write_dataset: write_dataset

In duckdbfs: High Performance Remote File System, Database and 'Geospatial' Access Using 'duckdb'

write_dataset

Description

write_dataset

Usage

write_dataset(
  dataset,
  path,
  conn = cached_connection(),
  format = c("parquet", "csv"),
  partitioning = dplyr::group_vars(dataset),
  overwrite = TRUE,
  options = list(),
  ...
)

Arguments

dataset	a remote tbl object from open_dataset, or an in-memory data.frame.
path	a local file path or S3 path with write credentials
conn	duckdbfs database connection
format	export format
partitioning	names of columns to use as partition variables
overwrite	allow overwriting of existing files?
options	Additional arguments to COPY, see https://duckdb.org/docs/stable/sql/statements/copy.html#copy–to-options Note, uses duckdb native syntax, e.g. c("PER_THREAD_OUTPUT false"), for named arguments, see examples. (Recall SQL is case-insensitive).
...	additional arguments to duckdb_s3_config()

Value

Returns the path, invisibly.

Examples

  write_dataset(mtcars, tempfile())


write_dataset(mtcars, tempdir())
write_dataset(mtcars, tempdir(), options = c("PER_THREAD_OUTPUT FALSE", "RETURN_STATS TRUE"))

duckdbfs documentation built on Aug. 8, 2025, 6:57 p.m.