You are quite right, the pixelBounds object is the one you want. However, even though it is mentioned in the documentation, it is not documented and the objects property names are not fixed.
I have opened an issue to try to get this documented. If you add a comment, the google maps team might document it - making life easier for us. You can find it here - http://code.google.com/p/gmaps-api-issues/issues/detail?id=2860 - Simon On Nov 20, 9:34 am, Peter van der Zee <qfo...@gmail.com> wrote: > Hm, that is interesting indeed. However, that would mean that I'd need > an event fired for every marker. But maybe there's another way of > getting to that object/information... It's weird that they (obviously) > have the information but don't expose it... Thanks! > > On Nov 19, 11:47 pm, CroNiX <cronix...@gmail.com> wrote: > > > This is an undocumented marker property I was just playing with > > (pixelBounds), but it seems to be what you want...but its > > undocumented...so unofficial...ugh. But the marker is supplying the > > correct properties but no "official" method to retrieve them. > > > google.maps.event.addListener(marker, 'click', function() { > > var pb = this.pixelBounds; > > var iconX = Math.abs(pb.l) + Math.abs(pb.q); > > var iconY = Math.abs(pb.n) + Math.abs(pb.o); > > alert('Width: ' + iconX + ', Height: ' + iconY); > > > }); > > > On Nov 19, 2:18 pm, CroNiX <cronix...@gmail.com> wrote: > > > > Is it the pixel coordinates you are after for the anchor location? > > > This might help. > > > > var containerHelper = null;//set in global space > > > > //call this somewhere in your map init function to create an overlay > > > that you can use to translate the coordinates > > > function init_helper_container(){ > > > containerHelper = null; > > > containerHelper = new google.maps.OverlayView(); > > > containerHelper.setMap(map); > > > containerHelper.draw = function () { > > > if (!this.ready) { > > > this.ready = true; > > > google.maps.event.trigger(this, 'ready'); > > > } > > > }; > > > > } > > > > ... > > > > google.maps.event.addListener(marker, 'click', function() { > > > var markerLoc = > > > containerHelper.getProjection().fromLatLngToContainerPixel(this.getPosition()); > > > alert('x=' + markerLoc.x + ', y=' + markerLoc.y); > > > > }); > > > > On Nov 19, 12:26 pm, Peter van derZee<qfo...@gmail.com> wrote: > > > > > I'm trying to port my marker cluster script (http://cm.qfox.nl) from > > > > v2 to v3 but it seems that's pretty much impossible because the api > > > > doesn't expose enough details. > > > > > The biggest problem seems to be getting the actual size of the marker > > > > and the anchor of the marker. I've searched the group, but can't > > > > really find related posts (which was surprising to me). > > > > > Basically I need to know precisely what area of the map is covered by > > > > a given marker in order to compare these regions to each other. > > > > However, there are is no api to give me this information. > > > > > So this is what I have right now... > > > > > An icon can be set by .setIcon* (or implicitly through the > > > > constructor, but I'm going to assume both ways lead to the same > > > > result). This method accepts a string or a MarkerImage*. The .getIcon > > > > method of a marker returns whatever you give it, it is not normalized > > > > (quite annoying). The .getShape method is not set implicitly, so it's > > > > useless in this case. > > > > > Ok, so .getIcon returns a string (the url of the icon) or the > > > > MarkerImage. If it's the url, you can load it in an Image and get the > > > > width/height after it's loaded. Easy enough. > > > > > If it returns the MarkerImage, things get hairy. That object should > > > > have a coords property. I checked, it's not set. It does have five > > > > properties. Four of them are scrambled by the minifiers, but their > > > > values clearly show that they are the properties as given in the > > > > constructor. The other one is obviously anchor, which is undocumented > > > > but also unscrambled (which is hardly coincidental). Maybe it's just > > > > missing from the documentation, who knows. Of the five properties, one > > > > is type string. Two are points and two are sizes. Since anchor is a > > > > point, you can also infer the origin, which is "point"less. > > > > > So since I can get at least the url from MarkerImage, albeit through a > > > > hack, I can reasonably reliably get the size of the image used for the > > > > marker. It's not the end-goal, but at least it's something. > > > > > So my question is, am I overlooking anything here? I'd like to get the > > > > anchor and final size of the image used for a marker when a > > > > MarkerImage is supplied for an icon, rather than a regular string. As > > > > it stands now, that seems impossible. Any help? > > > > > But as it stands it seems impossible to create a generic cluster > > > > manager. You can make it safe for when .getIcon returns a string, but > > > > when it doesn't you're pretty much screwed. > > > > > * > > > > Marker:http://code.google.com/apis/maps/documentation/javascript/reference.h... > > > > * > > > > MarkerImage:http://code.google.com/apis/maps/documentation/javascript/reference.h... -- You received this message because you are subscribed to the Google Groups "Google Maps JavaScript API v3" group. To post to this group, send email to google-maps-js-api...@googlegroups.com. To unsubscribe from this group, send email to google-maps-js-api-v3+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/google-maps-js-api-v3?hl=en.
