Concatenate multiple dafs along one or more axes.

Mirrors Julia concatenate!(). For each concatenation axis, entries from each source are appended in source order. Non-concat axes must be identical across all sources and are copied once.

Usage

concatenate(
  destination,
  axis,
  sources,
  names = NULL,
  dataset_axis = "dataset",
  dataset_property = TRUE,
  prefix = FALSE,
  prefixed = NULL,
  empty = NULL,
  merge = NULL,
  sparse_if_saves_storage_fraction = 0.25,
  overwrite = FALSE
)

Arguments

destination: A DafWriter. Must be empty of the concatenation axes.
axis: A single axis name or a character vector of axis names.
sources: List of DafReaders to concatenate.
names: Optional character vector of unique data set names (defaults to each source's name prop).
dataset_axis: Name of the per-source axis to create. NULL disables.
dataset_property: If TRUE (default) and dataset_axis is non-NULL, create a same-named vector on every concatenation axis holding the source name for each entry.
prefix: Logical (single or per-axis). Prefix concat-axis entries with "<dataset_name>." to de-duplicate across sources.
prefixed: Optional character vector (or list of vectors per axis) of additional property names to prefix, beyond the heuristic (same-name or "<axis>.*" properties).
empty: Named list of fill values for missing per-source properties.
merge: Named list mapping property keys to a merge action ("SkipProperty", "LastValue", "CollectAxis").
sparse_if_saves_storage_fraction: Numeric (default 0.25); reserved for a future sparse-promotion heuristic.
overwrite: If TRUE, allow replacing pre-existing destination entries.

Value

Invisibly, the destination.

Examples

a <- memory_daf(name = "A"); add_axis(a, "cell", c("a1", "a2"))
b <- memory_daf(name = "B"); add_axis(b, "cell", c("b1"))
dest <- memory_daf(name = "dest")
concatenate(dest, "cell", list(a, b))