ArcGIS Python Toolbox – Reordering Parameters and Referencing by Name

arcgis-proarcpyparameterspython-toolbox

Is there a way to reference parameters in an ArcGIS Pro Python Toolbox by name, instead of by numeric index / positional argument?

I built a data entry tool in a Python Toolbox that has over 100 parameters. If I need to add, remove, or reorder any of these, it becomes a massive undertaking to edit every entry in updateParameters(), updateMessages(), and execute().

Update: I've deleted my attempt, because it's bad, and only user2856's accepted answer should be on record.

Best Answer

Don't use a global. Just create a dictionary whenever you need to access parameters, e.g. in updateParameters, updateMessages, execute, etc:

parameters = {p.name: p for p in parameters}

e.g.

def updateParameters(self, parameters):
    parameters = {p.name: p for p in parameters}
    etc...

def updateMessages(self, parameters):
    parameters = {p.name: p for p in parameters}
    etc...

def execute(self, parameters, messages):
    parameters = {p.name: p for p in parameters}
    etc...

Another option is to create a namedtuple (could be a good ArcGIS Pro Idea) so you can access by index (backwards compatibility) as well as name.

from collections import namedtuple

etc...

    def execute(self, parameters, messages):  
        parameters = namedtuple('Parameters', (p.name for p in parameters))(*parameters)
        # do something with parameters[0] or parameters.some_name
        etc...

There's an ArcGIS Idea to change the parameters type from a list to a NamedTuple that you can support.

Related Solutions

[GIS] Python toolbox – keep parameter value updated after re-altering parameters

Unfortunately the parameterType property is read only (see here and here), so you cannot change it. However, you can modify the updateMessages function to check for blank string. It might look something like below.

Try this (you will need to modify parameter indices, as mine are 0-2):

def updateParameters(self):
  """Modify the values and properties of parameters before internal
  validation is performed.  This method is called whenever a parameter
  has been changed."""
  if not self.params[0].hasBeenValidated: #if boolean 1 changes
      if self.params[0].value == True:
          self.params[2].enabled = True #string parameter
      else:
          self.params[2].enabled = False

  elif not self.params[1].hasBeenValidated: #if boolean 2 changes
      if self.params[1].value == True:
          if self.params[0] == False:
              self.params[2].enabled = False #string parameter
  return

def updateMessages(self):
  """Modify the messages created by internal validation for each tool
  parameter.  This method is called after internal validation."""
  if self.params[0].value == True:
      if self.params[2].value == '':
          self.params[2].setErrorMessage("String required!")
  return

You could always have the string parameter as required, default '', and check it in your code.

[GIS] Using multiValue in Python Toolbox with parameters interaction

Select All and Unselect All option in the tool window don't work.

The select out values don't change if the user unchecked several of them.

This is because you are resetting the values every time there is a value set for parameter 0 and 1 (done in the line parameters[2].value = vals). Comment this line out.

def updateParameters(self, parameters):
    """Modify the values and properties of parameters before internal
    validation is performed.  This method is called whenever a parameter
    has been changed."""

    if parameters[0].value and parameters[1].value:
        fc = str(parameters[0].value)
        col = str(parameters[1].value)
        vals = sorted(set(row[0] for row in arcpy.da.SearchCursor(fc,[col]) if row))
        parameters[2].filter.list = vals
        #parameters[2].value = vals
    return

The UpdateCursor part removes all records rather than reflecting the user interests.

I think this sounds a bit odd that you are actually removing features that user doesn't want to work with. Isn't it easier just to filter them out or create a new dataset that will contain the data user would like to work with? There is a risk your users will be deleting things and be unaware of this.

Anyways, this is the code that will remove all rows from the features class that were not selected by user:

def execute(self, parameters, messages):
    """The source code of the tool."""
    values = parameters[2].valueAsText
    values_to_remove = values.split(';')

    for x in values_to_remove:
        arcpy.AddMessage(x)

    fc = parameters[0].valueAsText
    field = parameters[1].valueAsText
    with arcpy.da.UpdateCursor(fc,[field]) as cursor:
        for row in cursor:
            if not row[0] in values_to_remove:
                cursor.deleteRow()

    return

Some notes:

never name your variables using built-in Python keywords such as class or pass;
never name your variables using built-in Python data types such as list or dict;

If you have a variable named list it might bite you later. The reason why all features are deleted in your tool is because you are resetting the values in the parameter values essentially unselecting all of them (will will force UpdateCursor to delete all rows).

Best Answer

Related Solutions

[GIS] Python toolbox – keep parameter value updated after re-altering parameters

[GIS] Using multiValue in Python Toolbox with parameters interaction

Related Question