mirror of
https://github.com/kuhyx/diet-guard.git
synced 2026-11-11 13:47:38 +01:00
Krzysztof Rudnicki
4a5e350baf
gatelock 0.4.0 (pinned here) stops the viewport following the app's own focus_set(): on 2026-08-03 that moved the screen locker's lock screen to mid-content and back with nobody touching the keyboard, and this gate uses the same viewport. Scrolling now happens only on the user's own keypress, wheel or click. That makes fitting the screen the gate's own responsibility rather than something a scrollbar covers for, and with a day of meals logged this one did not: 770px of content in a 739px pane. Trimmed the vertical rhythm (the title is one step down the scale, and the gaps around the description, suggestions, macros and footer are one step tighter) for 718px, with room for the dashboard to keep growing. Its layout test now requires the gate to *fit* rather than merely be reachable by scrolling, empty and populated. Dropped the 1024x600 tier: no machine runs it, this gate cannot reach it without dropping content, and a target nobody uses is a constraint that only ever costs. The probe presses Tab before moving focus, because a bare focus_set() no longer scrolls -- it would have been testing a path no user can take. The Tk fakes learned pack_forget(), which the viewport calls to take the scrollbar down whenever the content fits. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BsXnTHyuuCUZsqetTaajcF
79 lines
3.6 KiB
Python
79 lines
3.6 KiB
Python
"""The gate must be fully keyboard-reachable at every supported screen size.
|
|
|
|
This is the *gate*, not a report: it fails rather than printing numbers. It ran
|
|
red before the fix and green after, which is the only evidence that a check
|
|
tests anything at all.
|
|
|
|
What it caught, measured under a real 1366x768 X server: the Log Meal tab
|
|
required 866px inside a 739px content pane -- and 938px with a day of meals
|
|
logged -- while being ``place``-centered, so the overflow was sheared off the
|
|
top *and* the bottom simultaneously. That took the "Diet Gate" title and the
|
|
slot header (which meal you are being asked to log) off one edge and the calorie
|
|
headline plus the whole dashboard off the other, with no scrollbar and nothing
|
|
to indicate anything was missing. Inside a hard lock that cannot be left
|
|
without logging a meal.
|
|
|
|
Two independent causes, both needed fixing: every font size was passed to Tk as
|
|
*points* rather than pixels (~30% oversized), and there was no scroll container
|
|
anywhere in the package.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import pytest
|
|
|
|
from diet_guard.tests._gate_layout_probe import measure
|
|
|
|
# The screen the gate must fit: the primary machine's 1366x768 panel. It is
|
|
# landscape and SHORT -- phone-portrait test sizes exercise neither
|
|
# constraint, which is why the overflow survived a test suite that already
|
|
# pinned surface sizes. Panels below 768px get gatelock's best-effort
|
|
# compaction (``gatelock._density``) but are not gated: no machine runs one,
|
|
# and the gate cannot be trimmed to 600px without dropping content.
|
|
_REQUIRED_SIZES = [(1366, 768)]
|
|
|
|
|
|
@pytest.mark.parametrize(("screen_w", "screen_h"), _REQUIRED_SIZES)
|
|
@pytest.mark.parametrize("populated", [False, True])
|
|
def test_all_gate_content_is_keyboard_reachable(
|
|
screen_w: int, screen_h: int, *, populated: bool
|
|
) -> None:
|
|
"""No part of the gate is unreachable, empty or with a full day logged.
|
|
|
|
``populated`` matters: the dashboard grows one line per logged meal, so a
|
|
layout that fits on a fresh boot can still overflow by dinner. The empty
|
|
case is not the worst case.
|
|
"""
|
|
result = measure(screen_w, screen_h, populated=populated)
|
|
assert result.reachable, result.describe()
|
|
|
|
|
|
@pytest.mark.parametrize(("screen_w", "screen_h"), _REQUIRED_SIZES)
|
|
def test_overflow_is_always_scrollable(screen_w: int, screen_h: int) -> None:
|
|
"""Whenever content exceeds the viewport, the keyboard can scroll it.
|
|
|
|
Guards the specific regression: content may legitimately overflow, but it
|
|
may never overflow *without* a keyboard route to the rest of it.
|
|
"""
|
|
result = measure(screen_w, screen_h, populated=True)
|
|
if result.needs_scroll:
|
|
assert result.scrolls_by_keyboard, result.describe()
|
|
assert result.last_widget_visible_after_focus, result.describe()
|
|
|
|
|
|
@pytest.mark.parametrize("populated", [False, True])
|
|
def test_gate_fits_the_primary_screen(*, populated: bool) -> None:
|
|
"""The gate fits 1366x768 without scrolling, empty or with a day logged.
|
|
|
|
Scrolling is a correctness backstop, not the intended experience -- and
|
|
since 2026-08-03 it is also *only* reachable by the user's own keypress or
|
|
click, because a viewport that followed the app's own ``focus_set()``
|
|
scrolled the screen under a user who had touched nothing. A gate that
|
|
needs scrolling therefore needs the user to discover that it does.
|
|
|
|
``populated`` is the worst case: the dashboard grows one line per logged
|
|
meal, so a layout that fits on a fresh boot can still overflow by dinner.
|
|
"""
|
|
result = measure(1366, 768, populated=populated)
|
|
assert not result.needs_scroll, result.describe()
|