read.ctable: Read a complex table and return array in basic storagemode.

In garray: Generalized Array Arithmetic for Ragged Arrays with Named Margins

Description

A complex table has several row and colume headers, some of which indicate the hierarchy of the dimensions (thus the returned array may have more dimensions than row and column), some of which are real row.names and col.names that will be turn into dimnames, and some of which are additional attributes. Cells of non-header are all in a same format (like double). The original header are preserved as attributes. The dimnames strictly save the layout of the matrix, thus row.names and col.names should be carefully chosen for matching the actual dimension. Sometime aperm() is needed after this function. Sparse table, where dimnames is not complete (unbalanced) and needs non-fractionally recycling, is not supported.

Usage

read.ctable(file, header, row.names, col.names, ...,
  storagemode = "double")

Arguments

file	The name of the file to be read in, see ('file' in ?read.table).
header	hierarchy structure of the header, a list of length 2, assigning the index of row and col header, affecting parsing of the input table;
row.names, col.names	a list, whose elements can be a character vector assigning the name of one dimension directly, or a integer scalar which is the index (icol,irow) of row,col-header that will be extracted; since the row and the col headers can have hierarchy in the input table and the hierarchy will be organized into addisional dimensions, rownames indicate the names of dimensions that are from the row of the input table, while colnames, the col; in the input table, the header can be in two pattern: AAA and AOO; in the AAA pattern, the names are repeated for the cells that are in the same index in high dimension, while in the AOO pattern, the names appear once at the first and the other cells that are in the same index in high dimension are left blank; in the output array, row,col.names are not necessary dimnames[[1]] and dimnames[[2]]; if all elements of the list are a integer scalar, then the list can also be coded as a integer vector (since they are the same for lapply);
...	Further arguments to be passed to ‘read.table’.
storagemode	The storagemode of return matrix, usually 'double'.

garray documentation built on May 1, 2019, 8:47 p.m.