Steve Jaffe wrote:
> I'm trying to build up a list of data.frames by "appending" one by one.
>
> If x,y,z are data.frames, I can do
>
> somelist <- list(x, y, z) (or even somelist <- list(x=x, y=y, z=z) to get
> names)
>
> But if I start with
>
> somelist <- list(x,y) and now want to go from this to list(x,y,z) I'm stuck.
> I've tried
>
>
> append(somelist, z)
> c(somelist, z)
>
return a list whose elements are the elements of somelist and the
*elements* of z, not z itself -- z, a data frame, is a *list*, and it's
elements are the vectors that are its columns.
> list(somelist, z)
>
makes a two-element list, with somelist and z as the elements.
you may want to consult the 'help' pages, though they do not appear to
be particularly carefully crafted. ?append explains:
"
Vector Merging
Description:
Add elements to a vector.
Usage:
append(x, values, after = length(x))
Arguments:
x: the vector to be modified.
values: to be included in the modified vector.
"
which is suggestive (perhaps except for the name 'values') of that
*values*, rather than its elements, is included in the modified vector,
while the opposite is true:
append(list(1), 2)
# list of 1 and 2
append(list(1), list(2))
# list of 1 and 2, not of 1 and list(2)
furthermore, "add elements to a vector" is misleading, since the
following works:
append(data.frame(1), 2)
# list of (named!) 1 and 2
is.vector(data.frame(1))
# FALSE
?c says:
"
Usage:
c(..., recursive=FALSE)
Arguments:
...: objects to be concatenated.
recursive: logical. If 'recursive = TRUE', the function recursively
descends through lists (and pairlists) combining all their
elements into a vector.
Details:
The output type is determined from the highest type of the
components in the hierarchy NULL < raw < logical < integer < real
< complex < character < list < expression. Pairlists are treated
as lists, but non-vector components (such names and calls) are
treated as one-element lists which cannot be unlisted even if
'recursive = TRUE'.
"
this 'help' page explains what is supposed to happen when
recursive=TRUE, but it fails to tell you what is supposed to happen
when, by default, recursive=FALSE. one might expect that it's to mean
that lists (and pairlists) are *not* recursively descended though, which
to my simple mind would mean that a list argument is taken to be a
single *element* of the result, not a collection of elements to be
included, each separately, in the result -- which is *not* the case, "of
course":
c(list(1), 2)
# list of 1 and 2, not a list of list(1) and 2
this 'help' page also explains that non-vector elements are treated as
one-element lists which cannot be unlisted -- even if recursive=TRUE
(!), but it does not really hold:
c(1, data.frame(2,3))
# list of 1, 2, and 3, not a list of 1 and data.frame(2,3)
is.vector(data.frame(2,3))
# FALSE
the subtle trick here is that data frames (which are lists) are
un-dataframed into lists that happen to be is.vector-TRUE (take care,
not all is.list-TRUE objects are is.vector-TRUE), hence are treated as
vectors though they are not. (or maybe it's just that is.vector is
confused.)
vQ
______________________________________________
R-help@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.
