Setting and Retrieving Parameters

APDL parameters can be retrieved from and instance of Mapdl using the Mapdl.parameters. For example, if you wish to use the MAPDL’s Mapdl.get() to populate a parameter, you can then access the parameter with:

>>> from ansys.mapdl.core import launch_mapdl
>>> mapdl = launch_mapdl()
>>> mapdl.get('DEF_Y', 'NODE' , 2, 'U' ,'Y')
>>> mapdl.parameters['DEF_Y']

You can also set both scalar and array parameters from python objects using Mapdl.parameters with:

>>> mapdl.parameters['MY_ARRAY'] = np.arange(10000)
>>> mapdl.parameters['MY_ARRAY']
array([0.00000e+00, 1.00000e+00, 2.00000e+00, ..., 9.99997e+05,
       9.99998e+05, 9.99999e+05])

>>> mapdl.parameters['MY_STRING'] = "helloworld"
>>> mapdl.parameters['MY_STRING']
"helloworld"

You can also access some built-in parameters normally accessed through the Mapdl.get(). For example, instead of getting the current routine with \*GET, ACTIVE, 0, ROUT, you can access it with:

>>> mapdl.parameters.routine
'Begin level'

For a full listing of the methods and attributes available to the Parameters class, please reference the Parameters.

Specially Named Parameters

Leading Underscored Parameters

The parameters starting with underscore ('_') are reserved parameters for MAPDL macros and routines. Their use is discouraged, and in PyMAPDL you cannot set them directly.

If you need to set one of these parameters, you can use Mapdl._run to avoid PyMAPDL parameter name checks. For example

>>> mapdl._run('_parameter=123')
'PARAMETER _PARAMETER =     123.00000000'

By default, this type of parameter cannot be seen when issuing Mapdl.parameters. However, you can change this by setting Mapdl.parameters.show_leading_underscore_parameters equal to True. For example:

>>> mapdl.parameters.show_leading_underscore_parameters=True
>>> mapdl.parameters
MAPDL Parameters
----------------
PORT                             : 50053.0
_RETURN                          : 0.0
_STATUS                          : 0.0
_UIQR                            : 17.0

Trailing Underscored Parameters

Parameters ending with an underscore are recommended for user routines and macros. You can set this type of parameter in PyMAPDL, but by default, they cannot be seen in Mapdl.parameters, unless Mapdl.parameters.show_trailing_underscore_parameters is set to True.

>>> mapdl.parameters['param_'] = 1.0
>>> mapdl.parameters
MAPDL Parameters
----------------
>>> mapdl.parameters.show_trailing_underscore_parameters=True
>>> mapdl.parameters
MAPDL Parameters
----------------
PARAM_                           : 1.0

Parameters with Leading and Trailing Underscore

These are a special type of parameter. They CANNOT be seen in Mapdl.parameters under any circumstances. Their use is not recommended.

You can still retrieve them using any of the normal methods to retrieve parameters. But you need to know the parameter name. For example:

>>> mapdl.parameters["_param_"] = 1.0
>>> mapdl.parameters
MAPDL Parameters
----------------
>>> print(mapdl.parameters['_param_'])
1.0

Issues when Importing and Exporting Numpy Arrays in MAPDL

Because of the way MAPDL is designed, there is no way to store an array where one or more dimension is zero. This can happens in Numpy arrays, where its first dimension can be set to zero.

>>> import numpy
>>> from ansys.mapdl.core import launch_mapdl
>>> mapdl = launch_mapdl()
>>> array40 = np.reshape([1, 2, 3, 4], (4,))
>>> array40
array([1, 2, 3, 4])

These types of array dimensions will be always converted to 1. For example:

>>> mapdl.parameters['mapdlarray40'] = array40
>>> mapdl.parameters['mapdlarray40']
array([[1.],
   [2.],
   [3.],
   [4.]])
>>> mapdl.parameters['mapdlarray40'].shape
(4, 1)

This means that when you pass two arrays, one with the second axis equal to zero (e.g. array40) and another one with the second axis equal to one, if later retrieved, they will have the same shape.

>>> array41 = np.reshape([1, 2, 3, 4], (4,1))
>>> array41
array([[1],
   [2],
   [3],
   [4]])
>>> mapdl.parameters['mapdlarray41'] = array41
>>> mapdl.parameters['mapdlarray41']
array([[1.],
   [2.],
   [3.],
   [4.]])
>>> np.allclose(mapdl.parameters['mapdlarray40'], mapdl.parameters['mapdlarray41'])
True