Compare commits

...
Author SHA1 Message Date
jheapsandClaude Opus 5 4c1ac222a8 Give Enter and Space back to the control that has focus
Found while checking the keyboard could still reach the dealer button: it could
reach it and then couldn't press it.

Enter and Space were bound to "advance" for the whole document, with an exception
only for text fields. Those two keys are how a keyboard presses the control it is
sitting on, so every button on the page lost them -- and not to nothing. Measured:

  Tab to Resolve Pot, press Enter   ->  the table dealt the next hand
  Tab to Resolve Pot, press Space   ->  Resolve Pot, correctly
  Tab to a dealer button, Enter     ->  nothing at all

The inconsistency is the shortcut handler's return value. It calls preventDefault
only when it handled the key, so when "advance" applied it ran instead of the button,
and when it didn't apply the browser's own activation got through. A control that
does something else every other press is worse than one that does nothing.

Enter and Space now go to a focused button, [role=button], link or summary, and the
shortcut only claims them when nothing pressable has focus. The letters stay live
with a button focused -- a button does nothing with "f", and a dealer whose hand is
on the keys shouldn't lose them to a stray focus ring.

Space still advances the street: dealt, f, f, c, c, Space -> flop. The action bar
rebuilds itself after every press, so focus falls back to the body and the dealer's
Space is never quietly aimed at the Fold they just clicked.

11 tests (163 total), and they fail on the old handler -- 4 of them, exactly the
Enter/Space-on-a-control cases.

Co-Authored-By: Claude Opus 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_015bnp7xKUPFFSpwkYNFaPgb
2026-09-01 12:16:15 -06:00
jheapsandClaude Opus 5 784315f2ea Put one dealer button on the table
A poker table has one dealer button. This one drew nine.

Every seat carried a puck: the real one in gold, and eight ghosts at 0.32 opacity
that existed to be pressed. Measured on a nine-seat table -- eight pale discs, all
eight showing the letter D, one of them the button. Finding it meant comparing nine
identical shapes for the one that's gold, when it should be the only thing of its
kind on the felt.

The ghosts also sat on their neighbours. Every remaining "something important is
covered" offender on a crowded table was a live seat covered by the seat-next-door's
ghost puck: 844x390, 915x412, 360x640 and 640x360 each reported one, and they were
all the same thing.

And a screen reader heard nine buttons named "D".

So a seat without the button doesn't draw one. Hovering a seat fades its slot in --
a preview of where the button would go, one click as before -- and on touch, where
there is no hover, "Give the button" has been in the seat's long-press menu all
along. Keyboard reaches it by focus, where the slot appears too.

pointer-events matters as much as the opacity: an invisible puck that still took
taps would be worse than a visible one.

  visible pucks    9 -> 1 at every size
  layout offenders 915x412 1->0, 360x640 1->0, 844x390 3->2, 640x360 3->2
  clean viewports  4 of 9 -> 6 of 9
  labels           nine "D" -> "Ann has the dealer button" and
                              "Give the dealer button to Ben"

The audit now checks it, because this is a rule about the table rather than about
CSS: one puck drawn, and no two announcing themselves the same way. Verified against
the old behaviour put back -- it reports 9 and 9.

