makeover_immediate/lib.rs
1//! The immediate-mode renderer for [`makeover_layout`].
2//!
3//! <!-- wiki: makeover-immediate -->
4//!
5//! Named for the mode, not the library, the way `makeover-tui` is named for
6//! the target and not for ratatui. Immediate mode is the constraint that
7//! actually separates this renderer from the other two, and egui is the
8//! backend it is written against.
9//!
10//! It is the harshest renderer the description has to survive: no
11//! `box-shadow`, no `inset`, no cascade, no retained tree to mutate, and
12//! `Visuals.widgets.*.bg_stroke` is a single stroke with no per-side control.
13//! A two-tone lit edge is not something egui can be configured into producing,
14//! so it gets painted by hand here, once, instead of in every consuming app.
15//!
16//! # What this crate does and does not own
17//!
18//! It owns the *expression*: two mitred polylines for a bevel and a `Frame`
19//! for a filled region. It owns no colours, no sizes and no substitutions:
20//! makeover derives `surface-well` and every consumer reads the real token.
21//! [`Palette`] is supplied by the caller,
22//! already resolved, and every radius, margin and stroke width arrives in
23//! [`FrameStyle`].
24//!
25//! That split is why the crate has no dependency on `makeover` itself: the app
26//! already resolves a theme, and coupling a renderer to a colour crate's
27//! version would buy nothing.
28//!
29//! # The cascade is the real difference
30//!
31//! A stylesheet can say "a pressed button inverts its bevel" once and let the
32//! cascade carry it. An immediate-mode renderer has nowhere to put that, so
33//! every call site decides. [`makeover_layout::Depth::pressed`] is what keeps
34//! the decision from being re-derived per widget.
35//!
36//! # Overlays
37//!
38//! [`Palette::cast`] is what overlaying means in immediate mode. [`frame`] hands
39//! the cast shadow to the `egui::Frame` for any depth whose fill is
40//! [`Fill::Overlay`], keyed off the fill rather than the variant.
41//!
42//! # The table
43//!
44//! [`table`] draws `makeover_layout::CellPart`. Two things it forces, both named
45//! where they land:
46//!
47//! - **`egui_extras`**, this crate's one dependency past egui. egui has no
48//! table, and `Grid` gives no per-column sizing, no sticky header and no
49//! scroll sync. A third answer here would reimplement that crate worse.
50//! - **[`Palette::action`]**, on the footing [`Palette::content`] sits on: a
51//! link in a cell needs the action intent.
52//!
53//! Narrowing works differently from the terminal's and the module header says
54//! why: a content column cannot be measured before the app's closure has drawn
55//! it, so `egui_extras` sizes it and the declared floor budgets it.
56//!
57//! Three things are host idiom rather than description, which is why they land
58//! here and not in `makeover-layout`, and all three are answered on a handle the
59//! app never sees: the `egui_extras` row and builder this crate owns. That is
60//! [`table::cell`]'s reasoning again: what the app cannot reach, the renderer
61//! owes it.
62//!
63//! - **A selected row.** [`table::Body::selected`], a predicate asked per row,
64//! because `set_selected` is a method on the row. Without it a file list has
65//! no way to show what is selected, which is most of what a file list does.
66//! - **Scrolling a row into view.** [`table::Body::scroll_to`], because
67//! `scroll_to_row` is a method on the builder. A keyboard cursor that moves
68//! off-screen and stays there is the bug this prevents.
69//! - **Dragging a divider.** [`table::TableStyle::resizable`], which is a knob
70//! because egui_extras offers two settings here and a renderer can honestly
71//! make either choice.
72//!
73//! Cells are centred on the row's centre line, always, because there is no
74//! second honest answer and egui's own default (top-aligned) is the one thing it
75//! cannot be. That is not a knob.
76//!
77//! [`table::Body`] is also what splits a table's per-frame facts from its
78//! description and from its style. A row count, a selection and a scroll request
79//! are none of them style, and none of them survive the frame.
80//!
81//! # The nodes that are not fields, tables or frames
82//!
83//! [`widget`] draws a meter, a token, a control and a figure. The four are
84//! ordinary nodes, so without them a screen walk has to draw them itself, one
85//! copy per consumer.
86//!
87//! [`Palette`] carries the three status intents together rather than one per
88//! widget, for the reason [`Palette::fill`] is an `Option`: `Tone` is five
89//! members wide, and a resolver missing one has to invent a colour, which is a
90//! substitution this crate does not make.
91//! # Forms
92//!
93//! The field vocabulary sits on top of the depth vocabulary:
94//! [`makeover_layout::Field`] rendered to egui widgets, in [`field`], and a set
95//! of them laid down a column in [`group`].
96//!
97//! `makeover-webview`'s form emitter is the precedent, followed rather than
98//! re-derived, including the parts that are bug fixes: a
99//! select handed a value none of its options carries keeps that value visible
100//! instead of silently reading as the first option, which is a save-the-wrong-
101//! thing bug goingson hit for real.
102//!
103//! What differs is forced by the mode and not chosen:
104//!
105//! - **The value arrives as a `&mut`.** [`Filling`] borrows the app's own field
106//! and the widget writes through it. There is no DOM to read back out of,
107//! which is also why the description deliberately does not carry the value.
108//! - **A text control is drawn as a well and a select is not.** The description
109//! holds that a well is for anything the user looks *into*, and a text field
110//! is its own example; a select and a checkbox are pressed rather than looked
111//! into, so they keep egui's own control painting.
112//! - **Focus is not describable, and egui owns all of it here.** **Reach**,
113//! **focus** and the **focus ring** are this renderer's three answers and
114//! egui already has all three: its own id stack decides what is reachable,
115//! its own state decides what holds the keyboard, and it paints exactly one
116//! ring. A description states none of them, and drawing a second ring on top
117//! of egui's would break the one-ring rule. The terms
118//! are defined once in `makeover_layout`'s crate header, "Reach, focus and
119//! the focus ring". [`makeover_layout::State::Disabled`] *is* drawn, because
120//! egui has no opinion about it until told.
121//! - **App-level chrome is not drawn here, and it is not this crate's to
122//! draw.** `quasi-router` names the affordances that outlive one screen: a
123//! `Chrome` of key bindings, and an `Outcome::Over` for a screen drawn over
124//! another. Both are answered by `quasi-webview` and `quasi-tui`, and neither
125//! is answerable here, because this crate depends on `makeover-layout` and
126//! not on `quasi-router` — it is the peer of `makeover-webview` and
127//! `makeover-tui`, one layer below the renderers that consume a `Screen`.
128//! What is missing is the egui crate at *that* layer, which does not exist:
129//! nothing renders a quasi `Screen` in egui at all, and chrome is one item on
130//! the list such a crate would owe. Said here because this is where a reader
131//! looks for it, and because the silent version reads as "egui does not need
132//! a palette" rather than "nobody has built the renderer yet".
133
134//! # An interval is a sixth control shape
135//!
136//! [`FieldKind::Interval`] is drawn as `Control::Spanned`: two drag boxes on one
137//! row with the word `to` between them.
138//!
139//! - **Dragged rather than typed**, because that is what these controls already
140//! were. audiofiles' six filter axes are `DragValue` pairs sharing an extent,
141//! a speed and a suffix, and describing them into two text boxes would be a
142//! port that cost the app a control.
143//! - **One row, not two wells stacked.** Two wells are two questions on screen
144//! whatever the description says, and the arrangement is the whole content of
145//! the kind.
146//! - **An empty end reads as the bound it stands for.** An unset minimum sits
147//! on the low edge and stores no filter, which is what the shipped control
148//! did; egui's `DragValue` has no empty state, and a text box in its place
149//! would cost the app a control. With no extent to fall back on it reads
150//! zero -- the one number this renderer invents, invented where the
151//! description declined to say anything.
