13. field
— Field data in pyFormex.¶
This module defines the Field class, which can be used to describe scalar and vectorial field data over a geometrical domain.

class
field.
Field
(geometry, fldtype, data, fldname=None)[source]¶ Scalar or vectorial field data defined over a geometric domain.
A scalar data field is a quantity having exactly one value in each point of some geometric domain. A vectorial field is a quantity having exactly nval values at each point, where nval >= 1. A vectorial field can also be considered as a collection of nval scalar fields: as far as the Field class is concerned, there is no relation between the nval components of a vectorial field.
The definition of a field is always tied to some geometric domain. Currently pyFormex allows fields to be defined on Formex and Mesh type geometries.
Fields should only be defined on geometries whose topology does not change anymore. This means that for Formex type, the shape of the coords attribute should not be changed, and for Mesh type, the shape of coords and the full contents of elems should not be changed. It is therefore best ot only add field data to geometry objects that will not be changed in place. Nearly all methods in pyFormex return a copy of the object, and the copy currently looses all the fields defined on the parent. In future however, selected transformations may inherit fields from the parent.
While Field class instances are usually created automatically by the
Geometry.addField()
method of some Geometry, it is possible to create Field instances yourself and manage the objects like you want. The Fields stored inside Geometry objects have some special features though, like being exported to a PGF file together with the geometry.A Field is defined by the following parameters:
geometry: Geometry instance: describes the geometrical domain over which the field is defined. Currently this has to be a
Formex
or aMesh
.fldtype: string: one of the following predefined field types:
 ‘node’: the field data are specified at the nodes of the geometry;
 ‘elemc’: the field data are constant per element;
 ‘elemn’: the field data vary over the element and are specified at the nodes of the elements;
 ‘elemg’: the field data are specified at a number of points of the elements, from which they can be inter or extrapolated;
The actually available field types depend on the type of the geometry object.
Formex
type has only ‘elemc’ and ‘elemn’. Mesh currently has ‘node’, ‘elemc’ and ‘elemn’.data: an array with the field values defined at the specified points. The required shape of the array depends on fldtype:
 ‘node’: ( nnodes, ) or ( nnodes, nval )
 ‘elemc’: ( nelems, ) or ( nelems, nval )
 ‘elemn’: ( nelems, nplex ) or (nelems, nplex, nval )
 ‘elemg’: ( nelems, ngp ) or (nelems, ngp, nval )
fldname: string: the name used to identify the field. Fields stored in a Geometry object can be retrieved using this name. See
Geometry.getField()
. If no name is specified, one is generated.
Example:
>>> from pyformex.formex import Formex >>> M = Formex('4:0123').replic(2).toMesh() >>> print(M.coords) [[ 0. 0. 0.] [ 0. 1. 0.] [ 1. 0. 0.] [ 1. 1. 0.] [ 2. 0. 0.] [ 2. 1. 0.]] >>> print(M.elems) [[0 2 3 1] [2 4 5 3]] >>> d = M.coords.distanceFromPlane([0.,0.,0.],[1.,0.,0.]) >>> f1 = Field(M,'node',d) >>> print(f1) Field 'field0', type 'node', shape (6,), nnodes=6, nelems=2, nplex=4 [ 0. 0. 1. 1. 2. 2.] >>> f2 = f1.convert('elemn') >>> print(f2) Field 'field1', type 'elemn', shape (2, 4), nnodes=6, nelems=2, nplex=4 [[ 0. 1. 1. 0.] [ 1. 2. 2. 1.]] >>> f3 = f2.convert('elemc') >>> print(f3) Field 'field2', type 'elemc', shape (2,), nnodes=6, nelems=2, nplex=4 [ 0.5 1.5] >>> d1 = M.coords.distanceFromPlane([0.,0.,0.],[0.,1.,0.]) >>> f4 = Field(M,'node',at.column_stack([d,d1])) >>> print(f4) Field 'field3', type 'node', shape (6, 2), nnodes=6, nelems=2, nplex=4 [[ 0. 0.] [ 0. 1.] [ 1. 0.] [ 1. 1.] [ 2. 0.] [ 2. 1.]] >>> f5 = f4.convert('elemn') >>> print(f5) Field 'field4', type 'elemn', shape (2, 4, 2), nnodes=6, nelems=2, nplex=4 [[[ 0. 0.] [ 1. 0.] [ 1. 1.] [ 0. 1.]] <BLANKLINE> [[ 1. 0.] [ 2. 0.] [ 2. 1.] [ 1. 1.]]] >>> f6 = f5.convert('elemc') >>> print(f6) Field 'field5', type 'elemc', shape (2, 2), nnodes=6, nelems=2, nplex=4 [[ 0.5 0.5] [ 1.5 0.5]] >>> print(f3.convert('elemn')) Field 'field6', type 'elemn', shape (2, 4), nnodes=6, nelems=2, nplex=4 [[ 0.5 0.5 0.5 0.5] [ 1.5 1.5 1.5 1.5]] >>> print(f3.convert('node')) Field 'field8', type 'node', shape (6, 1), nnodes=6, nelems=2, nplex=4 [[ 0.5] [ 0.5] [ 1. ] [ 1. ] [ 1.5] [ 1.5]] >>> print(f6.convert('elemn')) Field 'field9', type 'elemn', shape (2, 4, 2), nnodes=6, nelems=2, nplex=4 [[[ 0.5 0.5] [ 0.5 0.5] [ 0.5 0.5] [ 0.5 0.5]] <BLANKLINE> [[ 1.5 0.5] [ 1.5 0.5] [ 1.5 0.5] [ 1.5 0.5]]] >>> print(f6.convert('node')) Field 'field11', type 'node', shape (6, 2), nnodes=6, nelems=2, nplex=4 [[ 0.5 0.5] [ 0.5 0.5] [ 1. 0.5] [ 1. 0.5] [ 1.5 0.5] [ 1.5 0.5]]

comp
(i)[source]¶ Return the data component i of a vectorial Field.
Parameters:
 i: int: component number of a vectorial Field. If the Field is a scalar one, any value will return the full scalar data.
Returns an array representing the scalar data over the Geometry.

convert
(totype, toname=None)[source]¶ Convert a Field to another type.
Parameters:
 totype: string: target field type. Can be any of the available
field types. See
Field
class. If the target type is equal to the source type, a copy of the original Field will result. This may or may not be a shallow copy.  toname: string: the name of the target field. If not specified, a autoname is generated.
Returns a Field of type totype.
 totype: string: target field type. Can be any of the available
field types. See