Develop Reference

Setting multiple toggleable KML layers on GoogleMaps with custom icons

I would like to create a Google map with toggleable KML layers - every layer has its set of icons. Example of a map I would like to use can be seen here.
I am unable to set separate layers on the map so I can continue with making them toggleable one by one.
These are my steps for obtaining data for KML layers on Google map:
1. Select "Download KML"
2. I chose a layer ("Food stores" for example)
- I checked "Keep data up to date with network link KML" (Map will be maintained on Google maps and data should automatically refresh in my map on the website).
- I left unchecked "Export to a KML file" (because of the icons support)
Then I unzip downloaded files. Inside small code there is <href>http://www.google.com/maps/d/kml?mid=16i6f7Jvm754_jzkltP-Ks_2pKbU&lid=M97Oy53L0t4</href> which I include in url of sitesLayer when passing data for function loadKml() at the bottom of the following code.
The problem with this is that I instantly get all those icons on the map, even if I have downloaded and included only 1 layer - the map shows all 4 of them.
Am I missing something, in the code or in the steps for downloading KML files for separating layers/icons?
function initialize() {
// Map settings
var myLatlng = new google.maps.LatLng(0,0);
var mapOptions = {
zoom: 18,
center: myLatlng
};
var map = new google.maps.Map(document.getElementById('map'), mapOptions);
// Creating KML layers
loadKml = function(opts,map){
var layer = new google.maps.KmlLayer();
opts.preserveViewport = true;
if(map) {
opts.map = map;
}
google.maps.event.addListener(layer,'defaultviewport_changed',function() {
var map=this.getMap(),
bounds=map.get('kmlBounds') || this.getDefaultViewport();
bounds.union(this.getDefaultViewport());
map.set('kmlBounds',bounds);
map.fitBounds(bounds);
});
layer.setOptions(opts);
return layer;
}
// Setting KML layers
var sitesLayer = loadKml({
url: 'http://www.google.com/maps/d/kml?mid=1AQ5Is7NjBGjPZpLji5fbApYnegk&lid=z-AEWAAfH9c',
map: map
});
var sitesLayer2 = loadKml({
url: 'http://www.google.com/maps/d/kml?mid=16i6f7Jvm754_jzkltP-Ks_2pKbU&lid=jY6tGqHtk80',
map: map
});
}
google.maps.event.addDomListener(window, 'load', initialize);

OpenLayers WMS layer doesn't load

I use the following block of JavaScript to try to show a WMS layer. I'm using OpenLayers 2.8.
The map's base layer (Openstreetmap) shows correctly, it zooms to the correct area, the "pyramid" layer is shown in the layer switcher, but no request to its WMS service is ever made (so the fact that the URL, styles and params are dummies shouldn't matter -- it never even attempts to get them).
OpenLayers does try to get a WMS layer once I pan or zoom far enough so that the Gulf of Guinea is in view (but all my data is in the Netherlands). This suggests a projection problem (WGS84's (0, 0) point is there), but I don't understand why OpenLayers doesn't even try to fetch a map layer elsewhere. My data is in EPSG:3857 (Web Mercator) projection.
/*global $, OpenLayers */
(function () {
"use strict";
$(function () {
$(".map").each(function () {
var div = $(this);
var data_bounds = div.attr("data-bounds");
console.log("data_bounds: " + data_bounds);
if (data_bounds !== "") {
var map = new OpenLayers.Map(div.attr("id"), {
projection: "EPSG:3857"});
var extent = JSON.parse(data_bounds);
var bounds = new OpenLayers.Bounds(
extent.minx, extent.miny,
extent.maxx, extent.maxy);
map.addLayer(
new OpenLayers.Layer.OSM(
"OpenStreetMap NL",
"http://tile.openstreetmap.nl/tiles/${z}/${x}/${y}.png",
{buffer: 0}));
map.addLayer(
new OpenLayers.Layer.WMS(
"pyramid", "http://rasterserver.local:5000/wms", {
layers: "test",
styles: "test"
}, {
singleTile: true,
isBaseLayer: false,
displayInLayerSwitcher: true,
units: 'm'
}));
map.addControl(new OpenLayers.Control.LayerSwitcher());
map.zoomToExtent(bounds);
}
});
});
})();
Edit: the 'data_bounds' console print prints (with some added formatting):
data_bounds: {
"minx": 582918.5701295201,
"miny": 6923595.841021758,
"maxx": 821926.9006116659,
"maxy": 7079960.166533174
}
It zooms to the correct region in the north of the Netherlands, so I don't think the problem is there.
Since posting, I found out that if I don't use the OSM layer, and instead use the WMS layer as baselayer, it works. So perhaps there's some incompatibility with a OSM baselayer and a WMS layer added to it? But then I don't get that it does seem to do something near WGS84 (0, 0).

I eventually managed to fix this by giving the map an explicit maxExtent:
var extent = JSON.parse(data_bounds);
var bounds = new OpenLayers.Bounds(
extent.minx, extent.miny,
extent.maxx, extent.maxy);
var map = new OpenLayers.Map(div.attr("id"), {
projection: "EPSG:3857",
maxExtent: bounds
});
Oddly enough this doesn't limit the user's ability to pan and zoom around the world, but it does make the overlay work...