152//! - **The word rather than a dash**, which on a signed axis is a minus sign.
153//! audiofiles filters loudness in dBFS.
154//!
155//! `Axis` holds the four facts both boxes share, because they are one axis:
156//! reading `min`, `max`, `step` and `unit` once is what stops the two ends
157//! drifting apart.
158//!
159//! # A number draws its unit
160//!
161//! [`Field::unit`], and this host is the one with somewhere better than the
162//! label to put it: egui's `Slider` draws a suffix beside its readout.
163//!
164//! So a slider takes it as a suffix, inside the control. A typed number has no
165//! readout of its own and takes it as a muted label after the box. Every other
166//! kind ignores it, and the description says which those are --
167//! `FieldKind::measurable`, rather than a `matches!` kept here.
168//!
169//! # The slider's track is a curve
170//!
171//! `makeover-layout` says what a slider is: a fraction and a function
172//! taking numbers to numbers, with `min` and `max` being `f(0)` and `f(1)`
173//! rather than the control's extent. This host has the easiest job of the
174//! three, because egui already has the control -- `Slider::logarithmic` is a
175//! constant-ratio track, so the mapping is a builder call rather than an
176//! arithmetic of its own.
177//!
178//! Two things worth knowing. The granularity rides on the curve, so a range
179//! reads `Field::curve.step()` and every other kind reads `Field::step`; the
180//! step is in the value's own units under either curve, so the display
181//! precision derives from it directly. And the fallback for a ratio curve
182//! across zero is asked of `Curve::is_ratio` rather than matched on the
183//! variant, so this renderer and a terminal cannot disagree about when a
184//! logarithmic request is honoured.
185//!
186//! # The slider, the unanswered chooser, and the option that is not offered
187//! yet
188//!
189//! Three things a description can say here.
190//!
191//! - **[`FieldKind::Range`] is a fifth control shape**, `Control::Slid`, drawn
192//! with egui's `Slider`. A range missing an end falls back to a well rather
193//! than to invented bounds, which is what
194//! `makeover_layout::Field::bounded` is for.
195//! - **`Field::placeholder` reads on a chooser**, so a select with nothing
196//! chosen does not show an empty box.
197//! - **`Choice::unavailable` is drawn rather than dropped.** The option stays
198//! in the list, inert, with its precondition beside it instead of behind a
199//! hover — a greyed row with no reason reads as a dead end, which is the
200//! whole finding.
201//! - **`Choice::detail` is drawn under the option in a
202//! radio group and inside the row in a combo.** A closed chooser hides its
203//! list, so everything an option carries has to travel with its row; a group
204//! has a line to spare and putting a sentence beside the control instead
205//! would push every option's radio out of line with its neighbours.
206//!
207//! The value still arrives as a `&mut String` and a slider is a number, so the
208//! parse and the write-back are this renderer's, and the write happens only on
209//! a real drag: a value the app put there that this host cannot read survives
210//! being looked at.
211
212#![forbid(unsafe_code)]
213
214use egui::{
215 Color32, ComboBox, CornerRadius, DragValue, Margin, Painter, Rect, Response, RichText, Shape,
216 Slider, Stroke, TextEdit, Ui,
217};
218use makeover_layout::{
219 Bevel, Choice, Depth, Edge, Field, FieldKind, Fill, State, ThemeVariant, Tone,
220};
221use std::ops::RangeInclusive;
222
223/// Columns, narrowing, cell parts and the sort caret, over `egui_extras`.
224pub mod table;
225pub mod widget;
226
227/// The resolved colours this renderer needs, as flat values.
228///
229/// Built by the app from whatever it already uses to resolve a theme, then
230/// held and reused. Deliberately not a trait and not string-keyed: a bevel is
231/// painted per widget per frame, and a map lookup per edge is a cost with
232/// nothing to show for it.
233#[derive(Debug, Clone, Copy, PartialEq, Eq)]
234pub struct Palette {
235 /// `surface-page`.
236 pub page: Color32,
237 /// `surface-raised`.
238 pub raised: Color32,
239 /// `surface-overlay`.
240 pub overlay: Color32,
241 /// `surface-well`.
242 ///
243 /// Required, not optional. makeover derives it for every theme, so a
244 /// resolved palette without a well is not a thing that exists here.
245 /// `makeover-tui` keeps its own `Option` for a different reason, since a
246 /// terminal can have the colour and still be unable to show it.
247 pub well: Color32,
248 /// `surface-sunken`.
249 ///
250 /// A surface set back from the one it sits on, by colour and nothing else.
251 /// Not a well: a well is a hole with an edge, and this has no edge. An
252 /// immediate-mode renderer paints an arbitrary rect, so unlike
253 /// `makeover-tui` it has no excuse for declining this one.
254 ///
255 /// Required rather than optional, on the same footing as `well`: all 31
256 /// themes makeover embeds author it.
257 pub sunken: Color32,
258 /// `bevel-light`.
259 pub bevel_light: Color32,
260 /// `bevel-dark`.
261 pub bevel_dark: Color32,
262 /// `elevation`.
263 ///
264 /// What a surface that floats OVER the page is cast onto it with. The one
265 /// intent here that is about a surface's relationship to the page rather
266 /// than about the surface, which is why it is a translucent near-black on
267 /// every theme rather than something read off the palette's own ramp.
268 ///
269 /// **Only for a surface that overlays.** A menu, a tooltip, a modal. A
270 /// surface *in* the layout takes a bevel, and reaching for this on a panel
271 /// or a card is how a pre-Platinum look survives a conversion under a new
272 /// name.
273 ///
274 /// egui has a real answer for this where a terminal does not: see
275 /// [`Palette::cast`], which is the shadow to hand an
276 /// [`egui::Frame`](egui::Frame).
277 pub elevation: Color32,
278 /// `content`.
279 ///
280 /// Ordinary text.
281 pub content: Color32,
282 /// `content-secondary`.
283 ///
284 /// Inactive but usable: it still answers a press. The middle tone of the
285 /// three (wiki `three-tone-convention`), and the one an unchosen option in
286 /// a choice field takes.
287 ///
288 /// Not [`content_muted`](Self::content_muted), which carries a claim:
289 /// `State::Disabled` resolves to it, so a live control wearing it tells the
290 /// user it will not answer. `makeover-tui` draws the same widget the same
291 /// way from `makeover-tui@230bf63`.
292 ///
293 /// A step of `content` toward the page, derived at load by `makeover`
294 /// rather than authored, so it is read off the resolved theme here like
295 /// any other token and never re-derived.
296 pub content_secondary: Color32,
297 /// `content-muted`.
298 ///
299 /// A field's hint, and what
300 /// [`makeover_layout::State::Disabled`](makeover_layout::State::Disabled)
301 /// resolves to. Both readings come from the description rather than from
302 /// here: `State::Disabled` names this intent by token.
303 pub content_muted: Color32,
304 /// `action-primary`.
305 ///
306 /// What a control is drawn in.
307 ///
308 /// This is the intent [`CellPart`](makeover_layout::CellPart) exists to
309 /// separate. A cell holding a control that takes the cell's text colour is
310 /// the drift `CellPart` names.
311 pub action: Color32,
312 /// `danger`.
313 ///
314 /// A field's error message, a destructive control, a bar that has run over.
315 pub danger: Color32,
316 /// `success`.
317 ///
318 /// The three status intents arrive together and not one at a time:
319 /// [`Tone`] is five members wide and a resolver missing one has to invent
320 /// a colour for it, which is the substitution [`Palette::fill`] refuses.
321 pub success: Color32,
322 /// `warning`.
