Clustering with option minPoints 1

With option minPoints: 1, my assumption would be that a cluster could exist with 1 point in the cluster area, but since minPoints is 1, it would have zero neighbors.

Including another if (!clusterProperties) clusterProperties = this._map(p, true); after the for loop solves this by setting the clusterProperties when no neighbors are found.

Fork with change: https://github.com/michaelfarrelly/supercluster/commit/3dbbc52fc0c7bb8a20cc52ac338f74b3c4ddfca0

If this against the purpose of this property, then I can revoke this request. My projects goal was to have clustering only (no points) when zoomed out.

https://github.com/mapbox/supercluster/blob/main/index.js#L266-L282

for (const neighborId of neighborIds) {
	const b = tree.points[neighborId];

	if (b.zoom <= zoom) continue;
	b.zoom = zoom; // save the zoom (so it doesn't get processed twice)

	const numPoints2 = b.numPoints || 1;
	wx += b.x * numPoints2; // accumulate coordinates for calculating weighted center
	wy += b.y * numPoints2;

	b.parentId = id;

	if (reduce) {
		if (!clusterProperties) clusterProperties = this._map(p, true);
		reduce(clusterProperties, this._map(b));
	}
}

=>

for (const neighborId of neighborIds) {
	const b = tree.points[neighborId];

	if (b.zoom <= zoom) continue;
	b.zoom = zoom; // save the zoom (so it doesn't get processed twice)

	const numPoints2 = b.numPoints || 1;
	wx += b.x * numPoints2; // accumulate coordinates for calculating weighted center
	wy += b.y * numPoints2;

	b.parentId = id;

	if (reduce) {
		if (!clusterProperties) clusterProperties = this._map(p, true);
		reduce(clusterProperties, this._map(b));
	}
}

if (!clusterProperties) clusterProperties = this._map(p, true);

[shide1989]

Encountered same situation BUT : My opinion is that the minPointspurpose is different from the clusterOnly purpose where the getClustersmethod should only return clusters. The goal of the minPoints options is to define the minimum of points to be able to create a Cluster. Whereas a clusterOnly options would force the algorithm to return clusters.

IMO Both parameters could be used, for ex to have clusters with a minPoints = 3 and clustersOnly = truewhere it would form normal clusters of 3 points minimum, and the other points would be clusterized in clusters of 1 point.

[FR073N]

Same for me. clusterOnly can be a good property to have a proper display with clusters only.