R help - Jan 2012 - turning a list of vectors into a data.frame (as rows of the DF)?

As a newer R practicioner, it seems I stump myself weekly (at least) with issues
that have spinning my wheels.  Here is yet another... I'm trying to turn a
list of numeric vectors (of uneual length) inot a dataframe.  Each vector held
in the list represents a row, and there are some rows of unequal length.  I
would like NAs as placeholders for "missing" data in the shorter
vectors.  I think I'm missing something quite basic.
 
v1 <- c(1,2,3,4)
v2 <- c(1,2)
lst1 <- list(v1,v2)
 
Of course there is the intuitive:
 
as.data.frame(lst1)
 
However, the recycling rule (expectedly) reclycles 1,2 versus using NAs as
placeholders.
 
Then, looking into Teetor's R Cookbook, there is a piece of code that looked
(from the description) like it might do the trick:
 
do.call(rbind, Map(as.data.frame,lst1)
 
But I get the error:
Error in match.names(clabs, names(xi)) : 
  names do not match previous names
 
Thinking the source of the error had to do with the vectors of unequal lenght, I
tried Hadley's rbind.fill thusly:
 
library(reshape)
do.call(rbind.fill, Map(as.data.frame,lst1)
 
Which produced a dataset, but gain, not in the desired format.
 
Thanks in advance to anyone that can bring my frustrations to end!
C
	[[alternative HTML version deleted]]

Most methods take the rows of data.frame()s to be very significant
(indicating multiple values from a single observation) so what you're
doing seems like it may be "against the spirit of R", but if you want
a simple NA padding at the end, this should do it:

listToDF <- function(inputList, fill = NA){
    # Use fill = NULL for regular recycling behavior
    maxLen = max(sapply(inputList, length))
    for(i in seq_along(inputList))
        inputList[[i]] <- c(inputList[[i]], rep(fill, maxLen -
length(inputList[[i]])))
    return(as.data.frame(inputList))
}

listToDF(list(x = 1:4, y = 1:2))

Michael

On Wed, Jan 11, 2012 at 2:40 PM, Chris Conner <connerpharmd at yahoo.com>
wrote:
> As a newer R practicioner, it seems I stump myself weekly (at least) with
issues that have spinning my wheels.? Here is yet another... I'm trying to
turn a list of numeric vectors (of uneual length) inot a dataframe.? Each vector
held in the list represents a row, and there are some rows of unequal length.? I
would like NAs as placeholders for "missing" data in the shorter
vectors.? I think I'm missing something quite basic.
>
> v1 <- c(1,2,3,4)
> v2 <- c(1,2)
> lst1 <- list(v1,v2)
>
> Of course there is the intuitive:
>
> as.data.frame(lst1)
>
> However, the recycling rule (expectedly) reclycles 1,2 versus using NAs as
placeholders.
>
> Then, looking into Teetor's R Cookbook, there is a piece of code that
looked (from the description) like it might do the trick:
>
> do.call(rbind, Map(as.data.frame,lst1)
>
> But I get the error:
> Error in match.names(clabs, names(xi)) :
> ? names do not match previous names
>
> Thinking the source of the error had to do with the vectors of unequal
lenght, I tried Hadley's rbind.fill thusly:
>
> library(reshape)
> do.call(rbind.fill, Map(as.data.frame,lst1)
>
> Which produced a dataset, but gain, not in the desired format.
>
> Thanks in advance to anyone that can bring my frustrations to end!
> C
> ? ? ? ?[[alternative HTML version deleted]]
>
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

Perhaps the following does what you want.  It extends
each element of your list to a common length, converts
that to a matrix, then to a data.frame:

  f <- function(data) {
    nCol <- max(vapply(data, length, 0))
    data <- lapply(data, function(row) c(row, rep(NA, nCol-length(row))))
    data <- matrix(unlist(data), nrow=length(data), ncol=nCol, byrow=TRUE)
    data.frame(data)
  }
E.g.,
  > rawData <- list(c(1,2,3), c(11,12), integer(), 31)
  > f(rawData)
    X1 X2 X3
  1  1  2  3
  2 11 12 NA
  3 NA NA NA
  4 31 NA NA

What sort of data is this?  If it is longitudinal
it might be more straigtforward to store it as a
three-column data.frame (columns "subject", "time",
"value").