323 pub warning: Color32,
324 /// `info`.
325 pub info: Color32,
326 /// `border`, the authored line colour.
327 ///
328 /// A neutral badge's edge: a badge on the raised ground is raised too, so
329 /// the edge is what draws it.
330 pub border: Color32,
331 /// `info-surface`: raised with 12 percent of `info`. A badge's fill.
332 ///
333 /// The four tone surfaces arrive together for the status intents' reason.
334 pub info_surface: Color32,
335 /// `success-surface`.
336 pub success_surface: Color32,
337 /// `warning-surface`.
338 pub warning_surface: Color32,
339 /// `danger-surface`.
340 pub danger_surface: Color32,
341 /// `row-stripe`: every second row of a table, ink mixed into raised at 5
342 /// percent (wiki `table-model`).
343 pub row_stripe: Color32,
344 /// `row-hover`: the row under the pointer, at 9 percent.
345 pub row_hover: Color32,
346 /// `row-rule`: the hairline between rows and around a table, at 26
347 /// percent.
348 pub row_rule: Color32,
349 /// `row-selected`: a selected row, raised with 18 percent of the accent.
350 ///
351 /// A fill and never a foreground, so a row that is red for a failure stays
352 /// red while it is selected.
353 pub row_selected: Color32,
354}
355
356impl Palette {
357 /// Resolve a surface intent, or `None` for one this renderer does not know.
358 ///
359 /// A plain lookup, and no substitution.
360 ///
361 /// `Option`, because [`Fill`] is `#[non_exhaustive]` and a total function
362 /// over an open enum can only stay total by inventing a colour for a member
363 /// it has never heard of. Every member the description has today is
364 /// answered with `Some`.
365 #[must_use]
366 pub const fn fill(&self, fill: Fill) -> Option<Color32> {
367 match fill {
368 Fill::Page => Some(self.page),
369 Fill::Raised => Some(self.raised),
370 Fill::Overlay => Some(self.overlay),
371 Fill::Well => Some(self.well),
372 Fill::Sunken => Some(self.sunken),
373 _ => None,
374 }
375 }
376
377 /// The colour a [`Tone`] reads as.
378 ///
379 /// Total, unlike [`fill`](Self::fill), and the difference is not an
380 /// inconsistency. `Fill` is `#[non_exhaustive]` and `Tone` is not: the
381 /// description layer settled tone at five members and grows surfaces, so a
382 /// total function here cannot be made to invent a colour by an upstream
383 /// release the way a total `fill` could.
384 ///
385 /// [`Tone::Neutral`] is [`content`](Self::content) rather than a colour of
386 /// its own, which is what "an ordinary fact" means: a neutral badge is text
387 /// in a box, not a fifth status.
388 #[must_use]
389 pub const fn tone(&self, tone: Tone) -> Color32 {
390 match tone {
391 Tone::Neutral => self.content,
392 Tone::Info => self.info,
393 Tone::Success => self.success,
394 Tone::Warning => self.warning,
395 Tone::Danger => self.danger,
396 }
397 }
398
399 /// What a badge of this tone is filled with: its tone surface, or the
400 /// raised surface for [`Tone::Neutral`], as `makeover-webview` fills one.
401 #[must_use]
402 pub const fn tone_surface(&self, tone: Tone) -> Color32 {
403 match tone {
404 Tone::Neutral => self.raised,
405 Tone::Info => self.info_surface,
406 Tone::Success => self.success_surface,
407 Tone::Warning => self.warning_surface,
408 Tone::Danger => self.danger_surface,
409 }
410 }
411
412 /// What a badge of this tone is edged with: its tone, or
413 /// [`border`](Self::border) for [`Tone::Neutral`], whose tone is the ink.
414 #[must_use]
415 pub const fn tone_edge(&self, tone: Tone) -> Color32 {
416 match tone {
417 Tone::Neutral => self.border,
418 other => self.tone(other),
419 }
420 }
421
422 /// The cast shadow for a surface that overlays the page.
423 ///
424 /// What "overlaying" means in immediate mode, answered rather than skipped.
425 /// egui already paints shadows for its menus and windows through
426 /// [`egui::Frame::shadow`], so the honest port is to hand that machinery the
427 /// theme's tone instead of egui's own default, not to invent a painter here
428 /// the way [`paint_bevel`] had to.
429 ///
430 /// The geometry matches what `makeover-webview` composes, in points rather
431 /// than pixels: a small downward offset and a wide soft blur. A Platinum-era
432 /// menu sits just off the page rather than hovering above it.
433 ///
434 /// ```no_run
435 /// # let palette: makeover_immediate::Palette = unimplemented!();
436 /// # let ui: &mut egui::Ui = unimplemented!();
437 /// egui::Frame::popup(ui.style())
438 /// .shadow(palette.cast())
439 /// .show(ui, |ui| { ui.label("over the page"); });
440 /// ```
441 #[must_use]
442 pub const fn cast(&self) -> egui::Shadow {
443 egui::Shadow {
444 offset: [0, 2],
445 blur: 24,
446 spread: 0,
447 color: self.elevation,
448 }
449 }
450
451 /// Resolve a bevel edge intent.
452 #[must_use]
453 pub const fn edge(&self, edge: Edge) -> Color32 {
454 match edge {
455 Edge::Light => self.bevel_light,
456 Edge::Dark => self.bevel_dark,
457 }
458 }
459}
460
461/// The geometry a framed region is drawn with.
462///
463/// Every field is a value, which is why they all arrive from the caller:
464/// radius and border width belong to `makeover-geometry`, and margins come
465/// from its relational gaps.
466#[derive(Debug, Clone, Copy, PartialEq)]
467pub struct FrameStyle {
468 /// Corner radius. Square under the Platinum default.
469 pub radius: CornerRadius,
470 /// Inner margin between the frame and its contents.
471 pub margin: Margin,
472 /// Bevel stroke width, in points.
473 pub stroke: f32,
474}
475
476impl Default for FrameStyle {
477 /// A one-point square frame with no inner margin.
478 fn default() -> Self {
479 Self {
480 radius: CornerRadius::ZERO,
481 margin: Margin::ZERO,
482 stroke: 1.0,
483 }
484 }
485}
486
487/// Paint a two-tone edge just inside `rect`.
488///
489/// Fill first, bevel after: this adds two polylines and nothing else, so it
490/// composes over whatever is already there. That is what lets it go over an
491/// [`egui::TextEdit`] after `ui.add`, where the widget's own fill has landed.
492///
493/// Two three-point polylines meeting at opposite corners, rather than four
494/// segments, so egui mitres the corner joins instead of leaving a notch.
495///
496/// The dark polyline is drawn second, so the two corners where the runs meet
497/// take its tone. That is the right answer here rather than a concession.
498/// [`makeover_layout::Bevel`] holds those corners to belong to both edges, and
499/// a renderer with room to divide one should; at the default one-point stroke
500/// the corner is a one-point square, so the division is sub-pixel and
501/// antialiasing resolves it to the same blend the mitre already gives. Splitting
502/// it would add a seam and no information. `makeover-tui` does split, because a
503/// terminal cell is large enough that not splitting costs a visible cell of edge
504/// weight — the same rule, at a resolution where it has something to say.
505pub fn paint_bevel(painter: &Painter, rect: Rect, bevel: Bevel, palette: &Palette, stroke: f32) {
506 let (top_left, bottom_right) = bevel.edges();
507
508 // Inset by half a stroke so the line lands inside `rect` rather than
509 // straddling its edge, which on a fractional-scale display is the
510 // difference between one crisp pixel and two dim ones.
