Create Arcade expression for ArcGIS Dashboard Pie Chart widget

Caitlin_Todd_LCOR

Hello, 

I'm a beginner in all things Arcade and I'm having some trouble with an Arcade expression for ArcGIS Dashboard. I have a table of names that I will be pulling into a Pie Chart widget to show all repeated names as categories, so if John Doe is repeated 5 times and Jane Doe is repeated 3 times, then John Doe gets a bigger slice of the pie chart. But the widget has a max of 300 categories and my table has over 400 records, so I need to filter the data down to only the repeated names before it tries to render in the Pie Chart. Which leads me to trying to create a data expression. 

The part I'm stuck on is figuring out what function to use to identify repeat values in a field. Online research shows quite a bit about how to find unique values, but I really need the opposite of that. 

So far I have, 

 

 

var portal = Portal('https://someportal.com/')
var tablefs = FeatureSetByPortalItem(
    portal,
    'tableportalitemid',
    1,
    ['*'],
    false
);
// Create empty dictionary of all fields you want the pie chart to reference  
var dict = { 
  fields: [ 
    { name: "wholename", type: "esriFieldTypeString" }, 
    { name: "GlobalID", type: "esriFieldTypeGlobalID" },  
  ], 
  geometryType: "", 
  features: [], 
};

 

 

 

I think the rest of the logic would be something like;

//Create a list of all values in tablefs wholename field

//Where wholenames are not distinct, store into an array.

//Create more logic to loop through the tablefs and if wholename value exists in array then push all matching features into dictionary 

//Finally export dictionary to featureset to reference in Pie Chart Widget.

Thanks very much for any insight and advice, 

There may be a better way..

JohannesLindner

So, this is a case of me not knowing how Dashboard pie charts work... I thought you would supply a numeric and a label column, in which case the GroupBy would have sufficed. But you only supply one column and ArcGIS groups those values on its own.

Reading your question again, you correctly outlined the general process, although you can do it without using an extra dictionary:

// load your feature set
var dict = { 
  fields: [ 
    { name: "wholename", type: "esriFieldTypeString" }
  ], 
  geometryType: "", 
  features: [
      {"attributes": {"wholename": "abc"}},
      {"attributes": {"wholename": "abc"}},
      {"attributes": {"wholename": "def"}}, // singular name
      {"attributes": {"wholename": "ghi"}},
      {"attributes": {"wholename": "ghi"}},
      {"attributes": {"wholename": "ghi"}},
      ], 
};
var tablefs = FeatureSet(Text(dict))

// group by name, only keep multiple occurences
var grouped_fs = GroupBy(tablefs, "wholename", {"name": "NameCount", "expression": "wholename", "statistic": "Count"})
var duplicate_fs = Filter(grouped_fs, "NameCount > 1")

// construct an SQL filter statement with which to filter the original data
// wholename IN ('Name1', 'Name2', 'Name3')
var filter_names = []
for(var f in duplicate_fs) {
    Push(filter_names, f.wholename)
}
var filter_statement = `wholename IN ('${Concatenate(filter_names, "', '")}')`

// filter the original feature set with this statement and return the result
// this contains all the original rows with duplicate names
return Filter(tablefs, filter_statement)