inst/examples/quickstart_sparse_string.R

In tiledb: Universal Storage Engine for Sparse and Dense Multidimensional Arrays

#!/usr/bin/env Rscript
##
## cf quickstart_sparse_string.cc
##
## When run, this program will create a 2D sparse array with one dimension a
## string type, and the other an integer. This models closely what a dataframe
## looks like. The program will write some data to it, and read a slice of the
## data back.

library(tiledb)

## Name of the array to create.
array_name <- "quickstart_sparse_string_array"

## Path is either current directory, or a local config value is found
uri <- file.path(getOption("TileDB_Data_Path", "."), array_name)

create_array <- function(uri) {
  ## The array will be 4x4 with dimensions "rows" and "cols"
  ## "rows" is a string dimension type so domain and extend are NULL
  ## and space tiles 2x2
  dom <- tiledb_domain(dims = c(tiledb_dim("rows", NULL, NULL, "ASCII"),
                                tiledb_dim("cols", c(1L,4L), 4L, "INT32")))

  ## Sparse schema
  schema <- tiledb_array_schema(dom, attrs = c(tiledb_attr("a", type = "INT32")),
                                sparse = TRUE)

  ## Create the (empty) array on disk.
  tiledb_array_create(uri, schema)
  invisible(NULL)
}

write_array <- function(uri) {
  ## Write some simple data to cells ("a", 1), ("bb", 4) and ("c", 3).
  data <- c(1L, 2L, 3L)
  coords_rows <- c('a', 'bb', 'c')
  coords_cols <- c(1L, 4L, 3L)

  arr <- tiledb_array(uri, query_type="WRITE", is.sparse=TRUE)

  arr[] <- data.frame(rows=coords_rows, cols=coords_cols, a=data)
  invisible(NULL)
}

read_array <- function(uri) {
  ## opening in data.frame mode
  ## returning three vectors
  arr <- tiledb_array(uri, return_as="data.frame")
  dat <- arr[]
  print(dat)
}

if (dir.exists(uri)) unlink(uri, recursive=TRUE)
create_array(uri)
write_array(uri)
read_array(uri)