metaxtract 0.2.0

Metadata extractor and converter for scientific instrument file formats (AFM, EM, X-ray).

MetaXtract

Overview

MetaXtract is a Python package designed to extract metadata from various scientific file formats including:

• .h5 (HDF5)
• .xrdml (XRDML XML-based)
• .dm4 (DigitalMicrograph)
• .ibw (Igor Binary Wave)

Installation

From PyPI

# Basic installation (metadata extraction only)
pip install metaxtract

# With converters (xarray + netCDF4)
pip install "metaxtract[converters]"

# With every optional feature (converters, utils, checksums, viz)
pip install "metaxtract[all]"

From source

git clone https://gitlab.com/dataerai/MetaXtract.git
cd MetaXtract
pip install -e ".[dev]"     # editable install + tests, docs, lint, build tools

Optional Dependencies

• xarray ([converters]): Required for file conversion to netCDF format
• requests ([utils]): Required for the download_file() utility function
• blake3 ([checksums]): Optional faster checksum algorithm (falls back to sha256 if not available)
• plotly ([viz]): Required for interactive visualizations (metadata and data visualization)

Documentation

Full documentation is available at: https://dataerai.gitlab.io/MetaXtract/

Usage

Using the High-Level API

from metaxtract import extract_metadata, get_supported_formats

# Get list of supported formats
formats = get_supported_formats()
print(formats)

# Extract metadata from a file
metadata = extract_metadata("path/to/file.h5")
print(metadata)

Data Type Detection

from metaxtract import extract_metadata, detect_data_type

# Extract metadata
metadata = extract_metadata("file.ibw")

# Automatically detect data type
result = detect_data_type(metadata)
if result:
    print(f"Data Type: {result['type']}")
    print(f"Confidence: {result['confidence']:.2%}")

Visualization

from metaxtract import convert_file
from metaxtract.viz import visualize_data, visualize_metadata_tree

# Visualize metadata
metadata = extract_metadata("file.ibw")
fig = visualize_metadata_tree(metadata)
fig.show()

# Visualize data (automatically detects type and routes to appropriate visualization)
dataset = convert_file("file.ibw")
fig = visualize_data(dataset)
fig.show()

Utility Functions

from metaxtract.utils import download_file, benchmark_checksums

# Download a file
file_path = download_file("https://example.com/data.ibw")

# Benchmark checksum algorithms
results = benchmark_checksums("data.ibw", algorithms=['sha256', 'md5', 'blake3'])

Running the extractor

For h5 File type

from metaxtract.instruments.AFM.bandexcitation.h5 import H5
import json
from pprint import pprint

process_file = H5("path/to/file")
metadata = process_file.extract()

print(metadata)
pprint(metadata)  # Pretty Print likewise
print(json.dumps(metadata, indent=4))  # Print metadata in JSON format

For xrdml file type

from metaxtract.instruments.Xray.panalytical.xrdml import XRDML
import json
from pprint import pprint

process_file = XRDML("path/to/file")
metadata = process_file.extract()

print(metadata)
pprint(metadata)  # Pretty Print likewise
print(json.dumps(metadata, indent=4))  # Print metadata in JSON format

For dm4 file type

from metaxtract.instruments.EM.dm.dm4 import DM4
import json
from pprint import pprint

process_file = DM4("path/to/file")
metadata = process_file.extract()

print(metadata)
pprint(metadata)  # Pretty Print likewise
print(json.dumps(metadata, indent=4))  # Print metadata in JSON format

For ibw file type

from metaxtract.instruments.AFM.oxfordAFM.ibw import IBW
import json
from pprint import pprint

process_file = IBW("path/to/file")
metadata = process_file.extract()

print(metadata)
pprint(metadata)  # Pretty Print likewise
print(json.dumps(metadata, indent=4))  # Print metadata in JSON format

Links

• Source Code: https://gitlab.com/dataerai/MetaXtract
• Documentation: https://dataerai.gitlab.io/MetaXtract/
• Package Registry: https://gitlab.com/dataerai/MetaXtract/-/packages
• Issues: https://gitlab.com/dataerai/MetaXtract/-/issues

Contributing

Contributions are welcome! Please see CONTRIBUTING.rst for guidelines.

License

This project is licensed under the MIT License — see the LICENSE file for details.