Bill Dunlap
Spotfire, TIBCO Software
wdunlap tibco.com 
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at
r-project.org] On Behalf Of Chris Conner
> Sent: Wednesday, January 11, 2012 11:41 AM
> To: HelpR
> Subject: [R] turning a list of vectors into a data.frame (as rows of the
DF)?
> 
> As a newer R practicioner, it seems I stump myself weekly (at least) with
issues that have spinning my
> wheels.? Here is yet another... I'm trying to turn a list of numeric
vectors (of uneual length) inot a
> dataframe.? Each vector held in the list represents a row, and there are
some rows of unequal
> length.? I would like NAs as placeholders for "missing" data in
the shorter vectors.? I think I'm
> missing something quite basic.
> 
> v1 <- c(1,2,3,4)
> v2 <- c(1,2)
> lst1 <- list(v1,v2)
> 
> Of course there is the intuitive:
> 
> as.data.frame(lst1)
> 
> However, the recycling rule (expectedly) reclycles 1,2 versus using NAs as
placeholders.
> 
> Then, looking into Teetor's R Cookbook, there is a piece of code that
looked (from the description)
> like it might do the trick:
> 
> do.call(rbind, Map(as.data.frame,lst1)
> 
> But I get the error:
> Error in match.names(clabs, names(xi)) :
> ? names do not match previous names
> 
> Thinking the source of the error had to do with the vectors of unequal
lenght, I tried Hadley's
> rbind.fill thusly:
> 
> library(reshape)
> do.call(rbind.fill, Map(as.data.frame,lst1)
> 
> Which produced a dataset, but gain, not in the desired format.
> 
> Thanks in advance to anyone that can bring my frustrations to end!
> C
> 	[[alternative HTML version deleted]]

On Jan 11, 2012, at 1:40 PM, Chris Conner wrote:
> As a newer R practicioner, it seems I stump myself weekly (at least) with
issues that have spinning my wheels.  Here is yet another... I'm trying to
turn a list of numeric vectors (of uneual length) inot a dataframe.  Each vector
held in the list represents a row, and there are some rows of unequal length.  I
would like NAs as placeholders for "missing" data in the shorter
vectors.  I think I'm missing something quite basic.
>  
> v1 <- c(1,2,3,4)
> v2 <- c(1,2)
> lst1 <- list(v1,v2)
>  
> Of course there is the intuitive:
>  
> as.data.frame(lst1)
>  
> However, the recycling rule (expectedly) reclycles 1,2 versus using NAs as
placeholders.
>  
> Then, looking into Teetor's R Cookbook, there is a piece of code that
looked (from the description) like it might do the trick:
>  
> do.call(rbind, Map(as.data.frame,lst1)
>  
> But I get the error:
> Error in match.names(clabs, names(xi)) : 
>   names do not match previous names
>  
> Thinking the source of the error had to do with the vectors of unequal
lenght, I tried Hadley's rbind.fill thusly:
>  
> library(reshape)
> do.call(rbind.fill, Map(as.data.frame,lst1)
>  
> Which produced a dataset, but gain, not in the desired format.
>  
> Thanks in advance to anyone that can bring my frustrations to end!
> C
> 	[[alternative HTML version deleted]]

There may be an easier way, but try this:

list2df <- function(x)
{
  MAX.LEN <- max(sapply(x, length), na.rm = TRUE)
  DF <- data.frame(lapply(x, function(x) c(x, rep(NA, MAX.LEN - length(x)))))
  colnames(DF) <- paste("V", seq(ncol(DF)), sep = "")  
  DF
}

> list2df(lst1)
  V1 V2
1  1  1
2  2  2
3  3 NA
4  4 NA


HTH,

Marc Schwartz