LI3 - FlyData

A HIGLY specific database for travel data (flights, reservations, users, aircrafts, airports, passengers), written in C with GLib and ncurses with great performance to a non practical and not so logical extent. Built for Laboratórios de Informática III (2nd year, Software Engineering — Universidade do Minho).

---

Requirements

Dependency	Notes
GCC	compiled mainly in C99 but who cares
GNU Make	Standard build toolchain
GLib 2.x	Data structures and utilities
ncursesw	terminal UI

---

Build

From trabalho-pratico/:

make

This produces three binaries from the same sources:

Binary	Purpose
programa-principal	Batch query executor
programa-interativo	Interactive ncurses TUI
programa-testes	Automated regression tester

---

Usage

Batch mode

Reads queries from an input file and writes results to resultados/:

./programa-principal <dataset_path> <input_path>

./programa-principal dataset-fase-2/sem_erros inputs_fase2.txt
./programa-principal dataset-fase-1/sem_erros inputs_fase1.txt

Interactive TUI

./programa-interativo

Launches a terminal interface for selecting datasets and running queries interactively.

Regression testing

Runs all queries and diffs results against the expected outputs:

./programa-testes <dataset_path> <input_path> <expected_outputs_dir>

./programa-testes dataset-fase-2/sem_erros inputs_fase2.txt resultados-esperados

---

Cleaning

make clean                      # Remove build directory and binaries
make clean-resultados           # Remove generated query outputs in resultados/
make clean-resultadosesperados  # ⚠ Remove expected outputs — this deletes the regression ground truth

---

Repository Layout

trabalho-pratico/
├── src/
│   ├── main.c
│   ├── program.c
│   ├── controller/          # Query dispatch and execution
│   ├── model/               # Entities: aircraft, airport, flight, passenger, reservation
│   ├── parser/              # Per-entity CSV parsers (threaded)
│   ├── validator/           # Per-entity validation logic
│   ├── view/
│   │   ├── batch.c          # Batch output formatting
│   │   ├── csv_errors.c
│   │   └── interactive/     # ncurses TUI (menus, components, terminal)
│   ├── tests/               # Regression test runner logic
│   └── util/                # Shared utilities (date, radix sort, string pool, chunked array)
├── include/                 # Public headers mirroring src/ structure
├── assets/                  # ASCII art for TUI (logo, about screen)
├── dataset-fase-1/          # Phase 1 dataset (sem_erros/, com_erros/)   [gitignored]
├── dataset-fase-2/          # Phase 2 dataset (sem_erros/, com_erros/)   [gitignored]
├── resultados/              # Generated batch outputs                     [gitignored]
├── resultados-esperados/    # Expected outputs for regression (1/, 2/)   [gitignored]
├── inputs_fase1.txt         # Sample queries — Phase 1                   [gitignored]
├── inputs_fase2.txt         # Sample queries — Phase 2                   [gitignored]
├── inputs_prof.txt          # Professor query set                        [gitignored]
├── relatorio-fase1.pdf      # Phase 1 report (Portuguese)
├── relatorio-fase2.pdf      # Phase 2 report (Portuguese)
├── enunciado.pdf            # Project statement (Portuguese)
├── Doxyfile                 # Doxygen config
└── Makefile

Note: Gitignored items (datasets, inputs, outputs) are excluded to keep the repository slim. Obtain datasets separately and place them at the paths above before running

---

Documentation

Generate HTML API docs with Doxygen:

doxygen Doxyfile

Output is written to docs/html/. Open index.html in a browser.

---

Reports

• relatorio-fase1.pdf — Phase 1 design and implementation report
• relatorio-fase2.pdf — Phase 2 design and implementation report
• enunciado.pdf — Original project statement and requirements

---

Authors

• Ana Beatriz Rodrigues - A110627
• Guilherme Ferreira - A111042
• Gonçalo Pombal - A110681