Sounds like a bug/typo in the sample. The default value for layerOption is that only the top-most visible layer is identified. So the sample should probably specify the visible layers and set the layerOption to LAYER_OPTION_VISIBLE instead of all.
Sounds like a bug/typo in the sample. The default value for layerOption is that only the top-most visible layer is identified. So the sample should probably specify the visible layers and set the layerOption to LAYER_OPTION_VISIBLE instead of all.
I'm still not clear as to whether both parameters should be included or not. The API says that by specifying layerids then only those layers will be identified. It doesn't say anything about whether the layers are visible or not. Whereas the layer_option parameter is about visibility.
I'm still not clear as to whether both parameters should be included or not. The API says that by specifying layerids then only those layers will be identified. It doesn't say anything about whether the layers are visible or not. Whereas the layer_option parameter is about visibility.
I discovered that it does help to use both the layerids and layer_option_all. In my particular application I was identifying two featurelayers from one mapservice. An identify operation was only returning the results from one layer even though I specified both layers in identifyParams.layerIds = [3,4] and I knew that it should be picking up a feature from the second layer; When I added identifyParams.layerOption = esri.tasks.IdentifyParameters.LAYER_OPTION_ALL; the results from both layers came back.
So, the sample from the original posting may be correct, and all that might be needed is a clearer explanation of how these two parameters work together.
