laptop and a wrench

Bug

Running arcpy.management.SelectLayerByAttribute through multiple iterations, when the input is a feature class or table, results in each iteration being slower than the previous iteration.

Last Published: February 22, 2023 ArcGIS Pro
Bug ID Number BUG-000153589
SubmittedNovember 9, 2022
Last ModifiedOctober 7, 2024
Applies toArcGIS Pro
Version found3.0.2
Operating SystemWindows OS
Operating System Version10.0 64 Bit
StatusAs Designed

Additional Information

Using the dataset instead of a layer as an input to the SelectBy tools is for convenience reasons. But it incurs some performance problems, as captured and explained in this issue. If performance is a concern, see the workaround. When a dataset is passed as an input to the SelectBy tools, the first step is to generate a unique layer name. The heuristic is to create a layer based on the input name and if the layer exists, add an incrementing numeric value. So if the input dataset is 'city.shp', the tool generates 'cityLayer' in the first loop. The second loop also generates 'cityLayer' but as this returns 'True' to the existence check, the name of the layer is renamed to 'cityLayer1'. The third loop hits two name collisions before settling on 'cityLayer2',  and so on. By the 100th iteration, we have 99 existence checks returning 'True', allowing to settle on 'cityLayer99'. The slowdown in the later parts of the iteration is due to this.

Workaround

Use arcpy.management.MakeFeatureLayer to create a feature layer to be used as an input for arcpy.management.SelectLayerByAttribute. Refer to the following ArcMap documentation for more information: https://desktop.arcgis.com/en/arcmap/latest/tools/data-management-toolbox/select-layer-by-attribute.htm

Steps to Reproduce

Bug ID: BUG-000153589

Software:

  • ArcGIS Pro

Get notified when the status of a bug changes

Download the Esri Support App

Discover more on this topic

Get help from ArcGIS experts

Contact technical support

Download the Esri Support App

Go to download options