KeyFile¶

Ressources:

This class is for handling generic Keywords in a KeyFile. The class simply provides comfortable means to access or manipulate the lines of the Keyword. This has the advantage, that writing the Keyword will yield the exact same card including comments except for the changes.

class qd.cae.dyna.KeyFile(filepath="", read_keywords=True, parse_mesh=True, load_includes=True, encryption_detection=0.7)¶

Parameters

filepathstr: path to the keyfile
read_keywordsbool: whether to read all keywords
parse_meshbool: whether the mesh keywords shall be parsed
load_includesbool: load all includes within the file
encryption_detectionfloat: detection threshold for encrypted include files. Must be between 0 and 1.

Returns

keyfileKeyFile: instance

Raises

ValueError: in case of a wrong filepath or invalid encryption threshold
RuntimeError: if anything goes wrong during reading

Notes

If read_keywords=True every keyword found will be loaded and made accessible by the generic Keyword class. If at the same time parse_mesh=False then also the mesh itself is treated as generic keywords.

If parse_mesh=True, then the mesh keywords are loaded and parsed. Also the mesh specific keyword classes are used (see the keyword classes).

The argument encryption_detection is used to skip encrypted include files. It is simply tested against the entropy of every include divided by 8 for normalization. Encrypted files usually have a very high entropy. The entropy of a file can be tested with the function qd.cae.dyna.get_file_entropy.

Examples

>>> # load keyfile and parse the mesh
>>> keyfile = KeyFile("path/to/keyfile", parse_mesh=True)
>>> # get some mesh data
>>> node = keyfile.get_nodeByIndex(0)
>>> # check the keywords
>>> keyfile.keys()
['*NODE', '*ELEMENT_SHELL', ...]
>>> # get all part keywords
>>> part_keywords = keyfile["*PART"]

One can also create empty keyfiles and built them.

>>> keyfile = KeyFile(parse_mesh=True)
>>> kw = keyfile.add_keyword("*NODE")
>>> kw.add_node(1, x=0, y=0, z=0)

Methods

`add_keyword`(lines[, position])	Parameters
`get_elementByIndex`(element_type, index)	Parameters
`get_element_coords`(element_filter)	Parameters
`get_element_energy`(element_filter)	Parameters
`get_element_history_vars`(element_type)	Parameters
`get_element_ids`(element_filter)	Parameters
`get_element_node_ids`(element_type, n_nodes)	Parameters
`get_element_plastic_strain`(element_filter)	Parameters
`get_element_strain`(element_filter)	Parameters
`get_element_stress`(element_filter)	Parameters
`get_element_stress_mises`(element_filter)	Parameters
`get_elements`([element_filter])	Parameters
`get_filepath`()	Returns
`get_include_dirs`()	Returns
`get_includes`()	Returns
`get_nElements`([element_filter])	Parameters
`get_nNodes`()	Returns
`get_nParts`()	Returns
`get_nodeByID`(id)	Parameters
`get_nodeByIndex`(index)	Parameters
`get_node_acceleration`()	Returns
`get_node_coords`()	Returns
`get_node_ids`()	Returns
`get_node_velocity`()	Returns
`get_nodes`()	Returns
`get_partByID`(id)	Parameters
`get_partByIndex`(index)	Parameters
`get_partByName`(name)	Parameters
`get_parts`()	Returns
`keys`()	Returns
`plot`(self[, export_filepath])	Plot the KeyFile geometry.
`plot_parts`(parts[, export_filepath])	Plot a selected group of parts
`remove_keyword`(name, index)	Parameters
`save`()	keys()

get_elementByID
get_end_keyword_position

__init__(filepath="", read_keywords=True, parse_mesh=True, load_includes=True, encryption_detection=0.7)¶

KeyFile(filepath=””, read_keywords=True, parse_mesh=True, load_includes=True, encryption_detection=0.7)

Parameters

filepathstr: path to the keyfile
read_keywordsbool: whether to read all keywords
parse_meshbool: whether the mesh keywords shall be parsed
load_includesbool: load all includes within the file
encryption_detectionfloat: detection threshold for encrypted include files. Must be between 0 and 1.

Returns

keyfileKeyFile: instance

Raises

ValueError: in case of a wrong filepath or invalid encryption threshold
RuntimeError: if anything goes wrong during reading

Notes

If read_keywords=True every keyword found will be loaded and made accessible by the generic Keyword class. If at the same time parse_mesh=False then also the mesh itself is treated as generic keywords.

If parse_mesh=True, then the mesh keywords are loaded and parsed. Also the mesh specific keyword classes are used (see the keyword classes).

