From Scripts to Systems: Beginner

Python Application Architecture for Instrument Control

A 3-day beginner track for engineers without a Python background. Takes you from script-level code through to structured, testable instrument control systems.

---

Structure

slides/               Presentation decks
  Day1_Script_to_Architecture.pptx             (38 slides)
  Day2_Python_Foundations_and_Functions.pptx    (54 slides)
  Day3_ABC_Patterns_and_Instrumentation.pptx   (49 slides)
tasks/                Tiered exercises: work through these during the course
  day2_tasks.py       10 tasks covering Python fundamentals
  day3_tasks.py       6 tasks + capstone covering patterns & instrumentation
solutions/            Reference solutions: check your work after each task
  day2_solutions.py
  day3_solutions.py

---

Prerequisites

• Python 3.10+
• A text editor or IDE (VS Code recommended)
• No hardware required; all instrument examples use simulators
• No additional packages needed for Day 2
• Day 3 references pyserial but exercises use simulated instruments

---

Day 1: From Script to Architecture

Why scripts break, and how to fix them with objects, classes, and patterns.

Topic	What You'll Learn
Why Scripts Break	Global state, tight coupling, the "200-line wall"
OOP from Scratch	Classes, __init__, self, state and responsibility
Composition vs Inheritance	When to use each, how to combine objects
Seams & Refactoring	Finding the cut points in messy code
Adapter Pattern (intro)	Making hardware swappable via an ABC interface
Strategy Pattern (intro)	Making measurement modes swappable

Day 1 is the same in both tracks. If you have already done Day 1 of the Advanced course, you can skip it.

Day 1 is slide-based with 4 progressive labs (refactoring messy scripts into clean classes). No task file; work directly from the lab instructions in the slides.

---

Day 2: Python Foundations and Functions

Covers the Python building blocks needed before tackling architecture patterns.

Topic	What You'll Learn
Data Types	int, float, str, bool, type conversion, f-strings
Collections	Lists, dicts, tuples, sets; when to use each
Control Flow	if/elif/else, for/while loops, break/continue
Comprehensions	List and dict comprehensions as Pythonic shorthand
Functions	def, parameters, defaults, type hints, docstrings
*args / **kwargs	Flexible arguments, unpacking, star expressions
Scope & Closures	Local vs global, LEGB rule, closures for capturing state
Lambda & HOF	Anonymous functions, map/filter/sorted with keys
Decorators	@timer, @log_call, functools.wraps, stacking
Error Handling	try/except/finally, custom exceptions, exception chaining

Tasks (Day 2)

Open tasks/day2_tasks.py and work through the exercises. Each task corresponds to a topic above. Check your work against solutions/day2_solutions.py.

---

Day 3: ABC, Patterns and Instrumentation

Applies Day 2 foundations to real instrument control patterns. Starts with a catch-up on decorators and error handling if those weren't fully covered on Day 2.

Topic	What You'll Learn
Catch-up	Decorators + error handling essentials (if needed from Day 2)
Abstract Base Classes	@abstractmethod, contracts, abstract @property
Instrumentation	Serial/SCPI communication, request-response pattern
Adapter Pattern	ABC + SimulatedAdapter + SerialAdapter; swappable hardware
Strategy Pattern	Measurement modes: single-shot, averaging, max
DataSink Pattern	CSV, print, and composite sinks for flexible output
Capstone	Mini data acquisition system wiring everything together

Tasks (Day 3)

Open tasks/day3_tasks.py and work through the exercises. The capstone at the end ties all the patterns together into one working system.

---

How the Exercises Work

Every task uses a three-tier model. Start at the tier that matches your experience and move up if you finish early.

Tier	Slides Colour	Who It's For
Beginner	Teal	New to Python; guided steps, core concepts
Intermediate	Blue	Some experience; less scaffolding, deeper understanding
Advanced	Yellow	Experienced; internals, edge cases, design patterns

---

Key Principle

"The controller does not care whether the instrument is real or simulated, whether the data goes to CSV or JSON, or whether we measure once or average a hundred readings."

By the end of Day 3, you'll have built a system where every component is swappable:

• Adapter abstracts the hardware: swap SimulatedAdapter for SerialAdapter without changing the controller
• Strategy abstracts measurement modes: swap SingleShot for Averaging in one line
• DataSink abstracts output: swap PrintSink for CsvSink or use both via CompositeSink

---

What Comes Next

After completing this beginner track, the next step is the advanced course, From Scripts to Systems: Advanced, which builds on these patterns to create a full desktop application with PySide6, threading, real-time plotting, and configuration management.

---

Need Help?

Ask your instructor, or check solutions/ to compare your approach.