On Mon, 23 Jun 2014, Rainer M Krug wrote:

> One more question concerning the variable transfer of tables:
>
> ,----
> | 	(let ((file (orgtbl-to-tsv value '(:fmt org-babel-R-quote-tsv-field)))
> | 	      (header (if (or (eq (nth 1 value) 'hline) colnames-p)
> | 			  "TRUE" "FALSE"))
> | 	      (row-names (if rownames-p "1" "NULL")))
> | 	  (if (= max min)
> | 	      (format "%s <- local({
> |                         con <- textConnection(
> |                           %S
> |                         )
> |                         res <- read.table(
> |                           con,
> |                           header    = %s,
> |                           row.names = %s,
> |                           sep       = \"\\t\",
> |                           as.is     = TRUE
> |                         )
> |                         close(con)
> |                         res
> |                         })" name file header row-names)
> | 	    (format "%s <- local({
> |                        con <- textConnection(
> |                          %S
> |                        )
> |                        res <- read.table(
> |                          con,
> |                          header    = %s,
> |                          row.names = %s,
> |                          sep       = \"\\t\",
> |                          as.is     = TRUE,
> |                          fill      = TRUE,
> |                          col.names = paste(\"V\", seq_len(%d), sep =\"\")
> |                        )
> |                        close(con)
> |                        res
> |                        })" name file header row-names max))))
> `----
>
> It seems that the two variable transfer routines only differ slightly:
>
> 1. routine: =header= is present (TRUE) and defined in table when passed
> as =colnames= while
>
> 2. routine: =header= is not present (FALSE) and =colnames= is generated
> as V1 .. Vn
>
> This generation is actually not needed as this is the R default value
> which is used when =colnames= is not supplied.
>
> Also: I do not undestand why the argument =fill= is TRUE in the second
> routine. From R:
>
> ,----
> |     fill: logical. If ‘TRUE’ then in case the rows have unequal length,
> |           blank fields are implicitly added.  See ‘Details’.
> `----
>
> If I understands tables in org correctly, this is not needed as the rows
> always have equal length?
>
> so if I am not mistaken the following lines could be deleted:
>
> ,----
> |                            row.names = %s,
> |                            sep       = \"\\t\",
> |                            as.is     = TRUE,
> | -                          fill      = TRUE,
> | -                          col.names = paste(\"V\", seq_len(%d), sep =\"\")
> `----
>
> Or am I missing something here?

You need not pass a table - this

#+BEGIN_SRC R	:var x='(1 2 3)(4 5))

will produce a data.frame with two rows with is.na(x[2,2]) being TRUE.


Re col.names, from help("read.table")

      The number of data columns is determined by looking at the first
      five lines of input (or the whole file if it has less than five
      lines), or from the length of 'col.names' if it is specified and
      is longer.  This could conceivably be wrong if 'fill' or
      'blank.lines.skip' are true, so specify 'col.names' if necessary
      (as in the 'Examples').


This block

#+BEGIN_SRC R :var x=(append (make-vector 10 '(1 2)) '((1 2 3)))
x
#+END_SRC

makes an 11 by 3 data.frame with all(is.na(x[,1:10]))

but taking away the col.names arg will cause it to make a 12 by 2 
data.frame with x[12,1]==3 and is.na(x[12,2]).

so leave the col.names as is, I think.

Without fill, read.table will throw an error in that block

Chuck

[rest deleted]