The argument encryption_detection is used to skip encrypted include files. It is simply tested against the entropy of every include divided by 8 for normalization. Encrypted files usually have a very high entropy. The entropy of a file can be tested with the function qd.cae.dyna.get_file_entropy.

Examples

>>> # load keyfile and parse the mesh
>>> keyfile = KeyFile("path/to/keyfile", parse_mesh=True)
>>> # get some mesh data
>>> node = keyfile.get_nodeByIndex(0)
>>> # check the keywords
>>> keyfile.keys()
['*NODE', '*ELEMENT_SHELL', ...]
>>> # get all part keywords
>>> part_keywords = keyfile["*PART"]

One can also create empty keyfiles and built them.

>>> keyfile = KeyFile(parse_mesh=True)
>>> kw = keyfile.add_keyword("*NODE")
>>> kw.add_node(1, x=0, y=0, z=0)

add_keyword(lines, position=-1)¶

Parameters

linesstr or list of str: keyword definition
positionint: where to position the keyword in the KeyFile, if negative it is appended to the end

Notes

When the KeyFile was created using parse_mesh=True, then if a mesh keyword is created, the corresponding classes are returned instead of a generic keyword:

NodeKeyword

ElementKeyword

PartKeyword

Examples

Adding *NODE without mesh parsing:

>>> # open a keyfile
>>> keyfile = KeyFile("path/to/keyfile.key")
>>> # get the node keyword as generic keyword
>>> kw = keyfile.add_keyword("*NODE")
>>> type(kw)
<class 'qd.cae.dyna_cpp.Keyword'>

Adding *NODE with mesh parsing support

>>> keyfile = KeyFile("path/to/keyfile.key", load_mesh=True)
>>> kw = type(keyfile.add_keyword(["$ Comment header",
>>>                                 "*NODE"]))
>>> type(kw)
<class 'qd.cae.dyna_cpp.NodeKeyword'>
>>> # now it's simple to add nodes
>>> node = kw.add_node(id=1, x=0, y=0, z=0)

get_elementByIndex(element_type, index)¶

Parameters

element_typeElement.type: type of the element. Must be beam, shell or solid.
indexint or list(int): element index or list of indexes

Returns

elementElement: Element(s) depending on the arguments

Raises

ValueError: if invalid element_type is given or index does not exist

Examples

>>> # single element
>>> elem = femfile.get_elementByIndex(Element.shell, 0)
>>> # multiple elements
>>> list_of_shells = femfile.get_elementByID(Element.shell, [0,1,2])
>>> 
>>> femfile.get_elementByIndex(Element.beam, 1).get_type()
type.beam
>>> femfile.get_elementByIndex(Element.solid, 1).get_type()
type.solid
>>> femfile.get_elementByIndex(Element.tshell, 1).get_type()
type.tshell

get_element_coords(element_filter)¶

Parameters

element_filterElement.type: optional type for filtering

Returns

fieldnp.ndarray: coords vector of the elements (nElems x nTimesteps x 3)

Notes

If an element does not have the respective result, 0 is set as default value.

Examples

>>> d3plot = D3plot("path/to/d3plot", read_states="coords")
>>> d3plot.get_nElements(Element.shell)
4969
>>> d3plot.get_nTimesteps()
32
>>> d3plot.get_element_coords(Element.shell).shape
(4969, 32. 3)

get_element_energy(element_filter)¶

Parameters

element_filterElement.type: optional type for filtering

Returns

fieldnp.ndarray: energy of the elements (nElems x nTimesteps)

Notes

If an element does not have the respective result, 0 is set as default value.

Examples

>>> d3plot = D3plot("path/to/d3plot", read_states="energy mean")
>>> d3plot.get_nElements(Element.shell)
4969
>>> d3plot.get_nTimesteps()
32
>>> d3plot.get_element_energy(Element.shell).shape
(4969, 32)

get_element_history_vars(element_type)¶

Parameters

element_typeElement.type: type of the element

Returns

fieldnp.ndarray: history_vars vector of the elements (nElems x nTimesteps x nHistoryVars)

Notes

IMPORTANT: You can not query the history vars without specifying an element type. In LS-Dyna e.g. history var 7 is different between shells and beams. To prevent wrong usage and bugginess, we enforce this policy here. We didn’t mess this one up!

If an element does not have the respective result, 0 is set as default value.

