Self-hosted, Self-Improving, Federated Everything App — dev management, agent orchestration, persistent mind, digital twin, second brain, and creative pipeline from a single dashboard over Tailscale to port the user into the OS
JavaScript ai-agentsdashboarddeveloper-toolsdigital-twinquantified-selfsecond-brainself-hostedtailscale
/**
* Client mirror of `server/lib/threejsModelRig.js`'s readiness rule, plus the
* joint lookup the preview needs to label a picked part.
*
* The server writes an authoritative report onto the record at generation time.
* The preview is handed a bare `spec` (and a record generated before the report
* shipped has none at all), so it derives the same verdict from the spec rather
* than inventing a second, looser one — a preview that says "articulation-ready"
* while the detail panel says "static" is worse than either answer alone.
*
* Keep the thresholds here and in the server module in step.
*/
const MIN_USEFUL_JOINTS = 2;
/**
* Client mirror of `resolveThreejsAttachments` in `server/lib/threejsModel.js`.
*
* A spec declares carried parts in two forms — the original anchor-less
* `attachmentPartIds` and the anchored `attachments` — and both mean the same
* thing, so the counts are derived from one merged list. An anchored entry wins
* over a bare id for the same part.
*
* @param {object|null} articulation a spec's `articulation` object, or null
* @returns {Array<{partId: string, anchored: boolean}>}
*/
function resolveAttachments(articulation) {
const entries = new Map();
const bareIds = Array.isArray(articulation?.attachmentPartIds) ? articulation.attachmentPartIds : [];
for (const partId of bareIds) {
if (typeof partId !== 'string' || entries.has(partId)) continue;
entries.set(partId, { partId, anchored: false });
}
const anchored = Array.isArray(articulation?.attachments) ? articulation.attachments : [];
for (const attachment of anchored) {
if (typeof attachment?.partId !== 'string') continue;
entries.set(attachment.partId, {
partId: attachment.partId,
anchored: Boolean(attachment.anchorPartId) || Boolean(attachment.anchorSocket),
});
}
return [...entries.values()];
}
/**
* @param {object|null} spec a validated sculpt spec, or null
* @returns {{articulationReady: boolean, jointCount: number, socketCount: number,
* attachmentCount: number, anchoredAttachmentCount: number,
* unanchoredAttachmentCount: number, jointsByPartId: object}}
*/
export function summarizeThreejsArticulation(spec) {
const joints = Array.isArray(spec?.articulation?.joints) ? spec.articulation.joints : [];
const attachments = resolveAttachments(spec?.articulation);
const unanchoredAttachmentCount = attachments.filter((attachment) => !attachment.anchored).length;
// Null-prototype: part ids are provider-authored and the id schema accepts
// `toString`, so a bare lookup on a plain object can hand back a function.
const jointsByPartId = Object.create(null);
for (const joint of joints) jointsByPartId[joint.partId] = joint;
const unpivoted = joints.filter((joint) => joint.parentJointId && !joint.pivotSocket);
return {
// An attachment that names nothing to hang from is a declaration with no
// relationship in it, so — same rule the server applies — it keeps the graph
// out of "ready" rather than being credited toward it.
articulationReady: joints.length >= MIN_USEFUL_JOINTS
&& unpivoted.length === 0
&& unanchoredAttachmentCount === 0,
jointCount: joints.length,
socketCount: new Set(joints.map((joint) => joint.pivotSocket).filter(Boolean)).size,
attachmentCount: attachments.length,
anchoredAttachmentCount: attachments.length - unanchoredAttachmentCount,
unanchoredAttachmentCount,
jointsByPartId,
};
}