# CreateFeatureClass_management (simple question)

1189
6
10-10-2018 01:31 PM
MVP Frequent Contributor

I was asked about a seemingly simple line of code in one of my scripts and I didn't have an immediate answer.  After a few pages of g-search, I'm still a bit stumped, but not afraid to ask for the answer!

facFC = arcpy.CreateFeatureclass_management("in_memory", "facFC", "POINT")[0]‍

What exactly is the [0] at the end of this statement for?

There's examples of having it there and other examples it's not found.

Thanks!

6 Replies
MVP Esteemed Contributor
CreateFeatureclass_management (out_path, out_name,
{geometry_type},
{template}, {has_m}, {has_z},
{spatial_reference},
{config_keyword},
{spatial_grid_1}, {spatial_grid_2}, {spatial_grid_3},
{out_alias})‍‍‍‍‍‍‍

the [0] is a slice, meaning get the first, which in your case would be the first... first of what is sketchy.

The above is the preferred by me, approach,

out_path…. the location where the featureclass is going

out_name .. the name to give the featureclass

geometry_type… point, line, polygon stuff

template … if you have a featureclass that you want to use for emulation

has_m, has_z - well? do you want them?

spatial_reference - specify it now! or you have to use the define projection tool

the rest... might be important but not critical if saving to a gdb which would have the info anyway.

Note!  there is no ….. blah = CreateFeatureclass

because everything is inside the function

MVP Frequent Contributor

I had a similarly "fuzzy" response when I got the question, I had a hunch that [0] is an index position but not exactly sure to what.

Note!  there is no ….. blah = CreateFeatureclass

I am able to reference facFC var on the next line in order to add a field:

facFC = arcpy.CreateFeatureclass_management("in_memory", "facFC", "POINT")[0]
arcpy.AddField_management(facFC, "NAME", "TEXT", None, None, 40, "", "NULLABLE", "NON_REQUIRED")‍‍

Thank you for the response!

MVP Esteemed Contributor

I never liked shortcuts that obscure …

It  is 'ok' but not recommended since putting in the featureclass name as a parameter doesn't hide the fact and relying on slicing might lead you to slice the wrong parameter the next time for the next tool that you use.

Esri Esteemed Contributor

I second what Dan_Patterson  mentioned.  I normally define the output and other variable at the start of the script and use in this case the os module to split the path and name in a path and a name. Something like this:

import arcpy
import os

fc_out = r'C:\some\path\to\outputfgdb.gdb\YourFeatureclass'

ws, fc_name = os.path.split(fc_out)
arcpy.CreateFeatureclass_management(ws, fc_name, etc)

arcpy.AddField_management(fc_out, YourFieldName, etc)‍‍‍‍‍‍‍‍‍
MVP Honored Contributor

I believe the [0] indicates the first item in the Result object returned from the tool, which happens to be the new feature class name.

MVP Esteemed Contributor

As Darren points out, nearly all geoprocessing tools return Result—Help | ArcGIS Desktop objects, which explains the ArcGIS/ArcPy side of your question.

For the Python side of your question, Python classes can emulate container objects by implementing certain special methods:  3. Data model - Emulating Container Types— Python 2.7.15 documentation.  Support for the indexing and slicing operator, i.e., [], is implemented using the object.__getitem__ special method.  For the ArcPy Result object, the __getitem__ special method either points to Result.getOutput() or implements the same code.

In short, Result[0] is the same as calling Result.getOutput(0) .