Examples

>>> d3plot = D3plot("path/to/d3plot", read_states="shell history 7 max")
>>> d3plot.get_nElements(Element.shell)
4969
>>> d3plot.get_nTimesteps()
32
>>> # Element type is required here!
>>> d3plot.get_element_history_vars(Element.shell).shape
(4969, 32. 10)

get_element_ids(element_filter)¶

Parameters

element_filterElement.type: optional element type for filtering

Returns

element_idsnp.ndarray: Ids of the elements

Notes

If the ids for all elements are queried (Element.none as filter type) then the returned order is:

Element.beam

Element.shell

Element.solid

Element.tshell

Examples

>>> d3plot = D3plot("path/to/d3plot")
>>> # get all ids (beam, shell, solid, tshell)
>>> d3plot.get_element_ids()
array([4687, 4688, 4689, 4690, ...])
>>> # filter only beams
>>> d3plot.get_element_ids(Element.beam)
array([9545, 9546, 9547, 9511, ...])

get_element_node_ids(element_type, n_nodes)¶

Parameters

element_typeElement.type: type of the element
n_nodesint: number of nodes

Returns

element_node_idsnp.ndarray: ids of the nodes for every element (shape elems x n_nodes).

Notes

The number of nodes must be specified since elements of one type may have different number of nodes which would result in a non-uniform matrix.

Examples

>>> d3plot = D3plot("path/to/d3plot")
>>> d3plot.get_element_node_ids(Element.shell, 4)
array([[347, 354, 343, 344],
       [354, 355, 342, 343],
       [349, 352, 356, 348]], dtype=int32)

get_element_plastic_strain(element_filter)¶

Parameters

element_filterElement.type: optional type for filtering

Returns

fieldnp.ndarray: effective plastic strain of the elements (nElems x nTimesteps)

Notes

If an element does not have the respective result, 0 is set as default value.

Examples

>>> d3plot = D3plot("path/to/d3plot", read_states="plastic_strain max")
>>> d3plot.get_nElements(Element.shell)
4969
>>> d3plot.get_nTimesteps()
32
>>> d3plot.get_plastic_strain(Element.shell).shape
(4969, 32)

get_element_strain(element_filter)¶

Parameters

element_filterElement.type: optional type for filtering

Returns

fieldnp.ndarray: strain vector of the elements (nElems x nTimesteps x 6)

Notes

If an element does not have the respective result, 0 is set as default value.

Examples

>>> d3plot = D3plot("path/to/d3plot", read_states="strain mean")
>>> d3plot.get_nElements(Element.shell)
4969
>>> d3plot.get_nTimesteps()
32
>>> d3plot.get_element_strain(Element.shell).shape
(4969, 32. 6)

get_element_stress(element_filter)¶

Parameters

element_filterElement.type: optional type for filtering

Returns

fieldnp.ndarray: stress vector of the elements (nElems x nTimesteps x 6)

Notes

If an element does not have the respective result, 0 is set as default value.

Examples

>>> d3plot = D3plot("path/to/d3plot", read_states="stress mean")
>>> d3plot.get_nElements(Element.shell)
4969
>>> d3plot.get_nTimesteps()
32
>>> d3plot.get_element_stress(Element.shell).shape
(4969, 32. 6)

get_element_stress_mises(element_filter)¶

Parameters

element_filterElement.type: optional type for filtering

Returns

fieldnp.ndarray: mises stress of the elements (nElems x nTimesteps)

Notes

If an element does not have the respective result, 0 is set as default value.

Examples

>>> d3plot = D3plot("path/to/d3plot", read_states="stress_mises mean")
>>> d3plot.get_nElements(Element.shell)
4969
>>> d3plot.get_nTimesteps()
32
>>> d3plot.get_element_stress_mises(Element.shell).shape
(4969, 32)

get_elements(element_filter=Element.none)¶

Parameters

element_filterElement.type: Optional element type filter. May be beam, shell or solid.

Returns

elementslist(Element): list of Elements

Raises

ValueError: if invalid element_filter is given.

Notes

Get the elements of the femfile. One may use a filter by type.

Examples

>>> all_elements = femfile.get_elements()
>>> shell_elements = femfile.get_elements(Element.shell)

get_filepath()¶

Returns

filepathstr: Filepath of the femfile.

Examples

>>> femfile.get_filepath()
"path/to/femfile"

get_include_dirs()¶

Returns

include_dirslist of str: list of directories searched for includes. This function can be used to check which directories the library recognizes.

Notes

Gets all include directories from every *INCLUDE_PATH or *INCLUDE_PATH_RELATIVE keyword.

Examples

>>> # open a keyfile
>>> keyfile = KeyFile("path/to/keyfile.key")
>>> # get all includes
>>> kf.get_include_dirs()
['path/to/includes', 'path/to/other/includes', ...]

get_includes()¶

Returns

includeslist of KeyFile: list of all loaded include files

Notes

In order to load includes, the load_includes=True option must be used in the constructor of the KeyFile or one can use the IncludeKeyword.load function, which every include keyword has.