511 let r = rect.shrink(stroke / 2.0);
512
513 painter.add(Shape::line(
514 vec![r.left_bottom(), r.left_top(), r.right_top()],
515 Stroke::new(stroke, palette.edge(top_left)),
516 ));
517 painter.add(Shape::line(
518 vec![r.right_top(), r.right_bottom(), r.left_bottom()],
519 Stroke::new(stroke, palette.edge(bottom_right)),
520 ));
521}
522
523/// Draw a region at a given [`Depth`]: its fill and its edge, together.
524///
525/// [`Depth::Flat`] gets neither, and inherits whatever it sits on. That is the
526/// difference between level-with and painted-the-same-colour, and it is the
527/// reason `Depth::fill` returns an [`Option`] rather than defaulting to the
528/// page.
529pub fn frame<R>(
530 ui: &mut Ui,
531 depth: Depth,
532 palette: &Palette,
533 style: FrameStyle,
534 add_contents: impl FnOnce(&mut Ui) -> R,
535) -> R {
536 let mut f = egui::Frame::new()
537 .corner_radius(style.radius)
538 .inner_margin(style.margin);
539 // Two ways there is no fill to paint, and they collapse to the same
540 // outcome: the depth names none (Depth::Flat), or it names one this
541 // renderer cannot resolve. Either way the frame goes unfilled and the
542 // bevel below carries the depth on its own, which is the rule this
543 // module already documents for Flat.
544 if let Some(fill) = depth.fill().and_then(|f| palette.fill(f)) {
545 f = f.fill(fill);
546 }
547 // A surface that overlays the page is cast onto it. [`Palette::cast`] has
548 // answered what that means here since 0.10.0 and nothing could reach it: a
549 // description had no way to say Overlay until makeover-layout 0.14.0, so
550 // the answer sat beside the question. Keyed off the fill rather than the
551 // variant, so it stays right for whatever else the description calls an
552 // overlay later.
553 if depth.fill() == Some(Fill::Overlay) {
554 f = f.shadow(palette.cast());
555 }
556 let framed = f.show(ui, add_contents);
557 if let Some(bevel) = depth.bevel() {
558 paint_bevel(
559 ui.painter(),
560 framed.response.rect,
561 bevel,
562 palette,
563 style.stroke,
564 );
565 }
566 framed.inner
567}
568
569/// The geometry a field group is drawn with.
570///
571/// Values again, for the reason [`FrameStyle`] is: every number here belongs to
572/// `makeover-geometry` and arrives already resolved.
573#[derive(Debug, Clone, Copy, PartialEq)]
574pub struct FieldStyle {
575 /// The well a text control sits in.
576 pub frame: FrameStyle,
577 /// Between a field's own parts: its label, its control, its hint and its
578 /// error.
579 pub gap: f32,
580 /// Between one field and the next.
581 pub group_gap: f32,
582 /// What marks a required field, appended to its label.
583 ///
584 /// A knob rather than a constant, because it is the one piece of *copy* in
585 /// this crate and copy is not a renderer's call. A webview does not need it
586 /// at all — it emits the `required` attribute and the browser answers — so
587 /// this renderer is the first place where a compulsory field either shows
588 /// that it is or silently does not.
589 pub required_marker: &'static str,
590}
591
592impl Default for FieldStyle {
593 /// The default frame, no gaps, and an asterisk.
594 fn default() -> Self {
595 Self {
596 frame: FrameStyle::default(),
597 gap: 0.0,
598 group_gap: 0.0,
599 required_marker: "*",
600 }
601 }
602}
603
604/// What the field currently holds, borrowed from wherever the app keeps it.
605///
606/// The immediate-mode counterpart of `makeover_webview::form::Value`, and the
607/// place the two renderers are forced apart: there the value is read back out
608/// of the DOM after the fact, and here the widget writes through this borrow as
609/// it is edited. Same reason the description carries neither.
610///
611/// An enum rather than a bag of options, on the reasoning
612/// `makeover_webview::form::Value` records: a checkbox holding a string is
613/// unsayable here, where a struct would let it be said and then have to cope.
614#[derive(Debug, Default)]
615pub enum Filling<'a> {
616 /// Nothing to edit. The control is drawn and does not answer.
617 #[default]
618 Absent,
619 /// The buffer behind anything that takes typed text, a select included:
620 /// what a select holds is the `value` of one of its [`Choice`]s.
621 ///
622 /// [`Choice`]: makeover_layout::Choice
623 Text(&'a mut String),
624 /// A checkbox, on or off.
625 On(&'a mut bool),
626 /// The two buffers behind a [`FieldKind::Interval`], lower first.
627 ///
628 /// Two buffers rather than one string with a separator, which is
629 /// [`makeover_layout::Field::upper_name`]'s reason one level down: an
630 /// interval is submitted under two names, so it is edited as two values,
631 /// and a delimiter this crate owned could appear inside either of them.
632 ///
633 /// Either end may be empty while the other stands. An open end is an
634 /// answer -- "over 120 BPM" -- rather than a half-filled box.
635 Between {
636 /// The lower end's buffer.
637 lower: &'a mut String,
638 /// The upper end's buffer.
639 upper: &'a mut String,
640 },
641}
642
643/// The label, marked if the field is compulsory.
644fn label_text(field: &Field<'_>, style: &FieldStyle) -> String {
645 if field.required {
646 format!("{} {}", field.label, style.required_marker)
647 } else {
648 field.label.to_owned()
649 }
650}
651
652/// The four shapes a control comes in here, which is fewer than there are
653/// kinds.
654///
655/// [`FieldKind`] is `#[non_exhaustive]` and grows; this does not, because the
656/// ways egui has of asking for a value do not. Reducing the open set to this
657/// closed one in one total function is what keeps a new kind from needing a new
658/// arm at every match below.
659#[derive(Debug, Clone, Copy, PartialEq, Eq)]
660enum Control {
661 /// Typed into, so it is drawn as a well: the user looks into it.
662 Typed,
663 /// Picked from a control that shows one option at a time. Pressed rather
664 /// than looked into, so egui's own control painting stands.
665 Chosen,
666 /// Picked from options that are all on screen at once.
667 ///
668 /// Apart from [`Chosen`](Self::Chosen) because the description holds them
669 /// apart, and holding them apart is the whole content of
670 /// [`FieldKind::Radio`]: same question, and an answer the user can read
671 /// without opening anything.
672 Listed,
673 /// Held on or off.
674 Toggled,
675 /// Dragged across an extent that is on screen the whole time.
676 ///
677 /// Apart from [`Typed`](Self::Typed) for the reason
678 /// [`FieldKind::Range`] is apart from `Number`: the two ends are what the
679 /// question means, so a well with a figure in it is not a quieter version
680 /// of this control, it is a different one.
681 Slid,
682 /// Picked from a list of themes that arrives grouped and marked.
683 ///
684 /// Apart from [`Chosen`](Self::Chosen) rather than folded into it, and the
685 /// distinction is the same one [`Listed`](Self::Listed) draws: it is not a
686 /// different question, it is a different amount of structure on screen. A
687 /// theme picker's rows carry a group heading and a contrast mark, and both
688 /// come from members [`Field::options`] does not have, so a shared arm
689 /// would be a `matches!` on the kind inside the loop rather than one arm
690 /// less.
691 Themed,
692 /// Two values dragged across one axis, drawn as one question.
693 ///
694 /// Apart from [`Typed`](Self::Typed) for the reason
695 /// [`FieldKind::Interval`] is apart from `Number`: two wells one under the
696 /// other are two questions on screen, whatever the description says, and
697 /// the arrangement is the whole content of the kind.
698 Spanned,
699}
700
701/// Which shape a kind takes.
