ptir5 0.1.1

Read-only Python library for PTIR5 files (HDF5-based format from Photothermal Spectroscopy Corp.)

ptir5

Read-only Python library for PTIR5 files — the HDF5-based native format of Photothermal Spectroscopy Corp.'s PTIR Studio software (v5.0+).

Installation

From PyPI:

pip install ptir5

From conda-forge:

conda install -c conda-forge ptir5

For local development and testing:

pip install -e .[dev]
pytest -q

Quick Start

import ptir5

with ptir5.open("sample.ptir") as f:
    # List all measurements
    for m in f.measurements:
        print(m.label, m.measurement_type, m.data.shape)

    # Access a specific spectrum
    spec = f.measurements[0]
    x = spec.x_values        # wavenumber axis
    y = spec.data             # signal values

    # Navigate the document tree
    if f.tree:
        for node, folders, leaves in f.tree.walk():
            for leaf in leaves:
                if leaf.measurement is not None:
                    print(leaf.name, leaf.measurement.data.shape)

Features

• Read-only access to all PTIR5 measurement types as numpy arrays
• 16 measurement types: O-PTIR spectra/images/hyperspectra, Raman spectra/hyperspectra, camera images, fluorescence images, FL-PTIR images, PTSRS spectra/images, and image stacks
• Hierarchical tree navigation matching PTIR Studio's document structure
• Flat measurement enumeration with type filtering and GUID lookup
• Background spectra access
• Generated data (ROI spectra, band images) from hyperspectral measurements
• Lazy loading — data is read from disk only when accessed
• Dict-like metadata with Python-native types

Data Types

Type	Class	Array Shape	Dtype
O-PTIR Spectrum	OPTIRSpectrum	(points,)	float32
Raman Spectrum	RamanSpectrum	(points,)	float32
Generated Spectrum	GeneratedSpectrum	(points,)	float32
PTSRS Spectrum	PTSRSSpectrum	(points,)	float32
O-PTIR Image	OPTIRImage	(height, width)	float32
Generated Image	GeneratedImage	(height, width)	float32
PTSRS Image	PTSRSImage	(height, width)	float32
Camera Image	CameraImage	(height, width, bpp)	uint8
Fluorescence Image	FluorescenceImage	(height, width, bpp)	uint8
FL-PTIR Image	FLPTIRImage	(height, width, bpp)	uint8
O-PTIR Hyperspectra	OPTIRHyperspectra	(points, height, width)	float32
Raman Hyperspectra	RamanHyperspectra	(points, height, width)	float32
O-PTIR Image Stack	OPTIRImageStack	(images, height, width)	float32
PTSRS Image Stack	PTSRSImageStack	(images, height, width)	float32
Camera Image Stack	CameraImageStack	(images, height, width, bpp)	uint8
FL-PTIR Image Stack (legacy)	FLPTIRImageStack	(images, height, width, 4) ¹	uint8
FL-PTIR Image Stack	FLPTIRImageStack	(images, height, width)	float32

¹ Legacy rank-4 storage encodes one float32 per pixel as four uint8 bytes. Use m.read_image(i) or m.data_float32 to get canonical float32 imagery for either storage format; m.data returns the raw on-disk dataset. See docs/data_types.md for details.

Requirements

• Python >= 3.11
• h5py >= 3.9
• numpy >= 1.24

License

MIT