# Chromium DevTools UI Engineering ## Objective and scope This document defines how to build Chromium DevTools UI. It aims at improving consistency, maintainability and extensibility of the current code **Consistency** here means to have one way to do one thing, in the context of this doc to have a single reusable component per repeated task. **Maintainability** is addressed here through the [separation of concerns](https://en.wikipedia.org/wiki/Separation_of_concerns) while avoiding unnecessary indirection. **Extensibility** requires the ease of understanding and imitation, i.e. being able to take an existing code, understand and use it as an example to solve another problem. Additionally, all the changes need to be applicable to the existing code **point-wise** instead of requiring extensive rewriting. ## Reusable web components Common UI primitives need to be implemented as web components. Examples include buttons, checkboxs, or data grids. Web components should not be used solely for encapsulation, we should not have single-use web components. Not all reusable code should be a web component: in the Application panel we have several similar views showing key-value pairs in a datagrid with a preview sidebar. Making this a truly reusable component will lead to an unjustifiable complexity. Instead we should have extracted a base class or helpers implementing the common functionality. In implementation we should prefer wrapping existing code under ui/legacy over new implementation. The former is the most feature-rich (including less obvious aspects like a11y) and has stood the test of time. We should however strive to expose a “HTML-native” API: e.g. toolbar doesn’t need an `items` setter if its child HTML elements could define its content. Even the data grid doesn’t need to expose data grid nodes, when `` and ``’s are enough. ## Model-View-Presenter architecture We should strictly separate business logic from UI logic and presentation. This means that most of the UI code should be centered around a presenter (a subclass of `UI.Widget`) that gets a view function injected. All the logic that is not related to the DevTools UI (i.e., that would stay the same if we were to rewrite DevTools as a command-line tool) should belong to the model layer. The presenter should make no assumptions about the details of the model or view code. It should not care what race conditions CDP exposes or how many layers of `

` the markup has. For testability and simplicity, the view function should be injectable into the presenter's constructor. The presenter should also provide a default view. The recommended pattern is to define the default rendering logic in a separate, exported function named `DEFAULT_VIEW`. The presenter's constructor can then use this as the default value for its `view` parameter. This approach has two main benefits: 1. **Testability**: In unit tests, we can pass a simple stub as the view function, which allows us to test the presenter's logic without any DOM manipulation. 2. **Clarity**: It cleanly separates the presenter's logic from its rendering logic. To test the `DEFAULT_VIEW` function itself, we should use screenshot and e2e tests. ## Declarative and orchestrated DOM updates We should no longer use imperative API to update DOM. Instead we rely on orchestrated rendering of lit-html templates. The view function described above should be a call to lit-html `render`. The view function should be called from `UI.Widget`’s `performUpdate` method, which by default is scheduled using `requestAnimationFrame`. To embed another presenter (`UI.Widget`) in the lit-html template, use `widget(, {foo: 1, bar: 2})` This will instantiate a `Widget` class with the web component as its `element` and, optionally, will set the properties provided in the second parameter. The widget won’t be re-instantiated on the subsequent template renders, but the properties would be updated. For this to work, the widget needs to accept `HTMLElement` as a sole constructor parameter and properties need to be public members or setters. For backwards compatibility, the first argument to `widget` can also be a factory function: `widget(element => new MyWidget(foo, bar, element))`. Similar to the class constructor version, `element` is the actual `` so the following two invocations of `widget` are equivalent: `widget(MyWidget)` and `widget(element => new MyWidget(element))`. ## Styling To prevent style conflicts in widgets without relying on shadow DOM, we use the CSS [`@scope`](https://developer.mozilla.org/en-US/docs/Web/CSS/@scope) at-rule for style encapsulation. This ensures that styles defined for a widget do not leak out and affect other components. The convention is to add the `@scope` rule directly into the widget's .css file. The scope's "lower boundary" is set to `(devtools-widget > *)`. This prevents the styles from cascading into the contents of any nested child widgets, while still allowing the parent to style the `` element itself. First, define the styles within an @scope block in your CSS file: ```css /* my-widget.css */ @scope to (devtools-widget > *) { /* Use :scope to style the widget's container element itself. */ :scope { width: 100%; box-shadow: none; } .title { font-size: 1.2em; color: var(--sys-color-tonal-on-container); } } ``` Then, import and use these styles in your widget's view function: ```ts /* myWidget.ts */ import {html} from 'lit-html'; import * as UI from '../../ui/legacy/legacy.js'; import myWidgetStyles from './myWidget.css.js'; render(html`

My Widget

... ${widget(NestedWidget)}

`, this.element); ``` In this example, the `.title` style will apply within the parent widget but will not leak into the nested widget. Because this convention relies on developer discipline, it is important to verify its correct application during code reviews. **Note:** Inside a ``, any `