GMUClusterRendererDelegate willRenderMarker called twice for each marker

[SAPIENTechnologies]

I have created the GMUClusterRendererDelegate in my code and implemented the willrenderMarker function. I notice that this function loops through the visible markers 2x for each marker. For instance, if I have 5 visible markers, the function gets called for each marker in succession, and then AGAIN for each marker in succession. Is this expected behavior?

[mountainvat]

The expected behaviour is willRenderMarker should be called only once for each marker being added to the map. However whenever a recluster needs to be done (zooming in/out far enough from current zoom) the markers will be recreated and willRenderMarker will be called again. Was that what happens? We will try to reproduce on our side too.

[mountainvat]

FYI: I've just briefly tried to reproduce on my side but failed. So I am waiting for more information.

[SAPIENTechnologies]

Its very odd. Here is the only place where I call cluster()

func clusterMarkers () {
    //clear out the marker array
    self.markers.removeAll()

    //clear out the cluster manager items
    mapViewController.clusterManager.clearItems()

    self.clusterManager.add(trucks)

    // Call cluster() after items have been added to perform the clustering
    // and rendering on map.
    mapViewController.clusterManager.cluster()
    }

I call clusterMarkers from other places but I know its only called once because of the self.markers.removeAll() call. When I look at the self.markers array in my willRendermarker delegate function, it does not get cleared out in between the duplicate render calls. In other words:

I start out with the markers array empty
I start out with 5 items in my trucks array.
I add the trucks array to cluster manager
I call cluster()

in my willRenderMarker delegate function, it renders each marker which I then add to the markers array. But then it renders each marker again, invalidating the first set of markers. But I end up with 10 markers in my markers array. The first 5 are now invalid on the map.

I have put in some code to get past the problem in the WillRenderMarker delegate function, but I can’t seem to figure out why the markers are being rendered a second time.

Thank you for any input.

Regards,
info@sapien.com

On Nov 17, 2016, at 1:25 PM, Son Nguyen notifications@github.com wrote:

The expected behaviour is willRenderMarker should be called only once for each marker being added to the map. However whenever a recluster needs to be done (zooming in/out far enough from current zoom) the markers will be recreated and willRenderMarker will be called again. Was that what happens? We will try to reproduce on our side too.

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub #58 (comment), or mute the thread https://github.com/notifications/unsubscribe-auth/AOoyM7ZU3IfrKsj8JKVtaf1FXFcQmS2Vks5q_MZdgaJpZM4K1n7T.

[mountainvat]

So in the willRenderMarker handler, can you log the map's camera zoom out and paste them here?

[SAPIENTechnologies]

Its always the same:

16.0
16.0
16.0
16.0
16.0
16.0
16.0
16.0
16.0
16.0
16.0
16.0

Regards,
info@sapien.com

On Nov 17, 2016, at 2:20 PM, Son Nguyen notifications@github.com wrote:

So in the willRenderMarker handler, can you log the map's camera zoom out and paste them here?

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub #58 (comment), or mute the thread https://github.com/notifications/unsubscribe-auth/AOoyM2tK6xkn8j4JWvSfzoLaOl00lmgxks5q_NMtgaJpZM4K1n7T.

[mountainvat]

There were 12 events in the log above. So sth strange here. Can you also log the camera lat,lng to see if they actually moved?

[SAPIENTechnologies]

There are 4 trucks in my viewport here is the output

ID: 2
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 3
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 5
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 6
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 2
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 3
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 5
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 6
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

Regards,
info@sapien.com

On Nov 17, 2016, at 2:52 PM, Son Nguyen notifications@github.com wrote:

There were 12 events in the log above. So sth strange here. Can you also log the camera lat,lng to see if they actually moved?

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub #58 (comment), or mute the thread https://github.com/notifications/unsubscribe-auth/AOoyMy5OO8K5aUBSjAkKDubTjK5nZ-1_ks5q_NqzgaJpZM4K1n7T.

[SAPIENTechnologies]

Anything new based on the last logging I sent?

here it is again

There are 4 trucks in my viewport here is the output

ID: 2
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 3
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 5
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 6
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 2
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 3
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 5
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

ID: 6
Zoom level: 16.0
Latitude: 38.2556198811391
Longitude: -122.27283205837

Regards,
info@sapien.com

On Nov 17, 2016, at 2:52 PM, Son Nguyen notifications@github.com wrote:

There were 12 events in the log above. So sth strange here. Can you also log the camera lat,lng to see if they actually moved?

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub #58 (comment), or mute the thread https://github.com/notifications/unsubscribe-auth/AOoyMy5OO8K5aUBSjAkKDubTjK5nZ-1_ks5q_NqzgaJpZM4K1n7T.

[mountainvat]

The camera did not change throughout so I am not sure what's going on there. Are you able to check out the source, open the workspace/GoogleMapsUtils.xcodeproj and run the DemoApp target and see if it reproduce for the CustomMarkerViewController demo?

I was unable to reproduce using that demo.

Or else can you create a small repro project that I can take a look?