702///
703/// The wildcard falls to [`Control::Typed`] on purpose: a kind added to the
704/// description since this renderer was built degrades to a text box, which
705/// accepts any value the others would, rather than to nothing drawn at all.
706///
707/// `FieldKind::File` lands there rather than growing a shape of its own. egui's
708/// honest answer is a button that opens a native picker, which is a fifth
709/// control and a file-dialog dependency, and no consumer of this crate asks for
710/// a file field. The membership test is that every renderer *could* answer
711/// honestly, not that each one does on the day. A path in a text box is not
712/// nothing.
713///
714/// `Field::accept` and `Field::multiple` land on the same position: both are the picker's arguments, and this
715/// renderer has no picker to give them to. They are not lost — the description
716/// still carries them, and the day the native dialog arrives here it is opened
717/// with them rather than with a filter written twice.
718///
719/// `FieldKind::Date` and `FieldKind::DateTime` land there too, on the same
720/// footing and with one thing owed. A
721/// calendar is a sixth control and bare `egui` has none, so a typed value is
722/// the honest answer here; what the app gets is the format the description
723/// names, `makeover_layout::DATE_FORMAT` and `DATETIME_FORMAT`, which is why
724/// those are constants rather than a sentence. audiofiles is the only consumer
725/// of this crate and asks for neither today. A calendar popup is the upgrade
726/// whenever one does.
727///
728/// `Field::as_instant` is carried and not honoured, on
729/// the same footing. It asks for the typed wall-clock value to be submitted as
730/// the moment it names, and this renderer has no submission to convert on: it
731/// draws the control and the app reads the value back, so the conversion would
732/// belong wherever that read happens rather than here. What the app gets is the
733/// local value in `DATETIME_FORMAT`, which is what it got before the member
734/// existed. No described site on this host asks for it today.
735/// One row of a closed chooser, as the single string it has room for.
736///
737/// A combo hides its list, so everything an option carries has to travel with
738/// the row it belongs to: there is no second line to put a detail on and no
739/// space beside the row to put a reason in. That is the same constraint a
740/// `<select>`'s option has in the webview, and it takes the same answer.
741///
742/// The order is what the option *is* before why it cannot be picked, which is
743/// the order the two read in and the order the radio group draws them in.
744fn combo_row(opt: &Choice<'_>) -> String {
745 let mut text = opt.label.to_owned();
746 for extra in [opt.detail, opt.unavailable].into_iter().flatten() {
747 text.push_str(" ");
748 text.push_str(extra);
749 }
750 text
751}
752
753/// How far an option's second line is inset, in points.
754///
755/// The width of egui's radio button plus the gap after it, so the line starts under the label rather than under the control. A
756/// magnitude, which is `makeover-geometry`'s subject and not this crate's --
757/// but this one is measured off a widget egui draws and sizes, so there is
758/// nothing for a spacing scale to say about it.
759const OPTION_DETAIL_INDENT: f32 = 22.0;
760
761const fn control_shape(kind: FieldKind) -> Control {
762 match kind {
763 FieldKind::Select => Control::Chosen,
764 FieldKind::Radio => Control::Listed,
765 FieldKind::Checkbox => Control::Toggled,
766 FieldKind::Range => Control::Slid,
767 FieldKind::Interval => Control::Spanned,
768 FieldKind::Theme => Control::Themed,
769 _ => Control::Typed,
770 }
771}
772
773/// The shape the field actually gets, which is the kind's unless the field is
774/// missing what that shape needs.
775///
776/// One case, and `makeover-layout` names it: a [`FieldKind::Range`] carries its
777/// extent in [`Field::min`] and [`Field::max`], and a range missing an end has
778/// nothing to slide across. egui's `Slider` demands a `RangeInclusive`, so
779/// inventing one would be this renderer picking bounds the app never stated and
780/// the user then dragging against them.
781///
782/// It falls back to [`Control::Typed`], which is where every kind this renderer
783/// cannot draw natively already lands: a number in a well is a true report of
784/// the value and takes any answer the slider would.
785fn shape_of(field: &Field<'_>) -> Control {
786 match control_shape(field.kind) {
787 Control::Slid if !field.bounded() => Control::Typed,
788 shape => shape,
789 }
790}
791
792/// The unit to draw beside this field's value, if there is one to draw.
793///
794/// Two conditions rather than one: the field has to carry a unit and its kind
795/// has to be one that means anything by it. `FieldKind::measurable` is the
796/// description answering the second, so this renderer does not keep its own
797/// list of which kinds are quantities -- which is the drift that predicate
798/// exists to stop.
799fn unit_of<'a>(field: &Field<'a>) -> Option<&'a str> {
800 field.unit.filter(|_| field.kind.measurable())
801}
802
803/// The two ends of a range, as egui wants them.
804///
805/// `None` when either end is missing or is not a number this host can read.
806/// The description carries the bounds as text on purpose — the bound of a date
807/// is a date — so parsing them is the renderer's job and failing to is a real
808/// outcome rather than an assertion.
809fn extent(field: &Field<'_>) -> Option<RangeInclusive<f64>> {
810 let min = field.min?.parse::<f64>().ok()?;
811 let max = field.max?.parse::<f64>().ok()?;
812 Some(min..=max)
813}
814
815/// How many decimals to write a dragged value back with.
816///
817/// Read off [`Field::step`], which is the only thing that says what
818/// granularity the question has: a step of `0.01` is a two-decimal question and
819/// a step of `1` is a whole-number one. Without a step the host's own
820/// granularity stands, and egui's is continuous, so the value is written back
821/// at whatever precision it round-trips at.
822fn decimals(step: Option<&str>) -> Option<usize> {
823 let step = step?;
824 Some(match step.split_once('.') {
825 Some((_, fraction)) => fraction.trim_end_matches('0').len(),
826 None => 0,
827 })
828}
829
830/// What a select shows for the value it currently holds.
831///
832/// A value no option carries stays on screen as itself rather than reading as
833/// whichever option happens to be first. goingson saved a backup retention of
834/// 10 against a 1/3/7/14/0 list and the browser silently showed it as 1, so the
835/// next save wrote a value nobody chose; `makeover-webview` grew the fix as a
836/// stray `<option>` and this is the same fix in the shape egui allows.
837///
838/// The empty value is the one case that reads as unanswered rather than as an
839/// answer, and [`chosen_text`] is what puts the field's ghost text there.
840fn shown_label<'a>(options: &'a [Choice<'a>], value: &'a str) -> &'a str {
841 options
842 .iter()
843 .find(|opt| opt.value == value)
844 .map_or(value, |opt| opt.label)
845}
846
847/// What a select's closed control reads, and in which tone.
848///
849/// A chooser with nothing chosen showed an empty box: `shown_label` falls back
850/// to the value, and the unanswered value is the empty string. So an app with
851/// an instruction to give — audiofiles' "Select device..." — had nowhere to put
852/// it but a disabled button elsewhere on the screen, which is the affordance
853/// this vocabulary keeps moving messages *off*.
854///
855/// [`Field::placeholder`] is already the description's word for "what the field
856/// reads while it is empty" and was honoured by the typed kinds alone, so
857/// nothing new is said here; the renderer is what had not caught up. Muted
858/// because it is not an answer, the same tone the typed kinds' ghost text takes
859/// three lines up.
860///
861/// A value no option carries but that is *not* empty stays as itself, in
862/// `content`: that is the goingson retention bug and it is a wrong answer
863/// rather than an absent one.
864///
865/// Returns the words and the tone rather than a built [`RichText`], because
866/// what it decides is both of them and only one of them is readable back off a
867/// `RichText`.
