Skip to content

ARCHITECTURE.md

Latest commit

 

History

History
83 lines (66 loc) · 2.16 KB

ARCHITECTURE.md

File metadata and controls

83 lines (66 loc) · 2.16 KB

Optees Architecture

Optees is a desktop, open-source tool to help technical and non-technical staff solve optimization problems with a clean, testable, and scalable architecture.

Goals

  • Clear separation of concerns (Clean Architecture)
  • Easy to test (TDD), easy to extend
  • Source-agnostic core (files, GUI, adapters)
  • Prefer robust, well-maintained libraries; don’t reinvent the wheel

Layered Model (6 layers)

Responsibilities

  • Domain Pure business logic & domain entities (Entities & Managers). No I/O, no library APIs.

  • Application Use Cases orchestrate flows (parse → preprocess → solve → analyze → format). Services wrap solver calls (LP, MILP, Knapsack…). No UI or file format knowledge.

  • Data Repositories + Adapters handling external resources: CSV/XLSX, MPS/MAT, Burkardt text, etc. Convert into canonical dicts for the Application layer.

  • Presentation GUI (e.g., PyQt) – Views/Controllers. Only talks to Application Use Cases and maps DTOs ⇄ UI widgets.

  • Core Project-specific shared utilities: logging, string management, configuration & constants, error types.

  • Utility Generic, reusable helpers: numerical routines, small algorithms. No project coupling.

Canonical LP Problem (example)

Adapters should convert external data into:

problem = {
  "sense": "min" | "max",
  "c": [...],
  "A_ub": [[...], ...], "b_ub": [...],     # optional
  "A_eq": [[...], ...], "b_eq": [...],     # optional
  "bounds": [[lb, ub], ...],               # optional; default (0, +inf)
  "var_names": ["x0","x1",...],            # optional
  "obj_offset": 0.0                        # optional
}

Canonical Knapsack (0/1) schema

knapsack = {
  "values": [v1, v2, ...],     # floats
  "weights": [w1, w2, ...],    # non-negative ints
  "capacity": int,             # non-negative
  # optional:
  "var_names": ["i0","i1",...]
}

Actually folders

assets/
src/optees/
  application/
  core/
  data/
    adapters/
    repositories/
  domain/
  presentation/
  utility/
    io_lpnetlib.py
    io_knapsack.py
    lp_utils.py
    knapsack_utils.py
tests/
  application/
  data/
  utility/
docs/