How to in python

datasetjson - Read and write CDISC Dataset JSON formatted datasets in R and Python

R/Pharma 2025 Workshop

2025-11-07

How to Read/Write Dataset-JSON using Python

Getting to Know dsjconvert for Dataset-JSON

1. Introduction to the dsjconvert package
2. Using dsjconvert: as a command-line tool
3. Using dsjconvert: as a Python package
4. Exercises: getting your hands dirty

Introduction to dsjconvert

• Converts SAS XPT and SAS7BDAT files to Dataset-JSON v1.1 in JSON or NDJSON
• Converts Dataset-JSON in JSON or NDJSON to SAS XPT
• Validates Dataset-JSON datasets
• Functions as a CLI application or a Python library

dsjconvert Features

• Multiple Input Formats: Converts XPT, SAS7BDAT, and Dataset-JSON datasets
• Dual Output Formats: Dataset-JSON and SAS XPT
• Dual Dataset-JSON Formats: Supports JSON and NDJSON Dataset-JSON formats
• Flexible Metadata: Use Define-XML metadata or auto-infer from source data
• Validation: Built-in validation against Dataset-JSON schemas
• Logging: Configurable logging levels for debugging

Usage: as a CLI Tool

Convert XPT files using defaults (NDJSON format):

dsjconvert -v -x

Convert SAS7BDAT files to JSON format:

dsjconvert -v -b --format json

Convert Dataset-JSON to XPT

dsjconvert -v --to-xpt --input-format ndjson

Usage: as a Python Library

from dsjconvert import XPTConverter, MetadataExtractor

# With Define-XML metadata
extractor = MetadataExtractor('path/to/define.xml')
converter = XPTConverter(
    metadata_extractor=extractor,
    output_format='ndjson',
    skip_validation=True
)
converter.convert_dataset('input.xpt', 'output_dir')

# Without Define-XML (auto-infer metadata)
converter = XPTConverter(output_format='ndjson')
converter.convert_dataset('input.xpt', 'output_dir')

Usage: as a Python Library

Convert multiple SAS7BDAT files to JSON format:

import os
from dsjconvert import SAS7BDATConverter

converter = SAS7BDATConverter(output_format='json')

# Get all SAS files
sas_dir = 'data'
sas_files = [f for f in os.listdir(sas_dir) if f.endswith('.sas7bdat')]

# Convert each file
for sas_file in sas_files:
    input_path = os.path.join(sas_dir, sas_file)
    output_path = converter.convert_dataset(input_path, 'output')
    print(f"Converted: {output_path}")

Usage: as a Python Library

Convert Dataset-JSON to SAS XPT

from dsjconvert import DatasetJSONToXPTConverter

converter = DatasetJSONToXPTConverter(input_format='ndjson')
xpt_path = converter.convert_dataset('input.ndjson', 'output_dir')

Questions?

• Questions?
• Let’s get into the exercises!
• The exercises focus on converting to and from Dataset-JSON
• Open exercises/02-python.py and work through the exercises

dsjconvert Exercises

1. Convert cm.xpt to CM.ndjson using the define.xml metadata
2. Convert vs.xpt to VS.ndjson without using the define.xml metadata
3. Convert dm.xpt to DM.json using the define.xml metadata
4. Convert MH.ndjson to mh.xpt using the define.xml metadata