Maps and scenes often combine multiple sources of information such as feature layers, map image layers, raster layers, and graphics. Labels and legends don't always provide enough information to work out what the map or scene is displaying. Use the identify methods to quickly answer the question: what is this item here? This will allow your users to explore and learn about the map or scene content by tapping or clicking it. Information returned can be shown in pop-ups or other UI components in your app.
You can identify the visible items at a specific point on screen, such as the following:
- Within all the layers in the map or scene, or only within a specific layer
- Within all the graphics overlays in the view or only within a specific graphics overlay
- Returning only the topmost item or all the items at that location
- Returning the feature, graphic, raster cell value, or other item at that location, or by returning pop-ups for pop-up-enabled layers
Identify methods are asynchronous, so that the UI thread of your application is not blocked waiting for results. This is especially important when working with data that resides on a server, as results do not return immediately. The results of an identify take into account the following:
- Symbology—Tapping on a large marker symbol used to draw a point, or a wide line symbol used to draw a polyline or outline of a polygon will include those items in the results; it's not just the geometry that is used.
- Visibility—The visibility of a graphics overlay or layer, and of an individual feature or graphic is checked, and results will only include the visible items. The opacity of a layer or graphic is ignored.
- Visible extent—Only items within the currently visible map or scene extent are included in identify results.
Identify is supported on feature layers, map image layers, Web Map Service (WMS) layers, graphics overlays, scene layers, raster layers, and also on tiled layers that are based on map services that support identify.
If time-based filtering is being used (a time extent has been set on the displaying map or scene view), identify will only return features that are within the time extent set on the geo view.
The sections below show you how to identify different items in the map or scene in different ways, but these approaches can be combined to provide general identify functionality if required.
Identify features in a feature layer
The steps below show you how to identify the features within a specific feature layer in the map or scene. Later in this topic, these steps are adapted to identify only the topmost feature, identify against multiple layers, different types of layers, and to identify graphics.
- Listen to a tap or click event on the map or scene view, and get the point representing the center of the tap or click.
- Call the required identify method, passing in the screen point from the previous step.
- Choose either to identify against a specific layer, or to identify against all layers.
- Specify a tolerance for the search radius of the identify operation. A tolerance of 0 identifies only items at the single pixel at the screen point. However, typically this level of precision is hard to achieve, so you can supply a tolerance around the screen point. A suitable tolerance for a tap operation on a touch screen should be equivalent to the size of the tip of a finger, and a smaller tolerance should be considered for mouse pointers.
- Specify whether to include pop-up information on the geo-elements returned. (If no pop-ups are defined for a layer or graphics overlay, pop-ups will not be included in the results, regardless of this parameter value.)
- Optionally, specify the maximum number of results per layer to return. This may be especially useful if identifying on service layers, as you can limit the amount of information returned to the client (the maximum results will also be limited by the server).
- Results consist of layer content information about the layer the results are from, and a list of AGSGeoElement. Each AGSGeoElement represents a feature identified at the given screen point. Iterate the results, access the geometry and attributes of the identified features, and use them as required. In the example below, each identified feature is selected.
Features are loadable, but when using identify methods, they are always returned already loaded.
Identify features in all feature layers
When you do not know which specific layer to identify on, you can identify items in any layer in the map or scene. Change the first workflow above so that you do not specify the layer to identify. This time, the results are returned as a list of AGSIdentifyLayerResult instead of a single result.
Layers that do not support identify or do not have any results based on the inputs are not included in the returned list.
Identify topmost item only
To identify only the topmost item at the screen point (or the topmost item per layer if identifying against all layers), change the first workflow above by removing the maximum number of results parameter. AGSGeoElements are still returned as a list, but the list will have only zero or one items.
Identify on group layers
The results returned when any group layers are in the map or scene depends on the identify operation you use.
When identifying on the map or scene using identifyLayersAtScreenPoint (plural layers), identification operates on all the child layers of any group layers independently of the group layer. There is no result for the group layer itself, only for the child layers. Effectively, this behavior treats the child layers as independent layers and otherwise ignores the group layers.
When identifying specifically on the group layer using identifyLayer (singular layer), identification operates on all the child layers of the group layer. The result is returned in an AGSIdentifyLayerResult object for the group layer. AGSIdentifyLayerResult.geoElements will not contain any results for the group layer itself, so instead, access results for each child layer through AGSIdentifyLayerResult.sublayerResults. This returns a list of AGSIdentifyLayerResult objects for each child layer with the elements identified in each layer.
Identify on map image layers
Identifying against map image layers and tiled map layers follows the same workflow as shown for feature layers. The following are differences when identifying against map image layers:
- Results are returned as AGSFeatures; unlike other features, however, they will not have a reference to a AGSFeatureTable.
- Map image layers may have one or more sublayers—identify results from map image layers reflect this structure, and return results for each sublayer separately. (Note that if you have specified a maximum number of results to return, this value applies per sublayer.)
Return pop-ups as results
Some types of layers support pop-ups. You can use an identify method to return pop-ups for map content at a screen point.
Identify features in a WMS layer
WMS layers differ from other layers, as they do not support returning individual attributes or geometry for a feature. WMS services perform identification on the server and return HTML documents describing identified features. You can access the returned HTML document string through the "HTML" entry in the feature's attributes. This HTML string is suitable for display in a web view.
It is impossible to get the geometry for an identified (or any other) WMS feature. An identified WMS feature's geometry will always be null. Consequently, WMS layers do not support feature selection/highlight.
Graphics are identified using methods different than those used for layers. You can choose to return graphics in a specific graphic overlay or for all graphics overlays; you can also limit results to only the topmost graphic in each graphics overlay.
Identify on raster layers
Identify on raster layers returns the AGSRasterCell value for a tapped location in an AGSMapView or AGSSceneView. The identified AGSRasterLayer can be local on the device or from a web service layer hosted on ArcGIS Online or ArcGIS Enterprise portals. Identify returns raster cell values to display in a simple callout, or if a pop-up is configured for the raster layer, the information can be displayed in a formatted UI.
For mosaicked images, a mosaic rule defines how the individual rasters are combined. When identifying images mosaicked from a collection of images, the values returned from identify can vary according to the mosaic rule settings. You can use the default rule defined with the service or, starting with 100.9.0, define rule settings to control how overlapping areas in the mosaic are handled. Rendering rules applied on the raster layer on the client, as well as information from attribute tables (if present), will also be represented in the identify results.