Develop Reference

How to change cluster zoom level on click with Leaflet map?

I have a leaflet map that has zoom levels 2-7 and uses the MarkerCluster plugin, by default I have the L.MarkerClusterGroup disable clustering a zoom level 2 (which means no clustering) and I'm trying to allow the user to click a button that then changes the clustering zoom level to 5. Is this possible?
I know I could do it by making two markercluster groups, one that has no clustering and one that has clustering and remove/add it based on click but that just seems incredibly messy. Really, there's several ways to do it but they are so incredibly clunky.
Code:
Default (2 is the lowest level of zoom):
var markers = new L.MarkerClusterGroup (
{
disableClusteringAtZoom: 2,
maxClusterRadius: 100,
animateAddingMarkers: true
});
What I want to do be able to do:
$('#mcluster').click(function() {
//do some code that sets the disableClusterAtZoom to 5
});

I could not find a way to disable clustering or set a new value for disableClustering at zoom, but I found a less clunky way of achieving this.
var markers = new L.LayerGroup(); //non cluster layer is added to map
markers.addTo(map);
var clusters = new L.MarkerClusterGroup (
{
disableClusteringAtZoom: 5,
maxClusterRadius: 100,
animateAddingMarkers: true
}); //cluster layer is set and waiting to be used
var clusterStatus = 'no'; //since non cluster group is on by default, the status for cluster is set to no
$('#mcluster').click(function( event ) {
if(clusterStatus === 'no'){
clusterStatus = 'yes';
var current1 = markers.getLayers(); //get current layers in markers
map.removeLayer(markers); // remove markers from map
clusters.clearLayers(); // clear any layers in clusters just in case
current1.forEach(function(item) { //loop through the current layers and add them to clusters
clusters.addLayer(item);
});
map.addLayer(clusters);
} else {
clusterStatus = 'no'; //we're turning off clustering here
var current2 = clusters.getLayers(); //same code as before just reversed
map.removeLayer(clusters);
markers.clearLayers();
current2.forEach(function(item) {
markers.addLayer(item);
});
map.addLayer(markers);
}
});
I'm sure there is a more elegant solution but with my still growing knowledge this is what I came up with.

I know you needed a solution a few months ago, but just to let you know that I released recently a sub-plugin for Leaflet.markercluster that can perform exactly what you are looking for (with a few extra code): Leaflet.MarkerCluster.Freezable (demo here).
var mcg = L.markerClusterGroup().addTo(map),
disableClusteringAtZoom = 2;
function changeClustering() {
if (map.getZoom() >= disableClusteringAtZoom) {
mcg.disableClustering(); // New method from sub-plugin.
} else {
mcg.enableClustering(); // New method from sub-plugin.
}
}
map.on("zoomend", changeClustering);
$('#mcluster').click(function () {
disableClusteringAtZoom = (disableClusteringAtZoom === 2) ? 5 : 2;
changeClustering();
});
mcg.addLayers(arrayOfMarkers);
// Initially disabled, as if disableClusteringAtZoom option were at 2.
changeClustering();
Demo: http://jsfiddle.net/fqnbwg3q/3/
Note: in the above demo I used a refinement to make sure the markers merge with animation when clustering is re-enabled. Simply use a timeout before using enableClustering():
// Use a timeout to trigger clustering after the zoom has ended,
// and make sure markers animate.
setTimeout(function () {
mcg.enableClustering();
}, 0);

Google Maps - get values from location array based on the map's current viewpoint

Edit:
Question = "is there a way to loop through the array and check if each location (long/lat) falls within the current viewport directly" (failing that get all markers within the viewport)
Background:
I have an array of locations (lat, long, id).
I want to:
On a Google Map, use the location array to display markers.
The user can scroll/zoom the map.
Have a button underneath the map, so when the user has decided on an area, he can click the button, and the code will return the ids (from the location array) that are contained within the viewport / map bounds.
There is a .contains for Google, so I guess you could potentially use that with something like
map.getBounds().contains and somehow reference each marker.getPosition()
but I wonder if there's a way to loop through the array and check if each location (long/lat) falls within the current viewport directly

