screen-locker/pyproject.toml
Krzysztof Rudnicki 3e77f4e232 Make every lock screen keyboard-operable, and add the missing installer
The locker was not running at all. workout-locker.service crash-looped ~1977
times in a single boot with ModuleNotFoundError while remaining `enabled`, and
has no successful start in the retained journal. A locker that cannot start is
a locker that does not lock.

Root cause: commit 9b420a9 removed the unit's WorkingDirectory and
PYTHONPATH=/opt/screen-locker on the (correct in principle) grounds that
"screen_locker is now pip-installed" -- but nothing guaranteed that install.
There was no installer, /opt/screen-locker no longer exists, and the package
was absent from every Python version's user site-packages. The unit lost its
fallback and gained a hard dependency on a manual step nobody re-ran.

install.sh fixes that as a script rather than a reminder, and *gates*: it fails
if the production interpreter cannot import screen_locker, because an installer
that reports success while the service still cannot start is precisely the
failure above. It also re-asserts an editable gatelock from ~/utils AFTER the
package install and verifies it resolves there -- pyproject pins gatelock to a
git tag, so a plain install silently replaces a local editable gatelock with a
frozen wheel, which was observed removing ScrollableSurface and taking
diet_guard down with it.

The keyboard work, all of it Tk defaults rather than local mistakes:

- Focus was invisible. Tk ships highlightcolor="#000000" -- a black ring on
  #211D1B -- and relief="flat" removes the depth cue too, so a keyboard user
  had no indication of where they were. Now set in the _ui_widgets factory, so
  a new call site cannot omit it.
- Enter did nothing. X11 Tk binds <space> on Button and never <Return>.
- tk.Text trapped Tab (inserts a literal tab, refocuses itself) and no-opped
  Shift-Tab; the only exits are the undocumented Ctrl+Tab / Ctrl+Shift+Tab.
- The lock and retry screens started with NOTHING focused. on_focus_ready was
  a documented no-op arguing "buttons only, nothing to focus" -- exactly
  backwards: buttons are what needs focusing when there is no other window on
  screen and a global grab is held.
- The heat-skip modal grabbed input fullscreen with no focus, no ring, no
  Return and no Escape. It now focuses the *declining* option, so a blind
  Return keeps the workout rather than silently skipping it.

The manual-workout form's scroll viewport was pointer-only in three ways at
once: tk.Canvas has no class key bindings, so ::tk::FocusOK rejected it as a
focus stop; nothing bound wheel, PageUp/PageDown or arrows, leaving only a
draggable thumb; and canvas clipping does not unmap children, so below-the-fold
fields stayed in the tab ring while Tk never scrolled to follow focus -- Tab led
to fields the user could neither see nor reveal. Its height was also a magic
0.7*toplevel above ~230-290px of fixed chrome, i.e. satisfied at 1080p and
violated at 768p, where SUBMIT/BACK were pushed off the bottom.

All of that is gone by deletion: the surface container is now a
gatelock.ScrollableSurface, so the inner viewport is redundant. build_surface
and _ensure_container both use it, which fixes the same symmetric-clipping
problem for EVERY screen rather than one at a time; center_when_fits keeps
short screens looking exactly as they did.

Re-arming after a repaint hooks into clear_container(), which all ~19 repaint
sites already call. "Remember to call settle after painting" is a rule that
would eventually be missed, and on a lock a missed re-arm means unreachable
content.

The aux windows hardcoded geometry("600x400") and ("700x450") while displaying
more than that -- the relaxed-day prompt's own button row requests 862px inside
a 700px window, sheared at both edges. That pre-dates this work by a wide
margin (Tk's default focus ring accounts for 4px of the 162px overflow). They
now size to content, floored at the old values and clamped to the display.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-07-30 16:25:45 +02:00