868///
869/// [`Field::placeholder`]: makeover_layout::Field::placeholder
870fn chosen_text<'a>(field: &'a Field<'a>, value: &'a str, palette: &Palette) -> (&'a str, Color32) {
871 match field.placeholder {
872 Some(ghost) if value.is_empty() => (ghost, palette.content_muted),
873 _ => (shown_label(field.options, value), palette.content),
874 }
875}
876
877/// What one option in a choice field is drawn in.
878///
879/// The chosen one is the emphasised thing and takes `content`; the rest take
880/// [`content_secondary`](Palette::content_secondary), because an option that is
881/// not chosen is still an option and pressing it chooses it. Muted would be the
882/// lie: [`State::Disabled`] resolves to it, so a five-option field read as one
883/// live row and four dead ones. `makeover-tui` draws it the same way
884/// (`makeover-tui@230bf63`); wiki `three-tone-convention` is the table.
885fn option_color(value: &str, option: &str, palette: &Palette) -> Color32 {
886 if value == option {
887 palette.content
888 } else {
889 palette.content_secondary
890 }
891}
892
893/// Which end of an interval a box is.
894///
895/// Named rather than a bool, because what it selects is not a side but a
896/// fallback: an empty end reads as the bound it stands for, and which bound
897/// that is depends on the end.
898#[derive(Debug, Clone, Copy, PartialEq, Eq)]
899enum Bound {
900 /// The lower end, falling back to the start of the extent.
901 Low,
902 /// The upper end, falling back to its end.
903 High,
904}
905
906/// The facts an interval's two boxes share.
907///
908/// One struct because they are one axis: [`Field::min`], [`Field::max`],
909/// [`Field::step`] and [`Field::unit`] describe the question rather than either
910/// end of it, so reading them once is what stops the two boxes drifting apart.
911struct Axis<'a> {
912 /// The extent both ends are dragged inside, when it is one this host can
913 /// read.
914 extent: Option<RangeInclusive<f64>>,
915 /// The granularity, as the description writes it.
916 step: Option<&'a str>,
917 /// What the axis is measured in.
918 unit: Option<&'a str>,
919}
920
921impl Axis<'_> {
922 /// What an empty end reads as: the bound it stands for.
923 ///
924 /// Zero with no extent to fall back on. That is the one number this
925 /// renderer invents, and it invents it where the description declined to
926 /// say anything: an unbounded interval has no edge for the end to sit on,
927 /// and a drag box has to start somewhere.
928 fn edge(&self, which: Bound) -> f64 {
929 self.extent.as_ref().map_or(0.0, |extent| match which {
930 Bound::Low => *extent.start(),
931 Bound::High => *extent.end(),
932 })
933 }
934
935 /// One end of the interval, as a drag box.
936 ///
937 /// # An empty end reads as its bound
938 ///
939 /// Which is what the shipped control did before it was described: an unset
940 /// minimum sits on the low edge and stores no filter. egui's `DragValue`
941 /// holds a number and has no empty state to offer, so the alternative was a
942 /// text box, and that would cost the app a control on the way into being
943 /// described.
944 ///
945 /// With no extent to fall back on, an empty end reads zero. That is the one
946 /// number this renderer invents, and it invents it where the description
947 /// declined to say anything: an unbounded interval has no edge for the end
948 /// to sit on, and a drag box has to start somewhere.
949 ///
950 /// Nothing is written back until the user drags, so a value the app put
951 /// there survives being looked at -- the same guarantee the slider makes.
952 fn end(&self, ui: &mut Ui, value: &mut String, which: Bound) -> Response {
953 let mut number = value.parse::<f64>().unwrap_or(self.edge(which));
954 let mut drag = DragValue::new(&mut number);
955 if let Some(extent) = self.extent.clone() {
956 drag = drag.range(extent);
957 }
958 if let Some(places) = decimals(self.step) {
959 drag = drag.max_decimals(places);
960 }
961 if let Some(step) = self.step.and_then(|s| s.parse::<f64>().ok()) {
962 drag = drag.speed(step);
963 }
964 // Inside the control, beside the readout, which is where `Field::unit`
965 // was decided to belong and where these boxes already put it.
966 if let Some(unit) = self.unit {
967 drag = drag.suffix(format!(" {unit}"));
968 }
969 let response = ui.add(drag);
970 if response.changed() {
971 *value = match decimals(self.step) {
972 Some(places) => format!("{number:.places$}"),
973 None => number.to_string(),
974 };
975 }
976 response
977 }
978}
979
980/// The control alone, without its label, hint or error.
981fn control(
982 ui: &mut Ui,
983 field: &Field<'_>,
984 filling: Filling<'_>,
985 palette: &Palette,
986 style: &FieldStyle,
987 named_by: Option<egui::Id>,
988) -> Response {
989 // The mismatch path: described as one thing and filled as another. Nothing
990 // here can fix it, so it is drawn as the empty, inert version of what was
991 // described — visible on screen, in the way an empty select is at the
992 // webview renderer, rather than reported in a log nobody reads.
993 let mut discard = String::new();
994 // The interval's second scratch buffer. Two ends means the mismatch path
995 // needs two places to write nothing to.
996 let mut spare = String::new();
997 let mut off = false;
998
999 match shape_of(field) {
1000 Control::Slid => {
1001 let value = match filling {
1002 Filling::Text(text) => text,
1003 _ => &mut discard,
1004 };
1005 // `shape_of` has already refused an unbounded range, so the extent
1006 // is only missing here if a bound is not a number — a date range,
1007 // say, which this control cannot draw either.
1008 let Some(extent) = extent(field) else {
1009 return ui.label(RichText::new(value.as_str()).color(palette.content));
1010 };
1011
1012 // A value the host cannot read starts at the low end rather than at
1013 // zero, which may be outside the extent entirely. Nothing is
1014 // written back until the user drags, so an unreadable value the app
1015 // put there survives being looked at.
1016 let mut number = value.parse::<f64>().unwrap_or(*extent.start());
1017 // The granularity is the curve's as of makeover-layout 0.32.0. It
1018 // is still in the value's own units, so the display precision is
1019 // read off it exactly as before.
1020 let step = field.curve.step();
1021 let mut slider = Slider::new(&mut number, extent.clone()).text("");
1022 if let Some(places) = decimals(step) {
1023 slider = slider.max_decimals(places);
1024 }
1025 if let Some(step) = step.and_then(|s| s.parse::<f64>().ok()) {
1026 slider = slider.step_by(step);
1027 }
1028 // egui's own constant-ratio track, which is this host's answer to
1029 // `Curve::Logarithmic`. `is_ratio` rather than a match on the
1030 // variant, because a ratio across zero is not one: makeover-layout
1031 // decides the fallback so that four renderers cannot disagree about
1032 // when it applies.
1033 if field.curve.is_ratio(*extent.start(), *extent.end()) {
1034 slider = slider.logarithmic(true);
1035 }
1036 // The unit goes inside the control, beside the readout egui already
1037 // draws. That placement is the argument `Field::unit` was decided
1038 // on: it is where these controls put it before they were described,
1039 // and it is the one a label could never reach.
1040 if let Some(unit) = unit_of(field) {
1041 slider = slider.suffix(format!(" {unit}"));
1042 }
1043 let response = ui.add(slider);
1044 if response.changed() {
1045 *value = match decimals(step) {
1046 Some(places) => format!("{number:.places$}"),
1047 None => number.to_string(),
1048 };
1049 }
1050 response
1051 }
1052 // One question, so one row. Two wells stacked would be two questions on
1053 // screen whatever the description said, which is the reading
1054 // `FieldKind::Interval` exists to prevent.
1055 //
1056 // Dragged rather than typed, because that is what these controls
1057 // already were: audiofiles' six filter axes are `DragValue` pairs with
1058 // a shared extent, a speed and a suffix, and a port that turned them
1059 // into text boxes would be a description costing the app a control.
