DAPL — Data Analysis Programming Language

Interpreter for DAPL (Data Analysis Programming Language), a small statically-typed language designed for tabular-data manipulation. Built as a course project at AGH for Teoria kompilacji i kompilatory (Compiler Theory & Compilers), June 2025.

Authors

Co-authored with Jan Rolka, Mateusz Sobiech, and Maciej Szymański.

Language at a glance

DAPL combines a familiar imperative core (functions, if/elif/else, while, for, scoped variables) with first-class support for Table and List. Types: int, double, string, bool, plus auto for inferred locals; the ^ operator reaches into an enclosing scope.

A quick taste — joining two tables:

Table <int:empid, string:name, int:deptid> Employees = [
    [1, "Ala",     10],
    [2, "Bartek",  20],
    [3, "Celina",  10],
    [4, "Damian",  30]
];
Table <int:deptid, string:deptname> Departments = [
    [10, "Kadry"],
    [20, "IT"],
    [30, "Sprzedaż"],
    [40, "Marketing"]
];
Employees.join(Departments, deptid, deptid).show();

Supported table operations: filtering, grouping, aggregations, lambda expressions, joins.

Architecture

Implemented in Python with ANTLR4 for lexing and parsing. The interpreter walks the parse tree using two visitor passes:

• FirstVisitor — declaration collection (functions, type signatures)
• VariableListner — execution

The compiler pipeline lives under src/:

src/
├── DAPL.g4                # ANTLR grammar
├── DAPLLexer.py           # generated lexer
├── DAPLParser.py          # generated parser
├── DAPLListener.py        # generated listener
├── DAPLVisitor.py         # generated visitor base
├── FirstVisitor.py        # declarations pass
├── VariableListner.py     # execution pass
├── ErrorListener.py       # parse-error reporting
├── errors.py              # runtime errors
├── main.py                # CLI entrypoint
├── program.py             # program object
└── examples/              # sample DAPL programs (numbered)

Setup & usage

python3 -m venv env
source env/bin/activate          # Linux/macOS — Windows: env\Scripts\activate
pip install -r requirements.txt  # antlr4-python3-runtime

Run a DAPL program:

cd src
python main.py examples/employees

Documentation

Two reports ship with the source, both in Polish:

• RAPORT_IMPLEMENTACJI.md — implementation report (architecture, design decisions, type system)
• DOKUMENTACJA_UZYTKOWNIKA.md — user manual (syntax, types, examples)

PDF copies (DAPL-raport_techniczny.pdf, DAPL-dokumentacja-uzytkownika.pdf) are also included.

A note

Course project — the language is intentionally narrow (no module system, limited I/O beyond CSV import) but the parser and interpreter scaffolding are straightforward to extend.