DBMS in C++

A file-based mini DBMS in C++ with an interactive CLI and a lightweight SQL parser. The system stores metadata and table data as CSV files inside the Databases folder.

Project Layout

SQL_IN_CPP/
  include/                # Header files
  src/
    caller/               # CLI entrypoint / callers
    core/                 # Database + table core logic
    handlers/             # SQL/query handlers (parser, joins)
  tests/                  # Query test input files
  docs/                   # Feature and SQL notes
  tools/                  # Standalone experiments/utilities
  Databases/              # Runtime storage (generated/updated at runtime)
  output/                 # Build output artifacts

What Is Implemented

• Multi-database management (create, select, list)
• Table lifecycle operations (create, rename, truncate, drop, list)
• Row insertion in two modes:
  • positional insert
  • insert with explicit column list (missing columns are stored as NULL)
• Data type validation at insert time
• SELECT queries with column projection
• JOIN queries:
  • INNER JOIN
  • LEFT JOIN
  • RIGHT JOIN
  • FULL JOIN
• SHOW commands:
  • SHOW TABLES
  • SHOW DATABASES
• Automatic bootstrapping of required storage files/folders at startup

Supported SQL Commands

The parser in src/handlers/sqlparser.cpp currently supports the following commands:

DDL

• CREATE TABLE
• RENAME TABLE ... TO ...
• DROP TABLE
• TRUNCATE TABLE

DML

• INSERT INTO table VALUES (...)
• INSERT INTO table (col1, col2, ...) VALUES (...)

DQL

• SELECT * FROM table
• SELECT col1, col2 FROM table
• SELECT ... FROM table1 <JOIN_TYPE> JOIN table2 ON colA = colB

Utility

• SHOW TABLES
• SHOW DATABASES

Data Types

The engine validates and stores these logical types:

SQL Type	Internal ID	Notes
INT	0	Parsed with stoi
FLOAT	1	Parsed with stof
BOOL	2	Accepts TRUE/FALSE/1/0
BOOLEAN	2	Alias of BOOL
STRING	3	Single-quoted strings supported
DATE	4	Expected format: YYYY-MM-DD

CLI Flow

The executable starts with:

1. Initialization of Databases/information_schema.csv and Databases/baseDb
2. Main menu:
   • Create Database
   • Select Database
   • Show Databases
   • Exit
3. Table operations menu (after selecting a database):
   • Create Table (via SQL)
   • Show Tables
   • Run SQL Query loop

Build and Run

# Build
make

# Optional: include mongo-c-driver libs if you need that integration
make USE_MONGO=1

# Run
./main

# Clean object files and binary
make clean

# Reset persisted database folder
make reset

On Windows, run the produced executable directly (for example main.exe).

Run in Browser via GitHub Pages

This repo can be deployed as an interactive WebAssembly app on GitHub Pages.

What is included

• A workflow at .github/workflows/deploy-pages.yml
• Automatic build on push to main/master and manual run support
• C++ sources compiled with Emscripten to index.html + .js + .wasm
• Databases/ preloaded into the browser filesystem at startup

First-time setup in your GitHub repo

1. Push this repository (including the workflow) to GitHub.
2. Open your repository on GitHub.
3. Go to Settings -> Pages.
4. Under Build and deployment, set Source to Deploy from a branch.
5. Set Branch to gh-pages and folder to /(root).
6. Go to the Actions tab and run (or wait for) Build and Publish GitHub Pages Site.

After deployment, your app will be available at:

https://<your-username>.github.io/<your-repo-name>/

Interaction notes

• The app runs fully in-browser as WebAssembly.
• The deployed page uses a minimal terminal-like UI (output pane + single input line).
• Input is captured asynchronously from the on-page terminal box (no browser prompt popups).
• Press Enter in the input box to send one full line to the running program.
• Runtime file changes are in browser memory for that session and are not pushed back to GitHub.

Example Queries

CREATE TABLE users (id INT, name STRING, age INT, is_active BOOL);
CREATE TABLE departments (id INT, department_name STRING);

INSERT INTO users (1, 'Alice', 30, TRUE);
INSERT INTO users (id, name) VALUES (2, 'Bob');

SELECT * FROM users;
SELECT id, name FROM users;

SELECT * FROM users INNER JOIN departments ON id = id;
SELECT * FROM users LEFT JOIN departments ON id = id;

SHOW TABLES;
SHOW DATABASES;

RENAME TABLE users TO app_users;
TRUNCATE TABLE app_users;
DROP TABLE app_users;

Storage Layout

Databases/
  information_schema.csv            # database_name,date_created
  <db_name>/
    tables.csv                      # table_name,date_created
    <table_name>/
      columns.csv                   # col_name,data_type
      data.csv                      # raw rows

Current Limitations

These are present in helper docs or placeholders but not implemented in active execution flow yet:

• WHERE, GROUP BY, HAVING, ORDER BY, LIMIT, OFFSET, DISTINCT
• UPDATE and DELETE
• ALTER TABLE
• CROSS JOIN and SELF JOIN (commented placeholders)
• PK/FK/constraint enforcement

Notes:

• Table and column identifiers are normalized to lowercase by the parser.
• JOIN queries print a temporary joined table result and then drop that temp table.

Key Source Files

• src/caller/main.cpp: menu-driven CLI
• src/handlers/sqlparser.cpp: query parsing and dispatch
• src/core/database.cpp: database creation/selection/listing
• src/core/table.cpp: table creation, insert, display, rename, drop, truncate
• src/handlers/join.cpp: inner/left/right/full join execution
• src/core/globals.cpp: datatype mappings and timestamp helper