Dynamic rectangles in Google Maps

First I'm pretty new to Javascript, so sorry if my question comes across poorly.
I'm creating an application in Flash to help users calculate their electrical costs. Then I'm taking this figure and write it to an xml file.
Now I'm looking to open a webpage and show a google map, and there is a rectangle drawn over the map which is generated dynamically from the number generated earlier and stored in the xml file.
I'm completely lost as to places to turn on how to achieve this. I've gotten my map on to my page, and it scales 100% as I want it to, but I can't figure out the dynamic rectangle part at all. Any ideas or pointers in the right direction greatly appreciated.

In this latest version, the XML file
<countries>
<country name="USA" lat="40.0" lng="-100.0" width="30.0"/>
<country name="France" lat="46.6" lng="2.7" width="10"/>
<country name="Germany" lat="51.1" lng="10.1" width="20"/>
</countries>
is loaded as soon as the map tiles finish loading. I could not get the getProjection to be called correctly if I did not wait for tile loading to finish. The docs state that getting the projection needs the map to be initialized, and recommends listening for projection_changed. Both ways work yet I still feel listening to tiles_loaded is safer and if something goes wrong with the xml loading it will get called again if the map is zoomed or panned a noticeable amount.
var map;
var xmlLoaded = false;
function initialize() {
var mapOptions = { center: new google.maps.LatLng(30.0, 0.0), zoom: 2,
mapTypeId: google.maps.MapTypeId.ROADMAP };
map = new google.maps.Map(document.getElementById("map_canvas"), mapOptions);
google.maps.event.addListener(map, 'tilesloaded', loadData);
}
function loadData() {
if(!xmlLoaded) {
$.ajax({
type: "GET",
url: "co2data.xml",
dataType: "xml",
success: function(xml) {
var countries = xml.documentElement.getElementsByTagName("country");
for(var i = 0, country; country = countries[i]; i++) {
var name = country.getAttribute("name");
var lat = parseFloat(country.getAttribute("lat"));
var lng = parseFloat(country.getAttribute("lng"));
var point = map.getProjection().fromLatLngToPoint(new google.maps.LatLng(lat,lng));
// width is really an arbitrary unit, relative to CO2 tonnage.
// equals the side of the drawn square.
// it is measured in google maps points units.
var width = parseFloat(country.getAttribute("width"));
makeCO2Rect(name, point, width);
}
xmlLoaded = true;
}
});
}
}
The rectangle is defined by width in points (the whole world is 256x256 points), so some conversion is needed when assigning their centers to the more conventional LatLng.
function rectParamsToBounds(point, width) {
var ctrX = point.x;
var ctrY = point.y;
var swX = ctrX - (width/2);
var swY = ctrY - (width/2);
var neX = ctrX + (width/2);
var neY = ctrY + (width/2);
return new google.maps.LatLngBounds(
map.getProjection().fromPointToLatLng(new google.maps.Point(swX, swY)),
map.getProjection().fromPointToLatLng(new google.maps.Point(neX, neY)));
}
Finally, a rectangle is created with a country name that goes into a MarkerWithLabel (using v1.1.5 here, you can hotlink to http://google-maps-utility-library-v3.googlecode.com/svn/tags/markerwithlabel/1.1.5/src/markerwithlabel_packed.js though I prefer saving a local copy)
Since dragging a rectangle appears impossible, a marker in its center works as a handle. When it's dragged, the associated rectangle moves with it.
function makeCO2Rect(name, point, width) {
var rect = new google.maps.Rectangle({
map: map,
bounds: rectParamsToBounds(point, width)
});
var marker = new MarkerWithLabel({
map: map,
position: map.getProjection().fromPointToLatLng(new google.maps.Point(point.x, point.y)),
draggable: true,
raiseOnDrag: false,
labelContent: name,
labelAnchor: new google.maps.Point(30, 0),
labelClass: "labels", // the CSS class for the label
labelStyle: {opacity: 1.0}
});
google.maps.event.addListener(marker, 'drag', function(event) {
var newLatLng = event.latLng;
var newPoint = map.getProjection().fromLatLngToPoint(newLatLng);
rect.setBounds(rectParamsToBounds(newPoint, width));
});
}
google.maps.event.addDomListener(window, 'load', initialize);
Styling the labels need to be done both in the .labels CSS class and the constructor, and rectangles have options like stroke color, thickness, opacity, and fill color.

If you just want to place a rectangular shape on the map, you can create a google.maps.Rectangleapi-doc. If you want to create a rectangular label on the map, you may be more interested in the InfoBox Utility Librarywiki-page.

