No test suite existed before this, which is exactly how a real
regression (the bold-span coordinate-squashing bug, and the
'Type#id:'/'<ref>: <value>' header collision, both from this session)
went unnoticed until manually re-triggered. One test per format,
cross-referenced against the full commit history so nothing already
shipped gets silently dropped by a future change:
tests/test_ocr.py: standard blocks, the calibration target line,
destruction reports (digit and letter id), train-arrival intel,
ad-hoc Enemy installations (+ their destroyed reports), Listening
Post/Coastal Battery, Marine Garrison fire-support requests, multi-
word RP names, bare-name-header targets, the bold-span coordinate-
squashing regression specifically, forward-observer report
triangulation, the '<ref>: <value>' clue grammar (+ its header-
collision regression specifically), 16-point compass tolerance, and
grid-only coordinates.
tests/test_solver.py: direct bearing+distance resolution, two-bearing
triangulation, genuine two-distance ambiguity, the nested-circles
compromise-point fallback, toleranced bearings never being used to
triangulate, and manual coord overrides clearing a stale note.
Runs via ============================= test session starts ==============================
platform linux -- Python 3.14.6, pytest-8.4.2, pluggy-1.6.0
rootdir: /home/dodox/Projects/FeNigma
configfile: pytest.ini
plugins: anyio-4.13.0
collected 21 items
tests/test_ocr.py ............... [ 71%]
tests/test_solver.py ...... [100%]
============================== 21 passed in 0.72s ============================== (pythonpath configured in pytest.ini), dev-only
dependency in requirements-dev.txt so the app itself stays
dependency-light. Documented in the README.
Selecting text in-game copies it to the clipboard, screenshots were
never the only way in and are now explicitly the fallback for when a
selection isn't practical. Also mention scout flight planning and the
clipboard auto-watch toggle, both added since the last README pass.
- src/ironnest_assist/ -> src/fenigma/
- run.sh invokes -m fenigma.app
- APP_ID: eu.dominik-roth.IronNestAssist -> eu.dominik-roth.FeNigma
- window title, IronNestApp class -> FeNigmaApp
- __init__.py docstring, README pgrep hint updated
IRON NEST (the game's own name, e.g. NEST_KEYWORD in ocr.py) is left
untouched, only our own project/app naming changed. Verified: clean
import under the new module path, install.sh still builds the venv
correctly, and the OCR regression sweep (9 known screenshots) still
passes.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
install.sh creates a --system-site-packages venv (so it can still see
the system-installed PyGObject bindings, which pip can't build
reliably) and pip-installs the rest of requirements.txt into it.
Checks for PyGObject (GTK4 + libadwaita) and tesseract up front and
prints per-distro package hints if either is missing, rather than
failing deep into pip install. run.sh now prefers .venv/bin/python3
when it exists, falling back to system python3 otherwise. README gets
an Install section describing this.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Shorter pitch focused on what the app actually automates for the
player: screenshot orders -> geo puzzle + trajectory solved -> ready
fire commands; screenshot the field log -> kills/new contacts picked
up automatically; strike planning with blast-radius preview. Mentions
the map's geometric derivation overlays and ambiguous-intersection
handling. Drops the old deep technical dump in favor of pointing at
code comments for internals. Adds icon.png (used in the header) and
showcase.png (embedded screenshot); icon_alt.png kept as an unused
alternate for now.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
GTK4/libadwaita desktop helper for IRON NEST: Heavy Turret Simulator.
Reads clipboard screenshots of the game's typewriter orders via Tesseract
OCR, parses absolute/relative entity positions, geometrically resolves
relative bearing/distance clues into map coordinates, and provides a
firing-commands sidebar with real ballistics (elevation/azimuth/powder
charge). Screen-reading only — no game files touched, no input injected.
- models.py: Board/Nest/Spotter/ReferencePoint/Target data model
- ocr.py: Tesseract preprocessing + typewriter-text parsing
- solver.py: bearing/distance geometric resolution (4 solvable shapes)
plus position-based dedup for generic contacts later identified more
specifically at the same resolved coord
- ballistics.py / shells.py: elevation/azimuth/charge math, shell types
- grid_widget.py: interactive map canvas (hidden entities and,
optionally, dead targets are excluded from the map view entirely
unless selected)
- firing_panel.py: drag-reorderable firing-command sidebar
- app.py: main window wiring it all together
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
