Python for beginners

Where do you get Python, and how do you know it installed?

From the python.org downloads page and nowhere else. On macOS and most Linux systems Python is already there and you need to install nothing; on Windows you install it.

After the install one job is left, and it is where most beginners get stuck: make sure the terminal can see Python. Open a terminal and type this:

python3 --version

If a version number is printed, you are done. On the very server where this lesson is being written, the answer to that command is Python 3.12.3.

And now the trap no tutorial mentions: on many systems there is no command called python at all. On this server python answers "command not found" while python3 works. So when a tutorial shows you python file.py and nothing happens, the mistake is not yours; add the 3.

Windows has a third story. Python own documentation says that when several versions are involved, the recommended command is py. So on Windows try py --version first, and then use that same py for the rest of the commands in this lesson.

py --version

If none of those three commands answers, the problem is not the install; it is that the operating system does not yet know where Python is. The fix differs per system, and the official Windows guide has the Windows version of it.

The first script: from an empty file to output in the terminal

Make a text file and call it hello.py. Any text editor works and today is not the day to shop for tools. Write one line in it:

print("Hello")

Now open a terminal in that folder and run it:

python3 hello.py

The word Hello is printed. That is all. If nothing was printed, check three things in this order: you are in the folder the file is in, the file name really ends in py and not in py.txt, and the command you typed is the one that answered in the previous section.

There is also one rule that separates Python from other languages, and it is better to meet it now: in Python the space at the start of a line means something. There are no braces and no semicolons; you show that something belongs to a block by indenting it, and the whole language is built on that convention.

if 5 > 3:
    print("bigger")

Those four spaces before print are not decoration, they are syntax. Take them away and Python raises IndentationError. Our advice is to set your editor to insert four spaces for tab, and then never think about it again.

Input and output: a program that talks to the user

A program that only prints is a note. input turns it into something that gets an answer back:

name = input("Your name: ")
print("Hello,", name)

Three things live in those two lines. input prints the message you give it and waits, it returns whatever the user typed, and you keep that answer in the variable name so you can use it later.

Now the most important point in this section, and if you do not know it, it becomes the first real error of your programming life: input always returns a string. Python own documentation says exactly that; even when the user types 17, what reaches you is the text "17" and not the number 17.

age = input("Age: ")
print(type(age))

The output of those two lines is <class 'str'>. Which means that if you want to add one to it, Python raises an error and you have to convert it first:

age = input("Age: ")
print(int(age) + 1)

Why is it like this? Because Python does not guess. If it turned strings into numbers by itself, a program taking a postcode or a card number would lose its leading zeros. The conversion is yours because only you know what that string is supposed to mean. This matters enough that the next lesson in this track is entirely about data types, and the order of the lessons is on the track page.

Your first real program: count the length of a text

Now we write something we genuinely use ourselves. The meta description of every page on this site has to be between 140 and 160 characters; that is our own house rule and it is written in this site content guide. Counting that by eye is hard work, and six lines of Python make it a ten second job.

text = input("Meta description: ")
n = len(text)

if n < 140:
    print(f"{n} characters. Too short, {140 - n} to go.")
elif n > 160:
    print(f"{n} characters. Too long by {n - 160}.")
else:
    print(f"{n} characters. Good.")

We ran this three times before shipping this lesson and its real output was: for a 90 character text, 90 characters. Too short, 50 to go.; for 150 characters, 150 characters. Good.; and for 200 characters, 200 characters. Too long by 40.

There are four new things in those six lines. len gives the length of a string. if, elif and else build a three way branch and exactly one of them runs. The f before the quote means you may put braces inside the string and Python works out what is in them; it is called an f-string and it is the easiest way to build text. And arithmetic works inside those braces too, as in {140 - n}.

One exercise worth doing right now: change the condition so that it counts words rather than characters. Hint: len(text.split()). If you changed that yourself and it worked, you are no longer somebody who has taken a tutorial; you are somebody who has written a program.

Got an error? Read the traceback from the bottom

The skill that separates a beginner from somebody who really writes programs is not writing code without errors; it is reading errors. When a program falls over, Python prints a traceback, and that frightening block of text is in fact the most precise friend you have.

Make this file, call it shop.py, and run it:

def price(n):
    return n * 1.09

print(price("100"))

And this is exactly what gets printed. The real run, not what ought to be printed:

Traceback (most recent call last):
  File "shop.py", line 4, in <module>
    print(price("100"))
          ^^^^^^^^^^^^
  File "shop.py", line 2, in price
    return n * 1.09
           ~~^~~~~~
TypeError: can't multiply sequence by non-int of type 'float'

Now the rule: read it from the bottom. The last line says what happened; here a data type was wrong. The line above it says where it happened: line 2, inside the function price. And above that, who called that place: line 4. So the error appeared on line 2 but its cause is on line 4, where we sent a string instead of a number.

That order holds for every traceback, which is why the habit of reading upwards is worth building today. Read the last sentence first, then look at the lowest line that belongs to your own file, and only then look at the rest.

The first three errors of anybody life are these: SyntaxError means you wrote the sentence wrongly and it is usually a missing bracket or colon; NameError means you called a name that does not exist and most of the time it is a typo; and TypeError is the one above, meaning the type of the data does not fit what you are doing with it.

What comes next: a plan that does not waste your time

The most important thing you need after this lesson is not another tutorial; it is a small program you actually need. But there are four things whose order matters, and the figure below shows that order.

First, the standard library. Python installs with a warehouse of ready code, and most everyday jobs already have a function. Before writing any loop, search the official Python tutorial once to see whether the job is already done.

Then installing outside libraries with pip, and with it the virtual environment. These two lines keep your project separate from the rest of the system:

python3 -m venv .venv
source .venv/bin/activate

On Windows the second line differs and the venv documentation has the correct form. Why bother? Because without it every project you build drops its libraries in the same shared place, and six months later two projects want two different versions of one library while your system has only one.

And finally, when your program passes a hundred lines, move to writing functions and then to classes. Not sooner. Learning classes before you have a pain that classes cure is the mistake that talks a lot of people out of programming.