You mean something like this (not tested), map is the google.maps.Map object and needs to be in scope. markersArray is the array of markers.
for (var i=0; i< markersArray.length; i++) {
if (map.getBounds().contains(markersArray[i].getPosition())) {
// the marker is in view
} else {
// the marker is not in view
}
}

http://jsfiddle.net/UA2g2/1/
Thanks geocodezip, you gave me the idea on how to solve it via looping through the array. I don't know if this is the most efficient way, but I put together some code that seems to do what I want - if you check the jsfiddle above and view console you can see that it logs when and which points are in the viewport.
$(document).ready(function(){
var myOptions = {
center: new google.maps.LatLng(51, -2),
zoom: 9,
mapTypeId: google.maps.MapTypeId.ROADMAP
};
var map = new google.maps.Map(document.getElementById("map_canvas"), myOptions);
var storeArray = new Array(["51.38254", "-2.362804", "ID1"], ["51.235249", "-2.297804","ID2"], ["51.086126", "-2.910767","ID3"]);
google.maps.event.addListener(map, 'idle', function() {
for (i = 0; i < storeArray.length; i++) {
marker = new google.maps.Marker({
position: new google.maps.LatLng(storeArray[i][0], storeArray[i][1]),
map: map
});
}
for (var i=0; i<storeArray.length; i++) {
if (map.getBounds().contains(new google.maps.LatLng(storeArray[i][0], storeArray[i][1]))) {
console.log("marker: " + storeArray[i][2]);
}
}
});
});

Google Maps check if circle area is visible within map boundaries

I have a Google Map which contains a circle area.
I want to know if the circle is currently visible within the map's boundaries.
What I found so far is to check whether the center of the circle is within the boundaries, but I want to check for the whole circle and not only it's center.
My code that checks if the map's current center is within the boundaries of the circle is this:
google.maps.event.addListener(map, 'bounds_changed', function()
{
var circleBounds = circle.getBounds();
console.log(circleBounds.contains(map.getCenter()));
});
So what i want is something like this, which of course is not correct:
circleBounds.contains(map.getBounds());

Determine what the northmost, southmost, westmost, and eastmost LATLNGs of the circle are. You can find this given the circle radius
Determine if all points are inside the viewport bounds
If true, then yes, the circle must be viewable!
and you really should go back to your old questions and ACCEPT them (click on the check mark outline next to the best answer). This is how you show your appreciation for the hard work your answerers provided.

This is an old question - dinosaur-old in internet years - but if you're comparing two bounds, then the following holds a fortiori, n'est-ce pas?:
if(a.getBounds().contains(b.getBounds().getNorthEast())
&& a.getBounds().contains(b.getBounds().getSouthWest()))
{console.log('B is within A... Bounds are RECTANGLES: \
You only need to test two \
diagonally-opposing corners')};
That's because for a rectangle R, SW is (max(x),min(y)); NE is (min(x),max(y)) - and therefore all (x,y) in R are covered by the test.
I sure hope that's the case - it's how I do all my bounds comparisons...

Thanks Tina CG Hoehr.
Just in case anyone else wants this, here is the code for my question:
// Get the bounds
var circleBounds = circle.getBounds();
var ne = circleBounds.getNorthEast(); // LatLng of the north-east corner
var sw = circleBounds.getSouthWest();
var nw = new google.maps.LatLng(ne.lat(), sw.lng());
var se = new google.maps.LatLng(sw.lat(), ne.lng());
google.maps.event.addListener(map, 'bounds_changed', function()
{
var mapBounds = map.getBounds();
// Log whether the circle is inside or outside of the map bounds
if(mapBounds.contains(ne))
{
console.log("northeast is viewable");
}
if(mapBounds.contains(sw))
{
console.log("southwest is viewable");
}
if(mapBounds.contains(nw))
{
console.log("northwest is viewable");
}
if(mapBounds.contains(se))
{
console.log("southeast is viewable");
}
});

google maps middle of a polyline (centroid?)

I have a list of polylines, just like google maps does when I click on the polyline I want an infowindow to show up just where I clicked, and it works just fine with this function
function mapsInfoWindow(polyline, content) {
google.maps.event.addListener(polyline, 'click', function(event) {
infowindow.content = content;
infowindow.position = event.latLng;
infowindow.open(map);
});
}
the problem comes when I click on the list(using the same function for that), event obviously doesn't have the latLng, but I'd like infowindow to show up in the middle of the polyline anyway, just like it does when you click on the list in the google maps link I mentioned before.
Tried LatLngBounds(); but that gives the actuall center of the area the polylines create, not the middle I need.
Any idea how to do it?

So this is the(bit hacky) solution.
Use http://www.geocodezip.com/scripts/v3_epoly.js library, then count the total length of you polyline(various ways), divide it in half and call epoly's .GetPointsAtDistance() function upon it.
This should return LatLng point, but it acts a bit weird sometimes, returning two points or even turning that point somehow "broken". So the most secure thing you can do is probably this:
var pointInHalf = polyline.GetPointsAtDistance(polylineLength);
var pointCoordinate = new google.maps.LatLng(pointInHalf[0].lat(), pointInHalf[0].lng());
Well, better than nothing.