Examples

>>> # open a keyfile
>>> keyfile = KeyFile("path/to/keyfile.key")
>>> # get all includes
>>> len(keyfile.get_includes())
7

get_nElements(element_filter=Element.none)¶

Parameters

element_filterElement.type: Optional element type filter. May be beam, shell or solid.

Returns

nElementsint: number of elements

Raises

ValueError: if invalid element_filter is given.

Examples

>>> femfile.get_nElements()
43156

get_nNodes()¶

Returns

nNodesint: number of nodes in the file

Examples

>>> femfile.get_nNodes()
43145

get_nParts()¶

Returns

nPartsint: number of parts in the database

Examples

>>> femfile.get_nParts()
7

get_nodeByID(id)¶

Parameters

idint or list(int): node id or list of node ids

Returns

nodeNode or list(Node): requested Node(s)

Raises

ValueError: if idx does not exist.

Examples

>>> # get by single id
>>> node = femfile.get_nodeByID(1)
>>> # get a list of nodes at once
>>> list_of_nodes = femfile.get_nodeByID( [1,2,3] )

get_nodeByIndex(index)¶

Parameters

indexint or list(int): internal node index or list inf indexes

Returns

nodeNode or list(Node): Node(s)

Raises

ValueError: if index larger femfile.get_nNodes().

Notes

The internal index starts at 0 and ends at femfile.get_nNodes().

Examples

>>> # single index
>>> node = femfile.get_nodeByIndex(1)
>>> # get a list of nodes at once
>>> list_of_nodes = femfile.get_nodeByIndex( [1,2,3] )

get_node_acceleration()¶

Returns

node_accelerationnp.ndarray: Accelration of all nodes

Examples

>>> d3plot.get_node_acceleration().shape
(4915,1,3)

get_node_coords()¶

Returns

node_coordsnp.ndarray: coordinates of all nodes

Examples

>>> d3plot.get_node_coords().shape
(4915, 3)

get_node_ids()¶

Returns

node_idsnp.ndarray: Ids of all nodes

Examples

>>> d3plot.get_node_ids().shape
(100001)

get_node_velocity()¶

Returns

node_velocitynp.ndarray: Velocity of all nodes

Examples

>>> d3plot.get_node_velocity().shape
(4915,1,3)

get_nodes()¶

Returns

nodeslist(Node): list of node objects

Examples

>>> list_of_nodes = femfile.get_nodes()

get_partByID(id)¶

Parameters

idint or list of int: id or ids of the part in the file

Returns

partsPart or list of Parts: output depending on arguments

Raises

ValueError: if some id does not exist.

Examples

>>> part = femfile.get_partByID(1)

get_partByIndex(index)¶

Parameters

indexint or list of int: index or indexes of the part in the file

Returns

partsPart or list of Parts: output depending on arguments

Raises

ValueError: if some index does not exist.

Examples

>>> part = femfile.get_partByIndex(0)

get_partByName(name)¶

Parameters

namestr: name of the part

Returns

partsPart: part with the given name

Raises

ValueError: if a part with name does not exist.

Examples

>>> part = femfile.get_partByName("Lower Bumper")

get_parts()¶

Returns

partslist(Part): list of all parts in the file

Examples

>>> list_of_all_parts = femfile.get_parts()

keys()¶

Returns

keyword_nameslist of str: list of keyword names saved in the file

Examples

>>> keyfile = KeyFile("path/to/keyfile.key")
>>> # get a list of all keyword names
>>> keyfile.keys()
['*BOUNDARY_SPC_SET_ID', '*PART_CONTACT', '*NODE', ...]

plot(self, export_filepath=None)¶: Plot the KeyFile geometry.

static plot_parts(parts, export_filepath=None)¶: Plot a selected group of parts

remove_keyword(name, index)¶

Parameters

namestr: name of keyword to remove
indexint: index of keyword in list

Notes

When using parse_mesh=True then all mesh keywords are using the following specific classes:

NodeKeyword

ElementKeyword

PartKeyword

These mesh keywords cannot be deleted, since deletion of mesh entities is not supported yet.

Examples

>>> # open a keyfile
>>> keyfile = KeyFile("path/to/keyfile.key")
>>> # remove the second keyword of section shell
>>> keyfile.remove_keyword("*SECTION_SHELL",1)
>>> # remove all keywords node keywords
>>> keyfile.remove_keyword("*NODE")

save()¶

keys()

Parameters

filepathstr: path for the output file

Raises

RuntimeError: if the output file can not be written

Examples

>>> # open a keyfile
>>> keyfile = KeyFile("path/to/keyfile.key")
>>> # and write the identical file again
>>> keyfile.save("path/to/identical_file.key")