fry-qt6-image-widget

fry-qt6-image-widget is a PyQt6 image viewing and annotation widget. The core class is FryCanvas, and it inherits from QtWidgets.QWidget, so it can be placed directly into a layout, a QMainWindow, a QScrollArea, or wrapped by the application as a QDockWidget when needed.

The package intentionally exposes a widget, not a dock. Dock behavior belongs to the host application.

Install

pip install fry-qt6-image-widget

For local development:

python -m pip install -e ".[dev]"
python -m pytest

Quick Start

import sys

import numpy as np
from PyQt6 import QtWidgets
from fry_qt6_image_widget import FryCanvas


app = QtWidgets.QApplication(sys.argv)

canvas = FryCanvas()
canvas.resize(900, 600)
canvas.setViewMode()

image = np.ones((480, 640, 3), dtype=np.uint8) * 255
canvas.show_image_np(image, description="white demo image")

canvas.show()
sys.exit(app.exec())

Common Imports

from fry_qt6_image_widget import (
    FryCanvas,
    FryCanvasMode,
    create_shape,
)

Backward-compatible imports are also available:

from fry_base_classes import FryCanvas

Typical Operations

canvas = FryCanvas()

canvas.show_image("demo.png")
canvas.show_image_np(image_np)
image_np = canvas.get_image_np()

canvas.set_canvas_mode("view")
canvas.set_canvas_mode("draw")
canvas.set_shape_type("rectangle")

shape = create_shape("rectangle", label="box")
canvas.append_shape(shape)
all_shapes = canvas.get_all_shapes_data()

Supported shape type strings include polygon, rectangle, circle, line, linestrip, point, points, mask, and brush.

Examples

• examples/basic_widget.py: create a FryCanvas widget and show a NumPy image.
• examples/embed_in_main_window.py: embed the widget into a normal QMainWindow with a small toolbar.
• examples/wrap_as_dock.py: optional host-app code showing how to wrap the widget as a dock outside the package.

Core Notes

The key concepts and integration notes are documented in docs/CORE_CONCEPTS.md.

Build a pip Package

python -m pip install build twine
python -m build
python -m twine check dist/*

This produces a source distribution and a wheel in dist/. To publish to PyPI, run:

python -m twine upload dist/*

Publishing needs a PyPI account and token, so the repository is prepared for upload but credentials are not stored in the project.