[SAPIENTechnologies]

This problem still exists. I have created a stripped down version of my project. Is there a private location that I can upload it to for you to take a look at?

[domesticmouse]

Please feel free to email me the repro, or link to it, at brettmorgan@google.com

[domesticmouse]

I'm a tad confused, at first blush this app doesn't appear to include Google-Maps-iOS-Utils in the Podfile. It includes a bunch of other detail that isn't helpful with respect to debugging the issue.

I strongly suggest you build a minimal reproduction sample that includes only the GoogleMaps and Google-Maps-iOS-Utils pods, and post it up as a github repo that I can clone and fix.

Thanks!

[SAPIENTechnologies]

Thats because Google-Maps-iOS-Utils is included as source. See the attached image. When I first incorporated that lib, it was the suggested way to do it. Is it now available as a pod? This is pretty much the mosts stripped I can get the app and have it be a real representation of the app. What I have found yesterday is that the size of the marker.userData heavily influences the speed of the clustering. I created a parallel array of “liteTrucks” which include only a truckIdentifier and a position and passed that array to the clusterManager. This increased the clustering speed tremendously. I am still having issues with willRenderMarker being called multiple times for the same marker though. Regards, info@sapien.com

…

On May 25, 2017, at 5:07 PM, Brett Morgan ***@***.***> wrote: I'm a tad confused, at first blush this app doesn't appear to include Google-Maps-iOS-Utils in the Podfile. It includes a bunch of other detail that isn't helpful with respect to debugging the issue. I strongly suggest you build a minimal reproduction sample that includes only the GoogleMaps and Google-Maps-iOS-Utils pods, and post it up as a github repo that I can clone and fix. Thanks! — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#58 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AOoyM4PGtsDwCWaNK02lv_fFeoPybbcIks5r9hengaJpZM4K1n7T>.

[SAPIENTechnologies]

Here is the reason I don’t use the podfile:… https://github.com/googlemaps/google-maps-ios-utils/blob/master/Swift.md Regards, info@sapien.com

…

On May 25, 2017, at 5:07 PM, Brett Morgan ***@***.***> wrote: I'm a tad confused, at first blush this app doesn't appear to include Google-Maps-iOS-Utils in the Podfile. It includes a bunch of other detail that isn't helpful with respect to debugging the issue. I strongly suggest you build a minimal reproduction sample that includes only the GoogleMaps and Google-Maps-iOS-Utils pods, and post it up as a github repo that I can clone and fix. Thanks! — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#58 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AOoyM4PGtsDwCWaNK02lv_fFeoPybbcIks5r9hengaJpZM4K1n7T>.

[domesticmouse]

The instructions in https://github.com/googlemaps/google-maps-ios-utils/blob/master/Swift.md are for the specific case for when you have a Podfile that has use_frameworks! declared in it. From memory your reproduction instance didn't have that in it. It also had a lot of other stuff. I like to get people to boil down their issues to a minimal reproduction, as that exercise frequently shows the problem directly, or failing that, limits the search I have to make to figure out the problem.

So, please create a minimal repro case of the issue you are facing. Thanks!

[PatoSalazarNascent]

Also have the same issue. Could not use the podfile as my app is using use_frameworks!. Will Render is getting called twice for each marker.

For one single marker and a cluster marker (of six clusterItems) if I print inside the willRender method I get

render
render
render
render

also calling cluster only once when a promise resolves.

    private func getHotpotsLocations() {
        hotspotProtocol.getItemsByBounds(bounds: getVisibleBounds())
            .then {
                [weak self = self]
                clusterItems -> Void in
                
                for item in clusterItems {
                    
                    if WifiHotspotsViewController.isCoordinateInvalid(coord: item.position) {
                        print("MAPKIT NAME ===> \(item.name)")
                        print("MAPKIT ADDRESS ===> \(item.address)")
                        print("MAPKIT LATITUDE ===> \(item.position.latitude)")
                        print("MAPKIT LONGITUDE ===> \(item.position.longitude)")
                        
                        continue
                    }
                    
                    self?.clusterManager.add(item)
                }
                
                self?.clusterManager.cluster()
            }
            .catch {
                error in
                
                print(error)
        }
    }

This is what is inside my willRender method

    internal func renderer(_ renderer: GMUClusterRenderer, willRenderMarker marker: GMSMarker) {
        print("render")
        if let clusterData = marker.userData as? GMUCluster {
            marker.iconView = UIImageView(image: getClusterIcon(itemsCount: clusterData.count))
        }
        else {
            marker.iconView = UIImageView(image: UIImage(named: "singlePin"))
        }
    }

[frios]

Do you call clusterManager.clearItems() sometime before this? I was looking at this call and saw that it calls clusterManager.algorithm.clearItems() then calls requestCluster(). That seemed to be the issue for me that created the duplicate willRender calls.

I fixed it by calling clusterManager.algorithm.clearItems() directly, bypassing the call to requestCluster().

[stale]

This issue has been automatically marked as stale because it has not had recent activity. Please comment here if it is still valid so that we can reprioritize. Thank you!

[stale]

Closing this. Please reopen if you believe it should be addressed. Thank you for your contribution.