Skip to contents

Write an object to a file based on its extension

pm_write_file.Rd

Writes an object to a file based on the file extension. Supports multiple file formats including tabular data (CSV, TSV), serialized R objects (RDS), and R data files (RData/RData).

Usage

pm_write_file(file, x, ..., object_names = NULL)

Arguments

file

Character. Path to the file to write

x

The object to write. For tabular formats (CSV, TSV, Parquet), must be a data.frame or similar tabular object. For RData files, can accept multiple objects via ....

...

Additional arguments passed to the underlying write function. For RData files, additional objects can be provided here.

object_names

Character vector. Optional. Internal parameter. For RData files, explicitly specify the names for objects. If provided, must match the number of objects (x + objects in ...). If not provided, names are extracted from the call. This parameter is intended for use when forwarding a ... call from a parent frame where object names need to be preserved. See PMData$write() for example usage.

Value

Invisibly returns the file path.

Details

Supported file formats:

  • CSV: Comma-separated values files (.csv) - uses write.csv(). Requires a data.frame.

  • TSV: Tab-separated values files (.tsv) - uses write.table() with sep = "\t". Requires a data.frame.

  • Parquet: Apache Parquet files (.parquet, .pqt) - requires the arrow package. Requires a data.frame.

  • RDS: R serialized data files (.rds) - uses saveRDS(). Accepts any R object.

  • RData/RData/Rda: R data files (.rdata, .RData, .rda, .Rda) - uses save(). Can accept multiple objects: pm_write_file("file.RData", obj1, obj2, obj3).

For tabular formats (CSV, TSV, Parquet), the function validates that the object is a data.frame or similar tabular structure.

The object_names parameter is an internal parameter used when forwarding ... calls from a parent frame. When a wrapper function (like PMData$write()) needs to preserve object names from the original call, it can capture the names and pass them via object_names. This ensures that RData files contain objects with the correct names even when called through wrapper functions. See the implementation of PMData$write() for an example of how to use this parameter.

Examples

# Example usage:
withr::with_tempdir({

# Write a data.frame to CSV
df <- data.frame(x = 1:5, y = letters[1:5])
pm_write_file("data.csv", df)

# Write a data.frame to TSV
pm_write_file("data.tsv", df)

# Write any object to RDS
my_list <- list(a = 1, b = 2)
pm_write_file("data.rds", my_list)

# Write multiple objects to RData
obj1 <- data.frame(x = 1:3)
obj2 <- c("a", "b", "c")
pm_write_file("data.RData", obj1, obj2, obj3 = 42)

})