Python error: use "*" in a function parameter

[ArnauForner]

Hello,

When using this function:

arcpy.da.FeatureClassToNumPyArray(featureClass, "*")

With an asterisk in the second parameter I get this error in Pyscripter:

TypeError: long() argument must be a string or a number, not 'NoneType'

While when I put the name of a field or a some fields I have no problem.

According to ESRI help, this parameter can be populated with:

A list (or tuple) of field names. For a single field, you can use a string instead of a list of strings.

Or use an asterisk (*) instead of a list of fields if you want to access all fields from the input table.

What am I doing wrong?

Thanks

[JamesCrandall]

What's wrong with listing out the fields?  It's a good practice to be explicit in the fields you want to work with anyway.  However, if you must use all of the fields then you can just easily build a list of them:

eflds = [f.name for f in arcpy.ListFields(MyFeatureclass)]‍

[JoshuaBixby]

You mention passing a field or some fields, what if you pass all fields like jamesfreddyc‌ suggests?

The error suggests that a NULL in the data (which is returned as NoneType) is failing to be coerced into a Long data type.  In order to troubleshoot anymore, the field that contains the problematic data needs to be isolated.

Does your feature class contain any NULL or empty geometries?

[ArnauForner]

Thanks Joshua, I found a null value in one of the fields

[JoshuaBixby]

Look at the skip_nulls and null_value parameters in the documentation.

[DanPatterson_Retired]

As a suggestion, when you are using parameters, read their case requirements explicitly

in your case

FeatureClassToNumPyArray (in_table, field_names, {where_clause}, {spatial_reference}, {explode_to_points}, {skip_nulls}, {null_value}).  

The keyword arguments (**kwargs) already has a default.  It has been my experience also, that if a kwarg has a default, don't specify it.  The default should really be an empty list/tuple and not a string, which it does.  You would have obtained all the fields if you had left out your second parameter altogether OR specified all the field names within a tuple as James showed you how to do.