Stage 14: CREATE TABLE & Schema Catalog

Concept

Replace the hardcoded "users" table with a dynamic schema system. The student implements CREATE TABLE and stores table definitions in a schema catalog — a special system table that persists across sessions.

What It Teaches

DDL vs DML: CREATE TABLE is a Data Definition Language command — it changes the database structure, not the data. The student's parser now handles two fundamentally different command types.
System catalog: Every real database has internal tables that describe the schema. PostgreSQL has pg_class, pg_attribute. SQLite has sqlite_master. The student builds their own.
Catalog storage: The catalog is stored in a reserved B-tree (page 0 or a special system page). On startup, the database reads the catalog to know which tables exist.
Schema persistence: Table definitions must survive across sessions, just like data. The catalog is stored using the same pager/B-tree infrastructure.

Learning Objectives

Extend the parser to recognize CREATE TABLE name (col1 type1, col2 type2, ...).
Define a catalog entry format: [table_name][column_count][columns...].
Store catalog entries in a dedicated B-tree (system table).
On startup, read the catalog to populate an in-memory table registry.
Implement the catalog list debug command.
Reject duplicate table creation with an error.
Verify that INSERT and SELECT still work after CREATE TABLE.

New SQL Syntax

CREATE TABLE users (id INT, name VARCHAR, email VARCHAR);
CREATE TABLE products (id INT, title VARCHAR, price INT);

Debug Command: `catalog list`

[CATALOG] Tables:
[CATALOG]   users (id INT, name VARCHAR, email VARCHAR)
[CATALOG]   products (id INT, title VARCHAR, price INT)

How to Run

make test-c-stage14