Co-Authored-By: Claude Opus 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_015bnp7xKUPFFSpwkYNFaPgb
2026-09-01 12:16:00 -06:00
6 changed files with 197 additions and 7 deletions
@@ -163,7 +163,7 @@
<template id="seatTemplate">
<div class="seat" draggable="true">
<div class="seatChairNumber"></div>
<button class="dealerChip" title="Make this seat the dealer button">D</button>
<button class="dealerChip" type="button" title="Make this seat the dealer button">D</button>
<span class="blindBadge sbBadge">SB</span>
<span class="blindBadge bbBadge">BB</span>
<div class="seatPuckName"></div>
+28 -5
View File
@@ -591,7 +591,6 @@ button,
font-size: 0.68rem;
cursor: pointer;
z-index: 2;
opacity: 0.32;
transition: opacity 0.15s ease;
/* A dealer puck should look like a puck — small, sitting on the rim of its seat.
But tapping it hands the button to that seat and re-posts the blinds, and 22px
@@ -622,16 +621,40 @@ button,
left: -18px;
}
.dealerChip:focus-visible {
opacity: 1;
/* A poker table has one dealer button. This one drew nine.
Every seat carried a puck — the button's, and eight ghosts at 0.32 opacity that
were there to be pressed. Measured on a nine-seat table: eight pale discs, all
eight showing the letter D, one of them real. Finding the button meant comparing
nine identical shapes for the one that's gold, when it should be the only thing
of its kind on the felt. The ghosts also sat on top of their neighbours: every
remaining "something important is covered" offender on a crowded table was a
seat covered by the seat-next-door's ghost puck.
So a seat that doesn't have the button doesn't draw one. The affordance is still
there — hovering a seat fades its slot in, which is a preview of where the button
would go — and on touch, where there is no hover, "Give the button" has always
been in the seat's long-press menu. Keyboard reaches it by focus.
pointer-events matters as much as the opacity: an invisible puck that still takes
taps would be worse than a visible one. */
.dealerChip:not(.active) {
opacity: 0;
pointer-events: none;
}
@media (hover: hover) {
.seat:hover .dealerChip {
opacity: 1;
.seat:hover .dealerChip:not(.active) {
opacity: 0.5;
pointer-events: auto;
}
}
.dealerChip:not(.active):focus-visible {
opacity: 1;
pointer-events: auto;
}
.dealerChip.active {
background: radial-gradient(circle at 35% 30%, #ffe9a8, #f5d547 55%, #cba428 100%);
border-color: #a9782f;
@@ -174,8 +174,18 @@ const Render = {
node.querySelector(".seatChairNumber").textContent = index + 1;
const dealerChip = node.querySelector(".dealerChip");
dealerChip.classList.toggle("active", index === dealerIndex);
const hasButton = index === dealerIndex;
dealerChip.classList.toggle("active", hasButton);
dealerChip.classList.toggle("hidden", !!this.state.dedicatedDealer);
// Every one of these announced itself as "D", so a screen reader heard a row
// of nine identical buttons and couldn't tell which seat held the real one.
// The label now says whose it is, or what pressing it would do.
const who = (player.name || "").trim() || `seat ${index + 1}`;
const chipLabel = hasButton
? `${who} has the dealer button`
: `Give the dealer button to ${who}`;
dealerChip.setAttribute("aria-label", chipLabel);
dealerChip.title = chipLabel;
// Tapping the puck is the same act as "Give the button" in the seat menu,
// so it goes through the same hook — otherwise this one path would move
// the button without moving the blinds that follow it.
@@ -40,6 +40,22 @@ const Shortcuts = {
return [...document.querySelectorAll('[id$="Overlay"]')].some(el => !el.hidden);
},
// Enter and Space are how a keyboard presses the control it is already sitting on.
// Claiming them for "advance" didn't just block that press — it ran a different
// action instead. Measured: Tab to Resolve Pot, press Enter, and the table dealt
// the next hand. Space on the same button worked, because "advance" happened to
// decline that time and the browser's own activation got through. A control that
// does something else every other press is worse than one that does nothing.
//
// Only these two keys are handed back. The letter shortcuts stay live with a
// button focused, because a button does nothing with "f" and the dealer's hand
// shouldn't have to leave them.
focusedControlOwnsKey(key) {
if (key !== "enter" && key !== " ") return false;
const el = document.activeElement;
return !!(el && el.matches && el.matches('button, [role="button"], a[href], summary'));
},
onKey(e) {
if (e.metaKey || e.ctrlKey || e.altKey) return;
@@ -47,6 +63,8 @@ const Shortcuts = {
const action = this.KEYS[key];
if (!action) return;
if (this.focusedControlOwnsKey(key)) return;
// Escape is the exception: it should still close the raise pad even if focus
// happens to be sitting in its amount field.
if (this.isTyping() && !(action === "cancel")) return;
+110
View File
@@ -0,0 +1,110 @@
// The keyboard shortcuts, and the one rule that keeps them from stealing a press.
//
// Shortcuts talks to `document` directly, so these tests stand a small fake one up
// rather than a whole DOM: the only things it asks about are what has focus and
// whether an overlay is open.
const test = require("node:test");
const assert = require("node:assert");
const path = require("path");
const Shortcuts = require(path.join(