1060 Control::Spanned => {
1061 let (lower, upper) = match filling {
1062 Filling::Between { lower, upper } => (lower, upper),
1063 _ => (&mut discard, &mut spare),
1064 };
1065 let axis = Axis {
1066 extent: extent(field),
1067 step: field.step,
1068 unit: unit_of(field),
1069 };
1070 ui.horizontal(|ui| {
1071 let low = axis.end(ui, lower, Bound::Low);
1072 // The word rather than a dash. A dash between two numbers is a
1073 // minus sign on a signed axis, and audiofiles filters loudness
1074 // in dBFS.
1075 ui.label(RichText::new("to").color(palette.content_secondary));
1076 let high = axis.end(ui, upper, Bound::High);
1077 // Both ends, by name. A union response carries the first id, so
1078 // labelling the union outside this arm names the lower box and
1079 // leaves the upper one announced as whatever number is in it --
1080 // which is how an interval half-kept the fix that gave every
1081 // other shape its question.
1082 if let Some(id) = named_by {
1083 low.clone().labelled_by(id);
1084 high.clone().labelled_by(id);
1085 }
1086 low | high
1087 })
1088 .inner
1089 }
1090 Control::Typed => {
1091 let text = match filling {
1092 Filling::Text(text) => text,
1093 _ => &mut discard,
1094 };
1095 // An empty frame and no margin: the well is this crate's, and egui's
1096 // own control background and padding would sit underneath it saying
1097 // something different about both.
1098 // Keyed on the description's own `multiline` and not on the
1099 // member: a markdown field is several lines by definition, and a
1100 // single-line edit would be a control the value cannot fit in. egui
1101 // does nothing else with the markdown, which is the honest answer
1102 // rather than a gap -- the source is text, and editing it as text
1103 // loses none of it.
1104 let mut edit = if field.kind.multiline() {
1105 TextEdit::multiline(text)
1106 } else {
1107 TextEdit::singleline(text)
1108 }
1109 .frame(egui::Frame::NONE)
1110 .margin(Margin::ZERO)
1111 .text_color(palette.content)
1112 .password(field.kind.confidential());
1113 if let Some(ghost) = field.placeholder {
1114 edit = edit.hint_text(RichText::new(ghost).color(palette.content_muted));
1115 }
1116 let response = frame(ui, Depth::Well, palette, style.frame, |ui| ui.add(edit));
1117 // A typed number has no readout of its own to sit beside, so the
1118 // unit follows the box. Muted, because it is a fact about the value
1119 // rather than a second thing to read.
1120 match unit_of(field) {
1121 Some(unit) => {
1122 ui.label(RichText::new(unit).color(palette.content_muted));
1123 response
1124 }
1125 None => response,
1126 }
1127 }
1128 Control::Toggled => {
1129 let on = match filling {
1130 Filling::On(on) => on,
1131 _ => &mut off,
1132 };
1133 ui.checkbox(on, RichText::new(field.label).color(palette.content))
1134 }
1135 Control::Listed => {
1136 let value = match filling {
1137 Filling::Text(text) => text,
1138 _ => &mut discard,
1139 };
1140 // No `shown_label` counterpart, and none is needed: a value no
1141 // option carries leaves every button unfilled, which is already
1142 // the honest report on screen. The select needs the fix because it
1143 // has one slot and must put *something* in it.
1144 let group = ui.vertical(|ui| {
1145 let mut answered: Option<Response> = None;
1146 for opt in field.options {
1147 // An option that cannot be picked yet is drawn and does not
1148 // answer, with the precondition beside it rather than
1149 // behind a hover: a greyed row with no reason reads as a
1150 // dead end, which is the state `Choice::unavailable` exists
1151 // to stop being sayable.
1152 let picked = if let Some(reason) = opt.unavailable {
1153 ui.horizontal(|ui| {
1154 let picked = ui
1155 .add_enabled_ui(false, |ui| {
1156 ui.radio_value(
1157 value,
1158 opt.value.to_owned(),
1159 RichText::new(opt.label).color(palette.content_muted),
1160 )
1161 })
1162 .inner;
1163 ui.label(RichText::new(reason).color(palette.content_muted));
1164 picked
1165 })
1166 .inner
1167 } else {
1168 ui.radio_value(
1169 value,
1170 opt.value.to_owned(),
1171 RichText::new(opt.label).color(option_color(value, opt.value, palette)),
1172 )
1173 };
1174 // What picking it means, under the option rather than
1175 // beside it. makeover-layout 0.39.0, and the placement is
1176 // the difference from the reason above: a precondition is
1177 // short enough to sit on the line, and a sentence saying
1178 // what a tier is would push every option's control out of
1179 // line with its neighbours.
1180 //
1181 // Indented past the radio, so it reads as belonging to the
1182 // option above rather than as a label of its own. Muted,
1183 // the same reading `.form-option-detail` takes in the
1184 // webview: the line orients the label, it does not compete
1185 // with it.
1186 if let Some(detail) = opt.detail {
1187 ui.horizontal(|ui| {
1188 ui.add_space(OPTION_DETAIL_INDENT);
1189 ui.label(RichText::new(detail).color(palette.content_muted));
1190 });
1191 }
1192 answered = Some(match answered {
1193 Some(prev) => prev.union(picked),
1194 None => picked,
1195 });
1196 }
1197 answered
1198 });
1199 // A group described with no options answers as its own empty area
1200 // rather than as no response at all, which keeps the caller's
1201 // `.changed()` chain working on a field whose option list has not
1202 // loaded yet.
1203 group.inner.unwrap_or(group.response)
1204 }
1205 Control::Chosen => {
1206 let value = match filling {
1207 Filling::Text(text) => text,
1208 _ => &mut discard,
1209 };
1210 let (shown, tone) = chosen_text(field, value, palette);
1211 ComboBox::from_id_salt(field.name)
1212 .selected_text(RichText::new(shown).color(tone))
1213 .show_ui(ui, |ui| {
1214 for opt in field.options {
1215 // Same rule as the radio group: shown, inert, and
1216 // saying why. A closed control hides its list, so the
1217 // reason has to travel with the row it belongs to.
1218 // A closed control hides its list, so everything an
1219 // option carries has to travel with the row it belongs
1220 // to. That is why both extra strings run into the text
1221 // here and neither does in the group above: a combo
1222 // row is a row, the way a `<select>`'s option is.
1223 let text = combo_row(opt);
1224 if opt.unavailable.is_some() {
1225 ui.add_enabled_ui(false, |ui| {
1226 ui.selectable_value(
1227 value,
1228 opt.value.to_owned(),
1229 RichText::new(text).color(palette.content_muted),
1230 );
1231 });
1232 continue;
1233 }
1234 ui.selectable_value(
1235 value,
1236 opt.value.to_owned(),
1237 RichText::new(text).color(option_color(value, opt.value, palette)),
1238 );
1239 }
1240 })
1241 .response
1242 }
1243 // The grouping comes out of the order rather than out of a group list:
1244 // `Field::themes` arrives sorted by variant, so the run of one variant
1245 // is the group and a heading opens whenever the variant changes. Same
1246 // walk as `makeover-webview`'s `<optgroup>` emission, which is what
1247 // keeps two renderers from disagreeing about where a group starts.
1248 Control::Themed => {
1249 let value = match filling {
1250 Filling::Text(text) => text,
1251 _ => &mut discard,
1252 };
1253 let shown = themed_text(field, value);
1254 ComboBox::from_id_salt(field.name)
1255 .selected_text(RichText::new(shown).color(palette.content))
1256 .show_ui(ui, |ui| {
1257 if let Some(follow) = field.follows {
1258 // First and outside every heading. It names no theme
1259 // and sits in no variant, so a heading over it would be
1260 // inventing a fourth variant for one row.
