Control structures and functions

Why this matters

Useful programs must decide, for example warning when the battery is low, and repeat, for example computing every leg of a route. As code grows, it must be split into functions with clear jobs that can be tested separately and reused. The autopilots that fly real aircraft are built from the same elements, only on a much larger scale.

Deciding with if

This example grades a battery against PX4’s default thresholds in the version checked (low 15%, critical 7%), covered in DRT 342 module 2.

def battery_status(percent):
    """Return "OK", "LOW" or "CRITICAL" for a battery percentage (0-100)."""
    if not 0 <= percent <= 100:
        raise ValueError("percent must be between 0 and 100")
    if percent <= 7:
        return "CRITICAL"
    if percent <= 15:
        return "LOW"
    return "OK"


for level in [100, 16, 15, 8, 7, 0]:
    print(level, battery_status(level))

100 OK
16 OK
15 LOW
8 LOW
7 CRITICAL
0 CRITICAL

Order matters. <= 7 must be checked before <= 15; swap them and 5% returns “LOW” instead of “CRITICAL”. Values exactly on a boundary, such as 15 and 7, must always be tested.

Handling errors

Given an impossible value such as 120%, a function should not guess. It should raise an exception so the caller knows, and the caller chooses how to handle it with try/except.

readings = [45, 120, 12]
for reading in readings:
    try:
        print(reading, battery_status(reading))
    except ValueError as err:
        print(reading, "rejected:", err)

45 OK
120 rejected: percent must be between 0 and 100
12 LOW

Catching everything with a bare except: and doing nothing is a dangerous habit, because it hides the real problem. Always name the error type you expect.

Loops and data structures

A list holds an ordered, changeable sequence; a tuple holds data that should not change, such as one coordinate; a dict holds name–value pairs.

import math

waypoints = [(0, 0), (200, 0), (200, 100), (0, 100), (0, 0)]
drone = {"id": "RSU-QUAD-01", "cruise_ms": 8.0}

total_m = 0.0
for start, end in zip(waypoints, waypoints[1:]):
    leg = math.dist(start, end)
    total_m += leg
    print(f"{start} -> {end}: {leg:.0f} m")

minutes, seconds = divmod(total_m / drone["cruise_ms"], 60)
print(f"total {total_m:.0f} m, {minutes:.0f} min {seconds:.0f} s")

(0, 0) -> (200, 0): 200 m
(200, 0) -> (200, 100): 100 m
(200, 100) -> (0, 100): 200 m
(0, 100) -> (0, 0): 100 m
total 600 m, 1 min 15 s

Reusable functions

A good function does one job, has a name that says what it does, has a docstring describing what it takes and returns (PEP 257), and does not quietly change outside values.

EARTH_RADIUS_M = 6_371_000.0
NM_TO_M = 1852


def distance_m(lat1, lon1, lat2, lon2, radius=EARTH_RADIUS_M):
    """Great-circle distance in metres between two points given in degrees."""
    phi1, phi2 = math.radians(lat1), math.radians(lat2)
    d_phi = phi2 - phi1
    d_lambda = math.radians(lon2 - lon1)
    a = math.sin(d_phi / 2) ** 2 + math.cos(phi1) * math.cos(phi2) * math.sin(d_lambda / 2) ** 2
    return 2 * radius * math.asin(math.sqrt(a))


def dms_to_deg(degrees, minutes, seconds):
    """Convert degrees, minutes and seconds to decimal degrees."""
    return degrees + minutes / 60 + seconds / 3600


vtr1_centre = (dms_to_deg(13, 45, 54), dms_to_deg(100, 32, 18))
site = (dms_to_deg(13, 50, 0), dms_to_deg(100, 40, 0))

d = distance_m(*vtr1_centre, *site)
print(f"{d / 1000:.1f} km, inside VTR1: {d <= 10 * NM_TO_M}")

15.8 km, inside VTR1: True

Testing functions

assert checks that a condition is true; if not, the program stops and reports it immediately. Use it with known answers, including boundary values and impossible inputs.

assert battery_status(16) == "OK"
assert battery_status(15) == "LOW"
assert battery_status(7) == "CRITICAL"
assert math.isclose(distance_m(0, 0, 0, 1), 111_194.9, abs_tol=0.1)

try:
    battery_status(-1)
except ValueError:
    print("all tests passed")

all tests passed

As a project grows, move tests into test_*.py files and run them with pytest, which requirements.txt already installs.

# test_battery.py
import pytest
from mission_tools import battery_status

def test_boundaries():
    assert battery_status(15) == "LOW"
    assert battery_status(7) == "CRITICAL"

def test_rejects_impossible_values():
    with pytest.raises(ValueError):
        battery_status(120)

Run python -m pytest in the work folder.

Module lab

Common mistakes

Summary

• if/elif/else makes decisions; the order of conditions changes the answer
• for loops over data; zip(a, a[1:]) pairs neighbouring items
• Lists, tuples and dicts hold different kinds of mission data
• Good functions do one job, have docstrings, and raise errors on impossible input
• Test with assert and pytest, focusing on boundaries and known answers

Check your understanding

1. What does battery_status(15) return, and why?
2. If percent <= 15 were checked before percent <= 7, what would 5 return?
3. What is list(zip([1, 2, 3], [2, 3]))?
4. How long is the route (0, 0) → (30, 40) → (30, 100) in metres?
5. Why should you not write def add_point(points=[])?