dev-blog

Let’s break down how to construct a data model for system elements in a procedural programming style. I’ll focus on principles, steps, and concrete rules you can follow, keeping everything aligned with the procedural mindset.

---

1️⃣ Identify the Core Entities (Structs)

In procedural programming, data lives in simple containers (structs in C, records in Pascal, etc.).

Rules:

• Each entity should represent a single “thing” in your system.
• Only include data, no functions inside.
• Prefer flat structures (nested only if truly necessary).
• Give meaningful names to structs and fields.

Example: Game Element

typedef struct {
    int id;
    float x, y;
    float width, height;
    int health;
} GameObject;

---

2️⃣ Define Separate Functions to Operate on Data

• Functions should take structs as input and either return new structs or modify the struct explicitly.
• Avoid hidden state changes; everything the function modifies should be passed explicitly.
• Keep each function single-purpose.

Example: Movement Function

GameObject MoveObject(GameObject obj, float dx, float dy){
    obj.x += dx;
    obj.y += dy;
    return obj;
}

Example: Damage Function

GameObject ApplyDamage(GameObject obj, int dmg){
    obj.health -= dmg;
    if(obj.health < 0) obj.health = 0;
    return obj;
}

---

3️⃣ Use IDs for References Instead of Pointers (Optional)

If entities need to reference each other:

• Store IDs instead of pointers to avoid hidden dependencies.
• Use a central lookup table (array or map) for entity access.

typedef struct {
    int id;
    int targetId; // points to another GameObject
} Missile;

---

4️⃣ Keep Data Normalized

• Avoid duplicating information.
• Split data into logical chunks (position, velocity, stats, etc.) if reused.

Example: Position and Velocity as Separate Structs

typedef struct { float x, y; } v2;
typedef struct { v2 pos; v2 vel; float mass; } Physics;
typedef struct { int id; Physics physics; int health; } Entity;

---

5️⃣ Design for Explicit Updates

• Each system (physics, rendering, AI) should explicitly update data.
• Use loops over arrays of structs instead of hidden internal loops.

for(int i = 0; i < entityCount; i++){
    entities[i] = SimulateEntity(entities[i], dt);
}

---

6️⃣ Keep Data Simple, Functions Composable

• Flat arrays are often better than deeply nested structures.
• Small, pure functions are easier to test and reuse.
• Avoid mixing responsibilities; separate physics, graphics, AI data.

---

✅ Procedural Data Modeling Checklist

1. Structs contain only data.
2. Functions take structs as input, return modified structs or explicitly mutate.
3. Use IDs or indices for references, not hidden pointers.
4. Keep data normalized; avoid duplication.
5. Systems operate over arrays/lists of structs.
6. Functions are small, testable, deterministic.
7. Avoid hidden state; all dependencies are explicit.

---

If you want, I can make a ready-to-use template for a procedural-style system element with multiple components (physics, graphics, AI) that follows all these rules. It would be like a “starter kit” for your procedural engine.

Do you want me to do that?