__dirname, "..", "..", "JoshHeaps.Net", "wwwroot", "js", "casino", "holdem", "shortcuts.js"));
// A stand-in for a focused element. `matches` answers the same selector list the
// real one would, which is the whole of what focusedControlOwnsKey looks at.
function element(tagName, { selectors = [], isContentEditable = false } = {}) {
return {
tagName,
isContentEditable,
matches: selector => selectors.some(s => selector.includes(s)),
};
}
const BUTTON = element("BUTTON", { selectors: ["button"] });
const SUMMARY = element("SUMMARY", { selectors: ["summary"] });
const LINK = element("A", { selectors: ["a[href]"] });
const DIV = element("DIV");
const TEXT_FIELD = element("INPUT");
// Runs one keypress against a fresh Shortcuts and reports which action fired.
function press(key, { focused = DIV, overlays = [] } = {}) {
const fired = [];
let defaultPrevented = false;
globalThis.document = {
activeElement: focused,
addEventListener() {},
querySelectorAll: () => overlays,
};
const handlers = {};
for (const action of Object.values(Shortcuts.KEYS)) {
handlers[action] = () => { fired.push(action); return true; };
}
Shortcuts.handlers = handlers;
Shortcuts.onKey({ key, preventDefault: () => { defaultPrevented = true; } });
return { fired: fired[0] || null, defaultPrevented };
}
test.after(() => { delete globalThis.document; });
test("Enter on a focused button is left to the button", () => {
assert.equal(press("Enter", { focused: BUTTON }).fired, null);
});
test("Space on a focused button is left to the button", () => {
assert.equal(press(" ", { focused: BUTTON }).fired, null);
});
test("a key left to the control is not swallowed either", () => {
// preventDefault would stop the browser activating the button, which is the
// half of the bug that turned "wrong action" into "no action".
assert.equal(press("Enter", { focused: BUTTON }).defaultPrevented, false);
});
test("summary and links own Enter and Space too", () => {
assert.equal(press("Enter", { focused: SUMMARY }).fired, null);
assert.equal(press(" ", { focused: LINK }).fired, null);
});
test("Space still advances when nothing pressable has focus", () => {
assert.equal(press(" ", { focused: DIV }).fired, "advance");
});
test("Enter still advances when nothing pressable has focus", () => {
assert.equal(press("Enter", { focused: DIV }).fired, "advance");
});
test("the letter shortcuts stay live with a button focused", () => {
// A button does nothing with "f", so there is no press to hand back — and a
// dealer whose hand is on the keys shouldn't lose them to a stray focus ring.
assert.equal(press("f", { focused: BUTTON }).fired, "fold");
assert.equal(press("c", { focused: BUTTON }).fired, "checkOrCall");
assert.equal(press("r", { focused: BUTTON }).fired, "raise");
assert.equal(press("a", { focused: BUTTON }).fired, "allIn");
assert.equal(press("u", { focused: BUTTON }).fired, "undo");
});
test("typing still silences everything but Escape", () => {
assert.equal(press("f", { focused: TEXT_FIELD }).fired, null);
assert.equal(press("Escape", { focused: TEXT_FIELD }).fired, "cancel");
});
test("an open overlay still owns the keyboard", () => {
assert.equal(press("f", { focused: DIV, overlays: [{ hidden: false }] }).fired, null);
assert.equal(press("f", { focused: DIV, overlays: [{ hidden: true }] }).fired, "fold");
});
test("a focused control wins before the overlay check, not after", () => {
// Both would decline here; the point is that neither path fires an action.
assert.equal(press("Enter", { focused: BUTTON, overlays: [{ hidden: false }] }).fired, null);
});
test("modified keys are never shortcuts", () => {
globalThis.document = { activeElement: DIV, addEventListener() {}, querySelectorAll: () => [] };
const fired = [];
Shortcuts.handlers = { advance: () => { fired.push("advance"); return true; } };
Shortcuts.onKey({ key: " ", ctrlKey: true, preventDefault() {} });
assert.deepEqual(fired, []);
});
+29
View File
@@ -514,6 +514,34 @@ const LayoutAudit = {
return { name: "a seat's centre belongs to that seat", offenders };
},
// A poker table has one dealer button. Nine pucks all showing "D", one of them
// gold, turns the most-read marker on the felt into a spot-the-difference — and
// a screen reader heard nine buttons named "D" with nothing to tell them apart.
// Both halves are checked here: what's drawn, and what it announces itself as.
oneDealerButton() {
const chips = [...document.querySelectorAll(".dealerChip")];
const shown = chips.filter(chip => {
const style = getComputedStyle(chip);
return style.display !== "none" && style.visibility !== "hidden" &&
parseFloat(style.opacity) > 0.05;
});
const offenders = [];
if (shown.length > 1) {
offenders.push(`${shown.length} dealer buttons drawn on the felt at once`);
}
const labels = chips.map(chip => chip.getAttribute("aria-label") || chip.textContent.trim());
const duplicated = labels.filter((label, i) => labels.indexOf(label) !== i);
if (duplicated.length) {
offenders.push(
`${duplicated.length + 1} dealer buttons announce themselves as "${duplicated[0]}"`
);
}
return { name: "the felt shows one dealer button", offenders };
},
// ---- Runner ----
runAll(options = {}) {
@@ -527,6 +555,7 @@ const LayoutAudit = {
this.horizontalOverflow(),
this.fieldFontSize(),
this.seatOwnership(),
this.oneDealerButton(),
];
if (options.includeTouchTargets) checks.push(this.touchTargets(options.minimumTarget));