xportrs

SAS Transport (XPT) V5/V8 format reader and writer in pure Rust.

Features

• Full V5 format support - Maximum compatibility with regulatory submission systems
• V8 format support - Extended limits for variable names (32 chars), labels (256 chars)
• IEEE ↔ IBM float conversion - Accurate floating-point conversion between formats
• All 28 SAS missing value codes - Standard (.) and special (.A-.Z, ._)
• Streaming API - Memory-efficient processing of large files
• Optional Polars integration - Direct DataFrame conversion with polars feature

Installation

cargo add xportrs

Or add to your Cargo.toml:

[dependencies]
xportrs = "0.0.1"

Optional Features

# Enable Polars DataFrame integration
xportrs = { version = "0.0.1", features = ["polars"] }

# Enable Serde serialization
xportrs = { version = "0.0.1", features = ["serde"] }

Format Comparison

Usage

Reading XPT Files

use std::path::Path;
use xportrs::{read_xpt, XptDataset};

// Read entire dataset
let dataset = read_xpt(Path::new("dm.xpt")) ?;
println!("Dataset: {} ({} rows)", dataset.name, dataset.num_rows());

// Access columns
for column in & dataset.columns {
println!("  {} ({})", column.name, column.xpt_type);
}

Writing XPT Files

use std::path::Path;
use xportrs::{XptDataset, XptColumn, XptValue, write_xpt};

// Create dataset with columns
let mut dataset = XptDataset::with_columns(
"DM",
vec![
    XptColumn::character("USUBJID", 20).with_label("Unique Subject ID"),
    XptColumn::numeric("AGE").with_label("Age in Years"),
],
);

// Add rows
dataset.add_row(vec![
    XptValue::character("STUDY-001"),
    XptValue::numeric(35.0),
]);

// Write to file
write_xpt(Path::new("dm.xpt"), & dataset) ?;

Streaming Large Files

use std::path::Path;
use xportrs::read_xpt_streaming;

// Process rows one at a time (memory efficient)
let reader = read_xpt_streaming(Path::new("large_file.xpt")) ?;
for observation in reader {
let row = observation ?;
// Process each row...
}

Missing Values

SAS supports 28 different missing value codes:

use xportrs::{XptValue, MissingValue};

// Standard missing (.)
let missing = XptValue::numeric_missing();

// Special missing (.A through .Z)
let missing_a = XptValue::numeric_missing_with(MissingValue::Special('A'));

// Check for missing
assert!(missing.is_missing());

Polars Integration

Enable the polars feature for DataFrame support:

[dependencies]
xportrs = { version = "0.0.1", features = ["polars"] }

use std::path::Path;
use xportrs::{read_xpt_to_dataframe, write_dataframe_to_xpt};

// Read XPT to DataFrame
let df = read_xpt_to_dataframe(Path::new("dm.xpt")) ?;

// Write DataFrame to XPT
write_dataframe_to_xpt( & df, "DM", Path::new("dm_out.xpt")) ?;

V8 Format

By default, files are written in V5 format. Use XptWriterOptions for V8:

use std::path::Path;
use xportrs::{XptWriterOptions, XptVersion, write_xpt_with_options};

let options = XptWriterOptions::default ().with_version(XptVersion::V8);
write_xpt_with_options(Path::new("dm.xpt"), & dataset, & options) ?;

MSRV

The minimum supported Rust version is 1.92.

License

MIT License - see LICENSE for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Related Projects

This crate is used by Trial Submission Studio, a desktop application for transforming clinical trial data into FDA-compliant CDISC formats.