从R中的分组数据帧重塑JSON输出

Question

I have an R dataframe of the form:

Country Region  Year    V1  V2
AAAA    XXXX    2001    12  13
BBBB    YYYY    2001    14  15
AAAA    XXXX    2002    36  56
AAAA    XXXX    1999    45  67

and would like to generate a JSON equivalent of the form:

[
  {"Country": "AAAA",
   "Region":"XXXX",
    "V1": [ [1999,45], [2001,12] , [2002,36] ],
    "V2":[ [1999,67], [2001,13] , [2002,56] ]
  },
  {"Country": "BBBB",
   "Region":"YYYY",
   "V1":[ [2001,14] ],
   "V2":[ [2001,15] ]
  }
]

I'm imagining this requires:

1. grouping by Country and Region
2. sorting by Year within the groups
3. for each of the remaining columns Vx in V1, V2 (an arbitrary number of columns which may have arbitrary names), generating a list that contains list elements [Year, Vx], for ordered Year

but am struggling to find a way to do it?

Answer 1

Here is another way to do this.

dat <- read.table(textConnection("Country Region  Year    V1  V2
AAAA    XXXX    2001    12  13
BBBB    YYYY    2001    14  15
AAAA    XXXX    2002    36  56
AAAA    XXXX    1999    45  67"), header = TRUE)

We add two helper functions to zip vectors together and a custom sort function which sorts a list by elements in a given position.

#' Pluck element
pluck_ = function (element){
  function(x) x[[element]]
}

#' Zip two vectors
zip_ <- function(..., names = F){
  x = list(...)
  y = lapply(seq_along(x[[1]]), function(i) lapply(x, pluck_(i)))
  if (names) names(y) = seq_along(y)
  return(y)
}

#' Sort a vector based on elements at a given position
sort_ <- function(v, i = 1){
  v[sort(sapply(v, '[[', i), index.return = T)$ix]
}

Time to put things together and use the split-apply-combine magic to get the output you seek.

library(plyr)
dat2 <- dlply(dat, .(Country, Region), function(d){
  list(
    Country = d$Country[1],
    Region = d$Region[1],
    V1 = sort_(zip_(d$Year, d$V1)),
    V2 = sort_(zip_(d$Year, d$V2))
  )
})
cat(rjson::toJSON(setNames(dat2, NULL)))

This gives you the output

[
  {"Country":"AAAA",
   "Region":"XXXX",
   "V1":[[1999,45],[2001,12],[2002,36]],
   "V2":[[1999,67],[2001,13],[2002,56]]
  },
  {"Country":"BBBB",
   "Region":"YYYY",
   "V1":[[2001,14]],
   "V2":[[2001,15]]
  }
]

Answer 2

Here's a sort of messy function to do this (you could easily add sorting by year of the V1 and V2 arrays):

dat <- read.table(textConnection(
'Country Region  Year    V1  V2
AAAA    XXXX    2001    12  13
BBBB    YYYY    2001    14  15
AAAA    XXXX    2002    36  56
AAAA    XXXX    1999    45  67'
), header=TRUE, stringsAsFactors=FALSE)

library(plyr); library(RJSONIO)
myfunc <- function(nn)
{
  tt <- split(nn, nn$Country)
  bar <- function(w){
    foo <- function(x, y, z) paste(x[y], x[z], sep=",")
    V1 <- as.character(apply(w, 1, foo, y="Year", z="V1"))
    V2 <- as.character(apply(w, 1, foo, y="Year", z="V2"))
    datlist <- list(Country = unique(w$Country), 
                    Region = unique(w$Region), 
                    V1 = V1, V2=V2)
  }
  datlist <- lapply(tt, bar)
  names(datlist) <- NULL
  RJSONIO::toJSON(datlist)
}

cat(myfunc(dat))

[
 {
   "Country": "AAAA",
   "Region": "XXXX",
   "V1": [ "2001,12", "2002,36", "1999,45" ],
   "V2": [ "2001,13", "2002,56", "1999,67" ] 
 },
 {
   "Country": "BBBB",
   "Region": "YYYY",
   "V1": "2001,14",
   "V2": "2001,15" 
  } 
]