1261 ui.selectable_value(
1262 value,
1263 follow.value.to_owned(),
1264 RichText::new(follow.label).color(option_color(
1265 value,
1266 follow.value,
1267 palette,
1268 )),
1269 );
1270 }
1271 let mut open: Option<ThemeVariant> = None;
1272 for theme in field.themes {
1273 if open != Some(theme.variant) {
1274 // A heading rather than a `selectable_value`: it is
1275 // not pickable, and egui has no inert row that
1276 // still reads as a row. `content_muted` is the tone
1277 // for something that is not an answer, which is the
1278 // ghost text's tone eight lines up.
1279 if open.is_some() {
1280 ui.separator();
1281 }
1282 ui.label(
1283 RichText::new(theme.variant.heading()).color(palette.content_muted),
1284 );
1285 open = Some(theme.variant);
1286 }
1287 ui.selectable_value(
1288 value,
1289 theme.id.to_owned(),
1290 RichText::new(format!("{} {}", theme.name, theme.contrast.badge()))
1291 .color(option_color(value, theme.id, palette)),
1292 );
1293 }
1294 })
1295 .response
1296 }
1297 }
1298}
1299
1300/// What a theme picker's closed control reads.
1301///
1302/// [`chosen_text`]'s counterpart, and it is separate for the reason
1303/// [`Control::Themed`] is: the label lives on a [`makeover_layout::ThemeChoice`]
1304/// rather than on a [`Choice`], and the follow row is a third place to look.
1305///
1306/// No placeholder arm. A theme picker is never unanswered in the way a select
1307/// is — an app that resolved a theme to paint this control with has one — and
1308/// falling back to the raw value is the honest report on a stored id whose
1309/// theme has since been deleted.
1310fn themed_text<'a>(field: &'a Field<'a>, value: &'a str) -> &'a str {
1311 if let Some(follow) = field.follows
1312 && follow.value == value
1313 {
1314 return follow.label;
1315 }
1316 field
1317 .themes
1318 .iter()
1319 .find(|theme| theme.id == value)
1320 .map_or(value, |theme| theme.name)
1321}
1322
1323/// One field, as the column the app drops into its form.
1324///
1325/// The anatomy is `makeover-webview`'s, so the two renderers put a form
1326/// together the same way: label, control, hint, error, top to bottom, with a
1327/// checkbox labelling itself instead of taking a label above.
1328///
1329/// Returns [`None`] for a [`FieldKind::Hidden`] field, which is what
1330/// [`FieldKind::visible`] means and is the honest answer here: a webview still
1331/// emits an input for it because the form submits, and an immediate-mode
1332/// renderer has no form and no submission, so a hidden field is a value the app
1333/// already holds and there is nothing to draw or to respond to.
1334///
1335/// `state` is the description's interaction axis.
1336/// [`State::Disabled`] greys the field and stops it answering, through
1337/// [`State::suppresses_interaction`] rather than through a second reading of
1338/// what disabled means. Focus is not on that axis and never reaches here: egui
1339/// owns reach, focus and the ring for this renderer, and one ring means not a
1340/// second one per renderer that happens to have opinions.
1341pub fn field(
1342 ui: &mut Ui,
1343 field: &Field<'_>,
1344 filling: Filling<'_>,
1345 state: Option<State>,
1346 palette: &Palette,
1347 style: &FieldStyle,
1348) -> Option<Response> {
1349 if !field.kind.visible() {
1350 return None;
1351 }
1352 let enabled = !state.is_some_and(State::suppresses_interaction);
1353 let text = if enabled {
1354 palette.content
1355 } else {
1356 palette.content_muted
1357 };
1358
1359 let response = ui
1360 .vertical(|ui| {
1361 ui.spacing_mut().item_spacing.y = style.gap;
1362
1363 // A checkbox labels itself, on the right of the box.
1364 // `FieldKind::labels_itself` is the description saying so, and both
1365 // webview apps special-cased it inline before it did.
1366 let named_by = (!field.kind.labels_itself())
1367 .then(|| ui.label(RichText::new(label_text(field, style)).color(text)));
1368
1369 let response = ui
1370 .add_enabled_ui(enabled, |ui| {
1371 control(
1372 ui,
1373 field,
1374 filling,
1375 palette,
1376 style,
1377 named_by.as_ref().map(|l| l.id),
1378 )
1379 })
1380 .inner;
1381
1382 // The label, attached rather than merely adjacent.
1383 //
1384 // Drawing it above the control and stopping there is what this did
1385 // until 2026-08-22, and it put an unnamed box in the accessibility
1386 // tree with some text near it: a screen reader announced a text
1387 // field with no question, and a prefilled one announced its own
1388 // contents instead. audiofiles' four name modals were the site that
1389 // measured it, through a harness reading what the panel drew.
1390 //
1391 // Worth stating why it was worth fixing here rather than in each
1392 // app: `Field::label` is a member the description carries so a
1393 // renderer does not have to guess, `makeover-webview` has always
1394 // named it in `aria-describedby`, and the two renderers disagreeing
1395 // about a fact the description states is the one thing this layer
1396 // exists to prevent. A checkbox is unaffected -- egui names one from
1397 // its own text, which is what `labels_itself` already says.
1398 let response = match named_by {
1399 Some(label) => response.labelled_by(label.id),
1400 None => response,
1401 };
1402
1403 // Standing help, then what the answer costs, then what is wrong
1404 // now. All three, in that order, for the reason the webview
1405 // renderer names all three in `aria-describedby`: an error
1406 // appearing must not take the hint away with it. This host has the
1407 // room, so unlike makeover-tui it never has to choose -- the
1408 // precedence rule on `Field::note` is for the renderer that does.
1409 if let Some(hint) = field.hint {
1410 ui.label(RichText::new(hint).color(palette.content_muted));
1411 }
1412 // The note carries its own tone, and `Palette::tone` is what
1413 // resolves it, so a Neutral note is ordinary content rather than
1414 // a colour this renderer picked.
1415 if let Some((tone, note)) = field.note {
1416 ui.label(RichText::new(note).color(palette.tone(tone)));
1417 }
1418 if let Some(error) = field.error {
1419 ui.label(RichText::new(error).color(palette.danger));
1420 }
1421 response
1422 })
1423 .inner;
1424
1425 Some(response)
1426}
1427
1428/// A set of fields, laid down a column.
1429///
1430/// `show_extended` is the disclosure, and it is a parameter rather than state
1431/// held here because the disclosure belongs to the *form* and not to any field:
1432/// [`Field::extended`] marks which fields are behind one, and the app owns
1433/// whether it is open. That is the same division `makeover-webview` draws when
1434/// it marks the group `data-extended` and emits no control to toggle it.
1435///
1436/// `draw` is called once per field that should be visible, in order. Taking a
1437/// callback rather than a slice of [`Filling`]s is what keeps the app's own
1438/// values borrowed one at a time: a form's fields usually live in different
1439/// structs, and a parallel array would have to be built each frame and kept in
1440/// step with the description by hand.
1441pub fn group<'a>(
1442 ui: &mut Ui,
1443 fields: &'a [Field<'a>],
1444 show_extended: bool,
1445 style: &FieldStyle,
1446 mut draw: impl FnMut(&mut Ui, &'a Field<'a>),
1447) {
1448 ui.vertical(|ui| {
1449 ui.spacing_mut().item_spacing.y = style.group_gap;
1450 for f in fields {
1451 if f.extended && !show_extended {
1452 continue;
1453 }
1454 draw(ui, f);
1455 }
1456 });
1457}
1458
1459#[cfg(test)]
1460mod tests;