 1061     J
# 4  Richard 1262     I
# 5   Thomas 1032     E
# 6    Craig 1896     A
# 7    David 1080     R
# 8     Aman 1099  NULL
# 9    Frank 1679     J
# 10  Robert 1690     E

str(res)
# 'data.frame': 10 obs. of  3 variables:
#   $ fname:List of 10
# ..$ : chr "Jack"
# ..$ : chr "Yogesh"
# ..$ : chr "Steven"
# ..$ : chr "Richard"
# ..$ : chr "Thomas"
# ..$ : chr "Craig"
# ..$ : chr "David"
# ..$ : chr "Aman"
# ..$ : chr "Frank"
# ..$ : chr "Robert"
# $ id   :List of 10
# ..$ : num 1551
# ..$ : num 1033
# ..$ : num 1061
# ..$ : num 1262
# ..$ : num 1032
# ..$ : num 1896
# ..$ : num 1080
# ..$ : num 1099
# ..$ : num 1679
# ..$ : num 1690
# $ mname:List of 10
# ..$ : chr "B"
# ..$ : NULL
# ..$ : chr "J"
# ..$ : chr "I"
# ..$ : chr "E"
# ..$ : chr "A"
# ..$ : chr "R"
# ..$ : NULL
# ..$ : chr "J"
# ..$ : chr "E"

这篇关于将嵌套数据框展开到父级的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！