Adding Event Listener causes Google Map to freeze

I am attempting to create a google map with markers on my page. I have an unordered list where each item has data attributes for latitude, longitude and title. Using JQuery I pull these values and produce a marker on the map for each item in the list. Everything seems to work OK except google maps will not load tiles as you pan around the map.
This is how I initialize the map:
var map;
// initialize google map
$(function () {
var myOptions = {
zoom: 10,
center: new google.maps.LatLng(CoordinatesLat, CoordinatesLong),
mapTypeId: google.maps.MapTypeId.ROADMAP,
mapTypeControl: false
}
// initiate map
map = new google.maps.Map($("#map")[0], myOptions);
// when map is loaded, add events and behaviors to it
google.maps.event.addListenerOnce(map, 'tilesloaded', addEventsToMap(".event")); //commenting this line prevents GMAP problems
});
FYI - before I was using maps.event.addListener() and the map would work momentarily and then become completely unresponsive. Changing it to maps.event.addListenerOnce() stopped the freezing but still has the tile loading problem.
And this is the callback, where evidently I've done something wrong:
//add events from event list to map. add behavior to events ie. mouseover
function addEventsToMap(selector) {
$(selector).each(function (i, $e) {
$e = $($e);
var latlng;
if ($e.attr("data-geo-lat") && $e.attr("data-geo-long")) {
latlng = new google.maps.LatLng($e.attr("data-geo-lat"), $e.attr("data-geo-long"));
$e.marker = new google.maps.Marker({
position: latlng,
map: map,
title: $e.attr("data-title")
});
google.maps.event.addListener($e.marker, 'click', function () { window.location = $e.attr("data-href"); });
google.maps.event.addListener($e.marker, 'mouseover', function () { $e.addClass("event-hover"); });
google.maps.event.addListener($e.marker, 'mouseout', function () { $e.removeClass("event-hover"); });
//when mouse hovers over item in list, center map on it's marker
$e.mouseenter(function () {
map.panTo(latlng);
});
}
});
}
Any idea what could be causing this?

I see a problem, though I'm not sure if it's the issue here. If you examine this line:
google.maps.event.addListenerOnce(map, 'tilesloaded', addEventsToMap(".event"));
What you are intending to do is call your 'addEventsToMap' once the map is loaded, but instead what you are doing is calling the function and then assigning the return value of that as a listener.
You need this instead, and I think it shouldn't crash anymore.
google.maps.event.addListenerOnce(map, 'tilesloaded', function() {
addEventsToMap(".event");
});
If that doesn't work, I would recommend looking at the Javascript Console of whatever browser you are using, to see what error is happening to make the map crash.

Retrieve position of a google maps v3 marker to Qt in a desktop app with QtWebKit

I'm building a Qt app with Python where you can point and click at a (google) map and get the coordinates of the location. The map is shown through a QWebView loading a simple HTML page and the user can create markers by clicking. Screenshot of the widget after clicking on the map.
However, I'm having trouble to retrieve the just-clicked location coordinates back to Qt (so that I can use them as variables -- and, for example, show them in the QLineEdits on the topleft corner above, as current location of the marker).
This is the relevant part of the HTML file:
<script type="text/javascript">
var map;
function initialize() {
var local = new google.maps.LatLng(-23.4,-40.3);
var myOptions = {
zoom: 5,
center: local,
mapTypeId: google.maps.MapTypeId.ROADMAP
}
map = new google.maps.Map(document.getElementById("map_canvas"), myOptions);
google.maps.event.addListener(map, 'rightclick', function(event) {
placeMarker(event.latLng);
});
}
function placeMarker(location) {
var clickedLocation = new google.maps.LatLng(location);
var marker = new google.maps.Marker({
position: location,
map: map
});
map.setCenter(location);
}
function dummyTxt() {
return 'This works.';
}
</script>
I've been trying with evaluateJavaScript, but was not able to retrieve the coordinates. I tried to created a function to access the position with marker.getPosition(), but with no luck. The dummy below works though..
newplace = QWebView.page().mainFrame().evaluateJavaScript(QString('dummyTxt()'))
>>> print newplace.toString()
This works.
Any suggestions on how to get the coordinates back to Qt?
Edit:
Here is the code that worked for me:
def update_geo(self):
# Capture coordinates of the last marker on the map.
mark = self.map.page().mainFrame().evaluateJavaScript('document.getElementById("markerlocation").value').toString()
# Convert string to list of floats, stripping parentheses.
marker = str(mark).strip('()').split(', ')
decimals = [float(c) for c in marker]
Full source in https://github.com/nelas/veliger/blob/master/veliger.py#L2374

I found a work around to make it work but I'm not pretty sure that it will be the right approach. Anyway, this is what I did:
Create a hidden input in the body section of your html document to save the position data of the marker:
<body>
(...)
<input id="locationData" type="hidden">
(...)
</body>
In the javascript code, save the position of the marker in the hidden input every time it's created:
function placeMarker(location) {
(...)
document.getElementById("locationData").value = marker.position;
(...)
}
In your Qt code, read the value of the hidden input with the instruction:
webView->page()->mainFrame()->findFirstElement("#locationData").evaluateJavaScript("this.value").toString();
I hope it helps!
Source: http://opendocs.net/qt/4.6/webkit-formextractor.html