232 lines
7.9 KiB
TOML

[project]
name = "screen-locker"
version = "1.0.0"
description = "Tkinter/systemd screen locker with workout tracking, sick-day management, and wake-alarm integration"
requires-python = ">=3.10"
# gatelock: shared lock-window/HMAC backend, also used by diet_guard and wake_alarm
# crdt-sync: shared CRDT merge scheme + GitHub-Contents-API sync transport,
# also used by diet_guard; here for phone-workout GitHub sync (Workstream C)
dependencies = [
"gatelock @ git+https://github.com/kuhyx/utils@gatelock-v0.3.0#subdirectory=gatelock",
"crdt-sync @ git+https://github.com/kuhyx/utils@crdt-sync-v0.1.0#subdirectory=crdt-sync",
]
[project.scripts]
screen-locker-status = "screen_locker.status_view:main"
# Optional extras. The `mcp` group is ONLY needed by the read-only MCP server
# (screen_locker._mcp), which Claude Code spawns via a dedicated venv (see
# scripts/setup_mcp.sh / .mcp.json). It is deliberately NOT a core dependency
# so the systemd/CLI system-python path stays dependency-light and never has
# to import the MCP SDK.
[project.optional-dependencies]
mcp = ["mcp>=1.9"]
[tool.setuptools.packages.find]
# This repo also holds stronglift_replacement/ (an unrelated Flutter app, not
# a Python package); without this, setuptools' automatic flat-layout
# discovery sees two top-level directories and refuses to guess which is the
# installable package.
include = ["screen_locker*"]
[tool.ruff]
target-version = "py310"
include = ["*.py", "**/*.py"]
exclude = [".git", ".venv", "__pycache__", "build", "dist", ".eggs"]
[tool.ruff.lint]
select = ["ALL"]
ignore = [
"D203",
"D213",
"COM812",
"ISC001",
"S603",
]
fixable = ["ALL"]
unfixable = []
[tool.ruff.lint.per-file-ignores]
"**/tests/**/*.py" = [
"ANN", "ARG", "D", "DTZ011", "E501", "FBT",
"PLC0415", "PLR0913", "PLR2004", "PTH", "RUF003", "S101", "SLF001",
]
"**/test_*.py" = [
"ANN", "ARG", "D", "DTZ011", "E501", "FBT",
"PLC0415", "PLR0913", "PLR2004", "PTH", "RUF003", "S101", "SLF001",
]
"screen_locker/_runnerup_verification.py" = ["S314"]
# URL is always built from the hardcoded https:// _WTTR_URL constant with
# only the (quoted) city interpolated — never user/attacker controlled.
# ruff's S310 can't verify that via static analysis of a variable URL, so
# it always flags urlopen() here regardless; a runtime scheme check was
# considered and rejected (it can't ever trigger, so it's just an untested
# branch and a crash risk in the lock's startup path for zero real benefit).
"screen_locker/_temperature.py" = ["S310"]
# PERF203 (try/except inside a loop) describes a retry loop exactly, and lands
# where its cost is irrelevant: each iteration makes a network call and then
# sleeps for seconds, so handler setup is noise. Hoisting the handler out of
# the loop would mean swallowing the error into a return value, which the
# repo's no-silent-failures rule rightly discourages.
"screen_locker/_sync_retry.py" = ["PERF203"]
"screen_locker/_shutdown_base.py" = ["SLF001"]
# run_status walks the weekly-status pipeline (weekly logic, RunnerUp scan,
# bonus, heat-skip) as a sequence of stages — genuinely branchy by design, same
# rationale as the validation-chain functions below; not a config-object case.
"screen_locker/_status.py" = ["C901", "PLR0912", "PLR0915", "SLF001", "T201"]
# _heat_skip.py opens a tkinter fullscreen window; tk's .attributes()/other
# calls take positional booleans (its API), which FBT003 always flags.
"screen_locker/_heat_skip.py" = ["FBT003"]
# StatusWindow.__init__ takes one arg per independently-mocked dependency
# (verifier_factory, temperature_fetcher, log_file, ...) — same
# dependency-injection rationale as _status_data.py's gather_status below.
"screen_locker/status_view.py" = ["PLC0415", "PLR0913", "SLF001", "T201"]
