Open source Loom alternative. Beautiful, shareable screen recordings.
Rust appcapcossloommacnextjsnextjs15open-sourceossreactrecordrustscreen-capturescreen-recorderscreenshotsolidjstauritauri-apptypescriptvite
/**
* The worked example the timeline post's demo runs on: one 2:48 product
* walkthrough, seven pieces of feedback on it, and a synthetic soundtrack.
* Nothing here talks to the network or to a real recording; the frames are
* drawn (see `MockFrame`) and the waveform is a deterministic function of
* time, so the demo is fully self-contained.
*/
/** Length of the fictional recording, in seconds. */
export const DEMO_DURATION = 168;
export type DemoCommentKind = "text" | "emoji" | "screen" | "camera" | "voice";
export interface DemoComment {
id: string;
author: string;
/** Avatar letter; the demo has no real images to sign. */
initials: string;
/** Tailwind background class for the avatar. */
tone: string;
/** Seconds into the recording. */
t: number;
kind: DemoCommentKind;
/** Comment body, the emoji for reactions, or what a recorded reply shows. */
text: string;
/** Media length for video/voice replies, in seconds. */
mediaDuration?: number;
}
/**
* One of everything the timeline can hold: text, an emoji reaction, a screen
* reply, a voice note and a camera reply. Sam and Priya sit three seconds
* apart on purpose, so the second lane earns its keep, while the recorded
* replies are spread out far enough that every glyph stays visible instead of
* folding into a cluster.
*/
export const DEMO_COMMENTS: readonly DemoComment[] = [
{
id: "c1",
author: "Maya",
initials: "M",
tone: "bg-blue-9",
t: 12,
kind: "text",
text: "Can we open on the problem rather than the dashboard?",
},
{
id: "c2",
author: "Dev",
initials: "D",
tone: "bg-emerald-500",
t: 44,
kind: "emoji",
text: "π₯",
},
{
id: "c3",
author: "Sam",
initials: "S",
tone: "bg-violet-500",
t: 72,
kind: "text",
text: "The logo flickers the moment the dialog opens.",
},
{
id: "c4",
author: "Priya",
initials: "P",
tone: "bg-amber-500",
t: 75,
kind: "screen",
text: "Screen recording of the flicker in Safari",
mediaDuration: 19,
},
{
id: "c5",
author: "Jo",
initials: "J",
tone: "bg-rose-500",
t: 96,
kind: "voice",
text: "Voice note talking through a fix",
mediaDuration: 14,
},
{
id: "c6",
author: "Ana",
initials: "A",
tone: "bg-teal-500",
t: 124,
kind: "camera",
text: "Camera reply with a suggestion",
mediaDuration: 26,
},
{
id: "c7",
author: "Alex",
initials: "A",
tone: "bg-sky-600",
t: 151,
kind: "text",
text: "Ship it. I'll take the copy pass.",
},
];
/**
* When someone is talking. The holes at 1:01β1:10 and 2:09β2:18 are the dead
* air the speech envelope makes visible.
*/
const SPEECH_SEGMENTS: readonly (readonly [number, number])[] = [
[2, 61],
[70, 129],
[138, 166],
];
const hash = (n: number) => {
const x = Math.sin(n * 12.9898 + 78.233) * 43758.5453;
return x - Math.floor(x);
};
const isSpeaking = (t: number) =>
SPEECH_SEGMENTS.some(([start, end]) => t >= start && t <= end);
/**
* Loudness at a moment, 0-1. Two layers so it reads as a voice rather than as
* noise: a slow syllable swell, and a per-50ms grain on top. Room tone under
* the pauses, because a flat zero looks like a broken render.
*/
export function speechEnergy(t: number): number {
const grain = hash(Math.round(t * 20));
if (!isSpeaking(t)) return 0.04 + 0.05 * grain;
const syllable = 0.55 + 0.45 * Math.sin(t * 2.7);
const phrase = 0.75 + 0.25 * Math.sin(t * 0.6 + 1.2);
return Math.min(1, 0.2 + 0.8 * syllable * phrase * (0.82 + 0.32 * grain));
}
export function formatClock(seconds: number): string {
const safe = Number.isFinite(seconds) ? Math.max(0, seconds) : 0;
const mins = Math.floor(safe / 60);
const secs = Math.floor(safe % 60);
return `${mins}:${secs.toString().padStart(2, "0")}`;
}
export const clamp = (value: number, min: number, max: number) =>
Math.min(max, Math.max(min, value));