Welcome to the DBMS project! This project features a subset of SQL and implements a database engine to process the SQL queries.

Follow the installation steps below:

git clone https://github.com/nat-leo/dbms.git

You NEED to update pip to the latest version, and use the latest version of Python (3.11 as of 9/18/23) Or you'll get build errors when running pip install. First, update Python, second:

pip install --upgrade pip

pip3 install --upgrade pip3

In order to get access to the REPL:

cd dbms
pip install .

OR if you're using python3 and pip3:

cd dbms
pip3 install .

And that should do it! Run either command below to start the REPL:

db -l

OR

db --login

db --login

eate or login to a user with the name test and password password, then enter details when prompted:

% db --login
user: test
password: password
WARNING: t_ignore contains a literal backslash '\'
enter Q to exit
sql > 

Create a table easily with SQL:

sql > CREATE TABLE table (col varchar(255))
None

Add some data to your newly created table:

sql > INSERT INTO table (col) VALUES (bob)
None
sql >  INSERT INTO table (col) VALUES (bill)   
None
sql >  INSERT INTO table (col) VALUES (babs)
None

Currently you can only enter one row at a time, and will add an empty row:

sql > INSERT INTO table (col) VALUES (bob, bill, babs)
An Exception Occurred: list index out of range
None

Read the data you just wrote, including the empty row:

sql > SELECT * FROM table
[{'col': 'bob'}, {'col': 'bill'}, {'col': 'babs'}, {'col': ''}]

To quit:

sql > q

There are a few quirks before we begin. First is that your data is stored in the engine dirctory (as opposed to var/lib or wherever):

dbms/ # the repo
├── ...
├── ...
├── ...
├── engine/ # the Database Engine
    ├── ...
    ├── ...
    ├── ...
    ├── database/ # user prompt in the REPL
        ├── hash.txt
        ├── salt.bin
        ├── table/
            ├── schema.json 
            └── table.bin # the actual data

database/ is what you enter into the user: prompt when you initially start up the repo with:

db --login

table/ is the directory that holds the relation's metadata and data. When you run CREATE TABLE table, you get this directory and the following:

schema.json holds the schema of your table, including the datatype and size in bytes of your columns.

table.bin has the actual data. SQL queries read, write, and modify this file.