scada-ui-kit 0.1.1

A high-performance, dark-mode SCADA UI widget library for PyQt6.

SCADA UI Kit

A robust, enterprise-grade industrial UI widget library for PyQt6, engineered for performance, deep customisation, and a smooth developer experience. The codebase is written to Fluent Python (2e) idioms: rich data-model methods, property-based APIs, deque-backed buffers, IntEnum state codes, Qt-signal observers, and fully typed public interfaces (PEP 561).

Features

• ScadaRadialGauge - dark-mode radial gauge with a configurable alert threshold that turns the needle red above 80% of range. Assign to .value and the widget repaints itself.
• ScadaStripChart - real-time scrolling telemetry chart backed by a fixed-length collections.deque for O(1) appends and automatic eviction of the oldest sample. Iterable and subscriptable like a normal sequence.
• ScadaIndicator + ScadaIndicatorMatrix - multi-state LED indicators laid out in a grid. The matrix implements the full mapping protocol, so you can write matrix["COOLANT"] = IndicatorState.RUNNING, iterate with for name in matrix, and check membership with "COOLANT" in matrix.
• ScadaToggle - heavy-duty actuator switch that emits toggled(bool) so you can wire control inputs straight to your indicator matrix or business logic.
• IndicatorState enum - OFFLINE, RUNNING, WARNING, FAULT. It's an IntEnum, so raw integers still work for back-compat.
• Typed - every public symbol ships with type hints and a PEP 561 py.typed marker, so mypy and pyright pick up the annotations out of the box.

Installation

From the project root:

pip install .

Or, for local development with live code reloads:

pip install -e .

PyQt6 is pulled in automatically as a dependency.

Quick Start

import sys
from PyQt6.QtWidgets import QApplication
from scada_ui_kit import ScadaRadialGauge

app = QApplication(sys.argv)
gauge = ScadaRadialGauge(title="REACTOR PRESSURE", unit="PSI", min_val=0, max_val=150)
gauge.value = 87        # property assignment - triggers the repaint automatically
gauge.show()
sys.exit(app.exec())

Mapping-style status updates

from scada_ui_kit import IndicatorState, ScadaIndicatorMatrix

matrix = ScadaIndicatorMatrix(columns=2)
for name in ("COOLANT", "CORE TEMP", "CONTAINMENT", "VENTILATION"):
    matrix.add_indicator(name)

matrix["COOLANT"] = IndicatorState.RUNNING
matrix["VENTILATION"] = IndicatorState.WARNING

assert "COOLANT" in matrix
for system in matrix:                         # __iter__ yields indicator names
    print(system, matrix[system].state)       # __getitem__ returns the widget

Toggle with observer wiring

from scada_ui_kit import IndicatorState, ScadaToggle

pump = ScadaToggle()
pump.toggled.connect(
    lambda on: matrix.__setitem__(
        "COOLANT",
        IndicatorState.RUNNING if on else IndicatorState.OFFLINE,
    )
)

Running the demos

The repository ships with five runnable scripts at the project root. Once the package is installed (pip install -e .):

python main_dashboard.py   # unified dashboard with all widgets and a live feed
python scada_gauge.py      # slider-driven gauge demo
python scada_chart.py      # sine-wave + noise strip chart
python scada_matrix.py     # indicator matrix showcase
python scada_toggle.py     # toggle wired to a live status label

Requirements

• Python 3.10+
• PyQt6 6.0+

Building distribution artifacts

The repo ships a small helper that produces an sdist and a wheel in ./dist/ and validates them with twine check before any upload.

pip install -e .[build]      # installs the `build` + `twine` tools locally
python build_package.py      # clean -> build -> twine check -> summary

Useful flags:

python build_package.py --clean-only   # just wipe dist/ and build/
python build_package.py --no-clean     # keep previous artifacts
python build_package.py --no-check     # skip the twine validation pass

A successful run prints something like:

[build] artifacts ready in ./dist/:
    scada_ui_kit-0.1.0-py3-none-any.whl                         12.3 KB
    scada_ui_kit-0.1.0.tar.gz                                   10.1 KB

The wheel is a single, self-contained file you can distribute to users - they install it with pip install scada_ui_kit-0.1.0-py3-none-any.whl.

Bumping the version

The package version lives in one place: __version__ in scada_ui_kit/__init__.py. pyproject.toml resolves it dynamically at build time via setuptools' attr directive, so drift is impossible by construction.

python bump_version.py patch        # 0.1.0 -> 0.1.1  (bug fix)
python bump_version.py minor        # 0.1.0 -> 0.2.0  (back-compatible feature)
python bump_version.py major        # 0.1.0 -> 1.0.0  (breaking change)
python bump_version.py 2.7.3        # explicit SemVer set
python bump_version.py patch -n     # dry-run: print the new version and exit

Git automation

python bump_version.py patch -c     # write + 'git add' + 'git commit'
python bump_version.py patch -c -t  # also create an annotated 'vX.Y.Z' tag

Safety checks that run before anything is written:

• If --commit or --tag is requested, git rev-parse --is-inside-work-tree verifies we're actually in a repo.
• If --tag is requested, git tag --list vX.Y.Z must be empty (no clobber).
• The commit is scoped with git commit -- scada_ui_kit/__init__.py so any unrelated already-staged changes don't get swept into the release commit.

After a successful --commit -t, push with:

git push && git push origin vX.Y.Z

Typical release flow

python bump_version.py patch -c -t        # write, commit, tag
python build_package.py                   # sdist + wheel + twine check
python publish.py                         # TestPyPI (safe default)
python publish.py --production            # real PyPI, with y/N prompt
git push && git push origin vX.Y.Z        # share the release commit/tag

Publishing to PyPI / TestPyPI

Once the artifacts in dist/ are validated, publish.py uploads them with sensible safety defaults:

python publish.py                    # uploads to TestPyPI (safe default)
python publish.py --production       # uploads to real PyPI; asks for 'yes'
python publish.py --production -y    # non-interactive production upload (CI)

Authentication is delegated to twine, so any of the standard mechanisms work:

• TWINE_USERNAME / TWINE_PASSWORD environment variables (recommended for CI; set TWINE_USERNAME=__token__ and TWINE_PASSWORD=<your PyPI API token>).
• A ~/.pypirc file with [pypi] and [testpypi] sections.
• The system keyring, if configured.

Safety features baked into the script:

• twine check runs before any upload so malformed metadata is caught locally.
• Production uploads require both --production and an explicit yes confirmation (or --yes for automation).
• Non-interactive environments (no TTY) without --yes are refused outright for production, preventing accidents in cron jobs and CI triggers.

License

MIT