From http://www.geocodezip.com/v3_polyline_example_geodesic_proj.html
Without extensions and assuming the polyline is a straight line.
It is possible to convert the lat/lng coordinates to point plane (x,y) postions and calculate the average between the two. This will give you a central pixel position. You can then convert this position back to a latlng for map plotting.
var startLatLng = startMarker.getPosition();
var endLatLng = endMarker.getPosition();
var startPoint = projection.fromLatLngToPoint(startLatLng);
var endPoint = projection.fromLatLngToPoint(endLatLng);
// Average
var midPoint = new google.maps.Point(
(startPoint.x + endPoint.x) / 2,
(startPoint.y + endPoint.y) / 2);
// Unproject
var midLatLng = projection.fromPointToLatLng(midPoint);
var midMarker = createMarker(midLatLng, "text");
More information on changing the projection http://code.google.com/apis/maps/documentation/javascript/reference.html#Projection

So firstly you need to use the geometry library which calculates distances. Add libraries=geometry to your JS call, e.g.
<script type="text/javascript" src="http://maps.googleapis.com/maps/api/js?libraries=geometry"></script>
Assuming you know the start point and end point for your polyline, you should be able to do this:
var inBetween = google.maps.geometry.spherical.interpolate(startLatlng, endLatlng, 0.5);
infowindow.position = inBetween;
I guess if you don't already know the start and end points, you could work it out from polyline.getPath().

to get the coordinates of your polyline you should do:
var widePath = new google.maps.Polyline({
path: waypointsCoordinates,
strokeColor: '#3366FF',
strokeOpacity: 0.0,
editable: true,
draggable: true,
strokeWeight: 3
});
and do:
var latLng [];
latLng = widePath.getPath().getArray();

Might be a bit old as well, but why not add the infobox on the click?
infowindow.setPosition(event.latLng);
infowindow.open(this.getMap());
If it's a click that is.

Setting a Google Maps viewport to automatically fit pane for locations of (n) markers of various locations

The approach I took thus far has been:
function addMarker( query ) {
var geocoder = new google.maps.Geocoder();
var afterGeocode = $.Deferred();
// Geocode 'query' which is the address of a location.
geocoder.geocode(
{ address: query },
function( results, status ){
if( status === 'OK' ){
afterGeocode.resolve( results ); // Activate deferred.
}
}
);
afterGeocode.then( function( results ){
var mOptions = {
position: results[0].geometry.location,
map: map
}
// Create and drop in marker.
var marker = new google.maps.Marker( mOptions );
marker.setAnimation( google.maps.Animation.DROP );
var current_bounds = map.getBounds(); // Get current bounds of map
// use the extend() function of the latlngbounds object
// to incorporate the location of the marker
var new_bounds = current_bounds.extend( results[0].geometry.location );
map.fitBounds( new_bounds ); // fit the map to those bounds
});
}
The problem I'm running into is that the map inexplicably zooms out by some amount, no matter if the new marker fits within the current viewport or not.
What am I doing wrong?
ADDENDUM
I added logs and an additional variable to capture the map bounds after the transition was made (new_new_bounds)
current_bounds = // Map bounds before anything is done.
{-112.39575760000002, 33.60691883366427},
{-112.39295444655761, 33.639099}
new_bounds = // From after the extend
{-112.39295444655761, 33.60691883366427},
{-112.39575760000002, 33.639099}
new_new_bounds = // From after the fitbounds
{-112.33942438265382, 33.588697452015374},
{-112.44928766390382, 33.657309727063996}

OK, so after much wrangling, it turns out that the problem was a map's bounds are not the same as a map's bounds after fitBounds(). What happens (I presume), is Google takes the bounds you give it in the fitBounds() method, and then pads them. Every time you send the current bounds to fitBounds(), You're not going to fit bounds(x,y), you're going to fit bounds(x+m,y+m) where m = the arbitrary margin.
That said, the best approach was this:
var current_bounds = map.getBounds();
var marker_pos = marker.getPosition();
if( !current_bounds.contains( marker_pos ) ){
var new_bounds = current_bounds.extend( marker_pos );
map.fitBounds( new_bounds );
}
So, the map will only fit bounds if a marker placed falls outside the current map bounds. Hope this helps anyone else who hits this problem.

A possible explanation is that you randomly placed your new marker into the gap of the z-curve. A z-curve recursivley subdivide the map into 4 smaller tiles but that's also the reason why there are gaps between the tiles. A better way would be to use a hilbert curve or a moore curve for map applications. There is a patented search algorithm covering this issue, I think it is called multidimensional range query in quadtrees. You want to look for Nick's hilbert curce quadtree spatial index blog.