# The verify-and-credit flow split out of status_view.py: it deliberately
# drives the bare verifier's own mixin methods (_verify_phone_workout,
# _verify_runnerup_workout, _apply_workout_credit) — the same single-source-of
# -truth reuse status_view.py above is exempted for — and _make_bare_verifier
# imports ScreenLocker inside the function to avoid an import cycle.
"screen_locker/_status_view_verify.py" = ["PLC0415", "SLF001"]
# explain_lock_decision takes one arg per real file path + sick history, and
# returns early per lock-decision-chain stage (see module docstring) — both
# are the deliberate design, not something to collapse into a config object.
"screen_locker/_compliance_state.py" = ["C901", "PLR0911", "PLR0913"]
# gather_status takes one arg per real file path, all independently
# overridable for tests — same rationale as _compliance_state.py above.
"screen_locker/_status_data.py" = ["PLR0913"]
# validate_manual_workout returns early per failed evidence-form check (same
# validation-chain shape as _compliance_state.py's explain_lock_decision) —
# deliberate design, not something to collapse into a config object.
"screen_locker/_manual_workout.py" = ["C901", "PLR0911"]
# _verify_phone_workout returns early per verification-chain stage (clock
# check, sync, ADB, HTTP, then terminal statuses) — same shape as the two
# ignores above, not something to collapse into a config object.
"screen_locker/_phone_verification.py" = ["C901", "PLR0911"]
"scripts/check_file_length.py" = ["PTH123"]
[tool.ruff.lint.pydocstyle]
convention = "google"
[tool.ruff.lint.isort]
force-single-line = false
force-sort-within-sections = true
known-first-party = ["screen_locker"]
[tool.ruff.lint.flake8-quotes]
docstring-quotes = "double"
inline-quotes = "double"
[tool.ruff.lint.flake8-tidy-imports]
ban-relative-imports = "all"
[tool.ruff.format]
quote-style = "double"
indent-style = "space"
skip-magic-trailing-comma = false
line-ending = "auto"
docstring-code-format = true
[tool.mypy]
python_version = "3.10"
strict = true
warn_return_any = true
warn_unused_configs = true
disallow_untyped_defs = true
disallow_incomplete_defs = true
check_untyped_defs = true
disallow_untyped_decorators = true
no_implicit_optional = true
warn_redundant_casts = true
warn_unused_ignores = true
warn_no_return = true
warn_unreachable = true
disallow_any_unimported = true
disallow_any_explicit = false
disallow_any_generics = true
disallow_subclassing_any = true
strict_equality = true
extra_checks = true
ignore_missing_imports = true
show_error_codes = true
color_output = true
exclude = [".venv/"]
[tool.pylint.main]
analyse-fallback-blocks = true
persistent = true
jobs = 0
py-version = "3.10"
ignore = [".venv", "__pycache__"]
ignore-patterns = [".*\\.pyi$"]
[tool.pylint.messages_control]
enable = "all"
disable = []
[tool.pylint.design]
min-public-methods = 0
max-module-lines = 1000
max-attributes = 10
[tool.pylint.typecheck]
generated-members = [
".*\\.assert_called_once_with",
".*\\.assert_called_once",
".*\\.assert_called",
".*\\.assert_not_called",
".*\\.assert_any_call",
".*\\.call_args",
".*\\.call_args_list",
".*\\.call_count",
]
[tool.bandit]
exclude_dirs = ["tests", ".venv"]
[tool.pytest.ini_options]
testpaths = ["screen_locker/tests"]
python_files = ["test_*.py", "*_test.py"]
python_classes = ["Test*"]
python_functions = ["test_*"]
addopts = [
"-v",
"--strict-markers",
"--strict-config",
"-ra",
"--cov=screen_locker",
"--cov-branch",
"--cov-report=term-missing",
"--cov-report=lcov",
# A fixture-only plugin module, not conftest.py (already at the
# repo's 400-line-per-file cap) -- see its own docstring.
"-p",
"screen_locker.tests.test_sync_fixtures",
]
filterwarnings = [
"error",
"ignore::DeprecationWarning",
"default::pytest.PytestUnraisableExceptionWarning",
]
[tool.coverage.run]
source = ["screen_locker"]
branch = true
omit = [
"*/__pycache__/*",
"*/tests/*",
"*/.venv/*",
]
[tool.coverage.report]
fail_under = 100
show_missing = true
skip_covered = false
exclude_lines = [
"pragma: no cover",
"raise NotImplementedError",
"raise AssertionError",
"if TYPE_CHECKING:",
'if __name__ == "__main__":',
]
partial_branches = ["pragma: no branch"]