Structs und Python-Klassen im Vergleich

A Mojo struct and a Python class both bundle data with methods, so they feel familiar. But under the hood they are quite different. 🔍

A struct's layout is fixed at compile time. The compiler knows every field and type ahead, so it generates tight, fast code.

Python classes are dynamic: you can add attributes at runtime. That flexibility is convenient but costs lookups and speed.

In a struct you cannot tack on a new field after definition. Mojo rejects it, which catches typos and keeps the layout honest.

Structs are value types, so assigning copies them. Python objects are references, so two names can point to the same object.

Because structs copy on assignment, changing one instance never silently mutates another. That removes a whole class of aliasing bugs.

Static layout and value semantics let Mojo skip runtime overhead. The result is performance close to C for your custom types.

Structs do not use class-style inheritance. Instead Mojo shares behavior through traits, which you will explore later in the path.

Every struct field needs a declared type, while Python attributes are untyped by default. Types are what let Mojo optimize so aggressively.

Reach for a struct when you want speed and a fixed shape. Keep Python classes when you need runtime flexibility from existing libraries.

The takeaway: structs borrow Python's friendly syntax but run on a static, compiled engine. You write familiar code that goes fast.

Let us compare structs and Python classes.

Structs trade Python's runtime flexibility for a static, value-based design that runs fast. You now know when each one shines. 🎯