femap-neutral-parser 0.14.0

Low-level data access

Under the hood, output vectors (block451) are organized as nested dictionaries [<vector title>][<LCID>]:

>>> pp(neu._output_vectors["displacements::t_total"][2])
{'abs_max': 2.578386,
 'calc_warn': True,
 'cent_total': True,
 'comp_dir': 1,
 'component_vec': [10002.0,
                   10003.0,
                   10004.0,
                   0.0,
                   0.0,
                   0.0,
                   0.0,
                   0.0,
                   0.0,
                   0.0],
 'id_max': 7,
 'id_min': 1,
 'max_val': 2.578386,
 'min_val': 0.0,
 'record': array([( 1, 0.000000e+00), ( 2, 2.045391e-01), ( 3, 0.000000e+00),
       ( 4, 1.468270e-02), ( 5, 9.231050e-05), ( 6, 6.276400e-01),
       ( 7, 2.578386e+00), ( 8, 1.025100e-01), ( 9, 2.578363e+00),
       (10, 1.916094e+00), (11, 1.100510e+00), (12, 2.389742e+00)],
      dtype=[('NodeID', '<i8'), ('displacements::t_total', '<f8')]),
 'vecID': 10001}

Aggregated Outputs

Slightly highest level than _output_vectors access, aggregated output is available using Parser.vectors() method. For example, to get all outputs for translations vectors:

>>> arr = neu.vectors(("displacements::t1", "displacements::t2", "displacements::t3"))
>>> arr
rec.array([( 1,  0.        , 0.,  0.000000e+00, 1),
           ( 2, -0.1870816 , 0.,  0.000000e+00, 1),
           ...
           (11,  0.        , 0., -1.100510e+00, 2),
           (12,  0.        , 0., -2.389742e+00, 2)],
          dtype=[('NodeID', '<i8'), ('displacements::t1', '<f8'), ('displacements::t2', '<f8'), ('displacements::t3', '<f8'), ('SubcaseID', '<i8')])

Returned value is a numpy structured array (https://numpy.org/doc/stable/user/basics.rec.html). If Pandas is available, you can request to have a DataFrame instead:

>>> neu.vectors(("displacements::t1", "displacements::t2", "displacements::t3"), asdf=True)
                   displacements::t1  displacements::t2  displacements::t3
SubcaseID NodeID
1         1       0.000000       0.0  0.000000
          2      -0.187082       0.0  0.000000
...
          11      0.000000       0.0 -0.956073
          12      0.000000       0.0 -1.602912
2         1       0.000000       0.0  0.000000
          2      -0.204539       0.0  0.000000
...
          11      0.000000       0.0 -1.100510
          12      0.000000       0.0 -2.389742

You can also request sub-cases IDs, or request raw headers:

>>> neu.vectors(("displacements::t1", "displacements::t2", "displacements::t3"), asdf=True, raw=True,
...              SubcaseIDs=2)
                  T1 translation  T2 translation  T3 translation
SubcaseID NodeID
2         1             0.000000             0.0        0.000000
          2            -0.204539             0.0        0.000000
...
          11            0.000000             0.0       -1.100510
          12            0.000000             0.0       -2.389742

High-Level access

At highest level, you can use the get method that already organize vectors for you:

>>> neu.get(what="displacements", asdf=True)
                        t1   t2        t3        r1        r2            r3
SubcaseID NodeID
1         1       0.000000  0.0  0.000000 -0.000432  0.008923  4.699029e-03
          2      -0.187082  0.0  0.000000  0.000432  0.008923  4.666047e-03
...
2         1       0.000000  0.0  0.000000 -0.000007  0.009755  5.137517e-03
          2      -0.204539  0.0  0.000000  0.000007  0.009755  5.101457e-03
...
          11      0.000000  0.0 -1.100510  0.000000  0.010988  1.006013e-06
          12      0.000000  0.0 -2.389742  0.000000  0.004149  1.360364e-08

One can get the list of high-level shortcuts using Parser.get_vectors() which will return a set of available headers:

>>> neu.get_vectors() == {
... 'elem_gpf',
... 'force_vectors',
... 'displacements',
... 'summed_gpf',
... 'mpc_forces',
... 'constraint_gpf',
... 'cbar_force',
... 'cbar_stress',
... 'cbush_force',
... 'cbush_stress',
... 'cbar_ms',
... 'applied_gpf',
... 'spc_forces',
... 'cbeam_force'
... }
True

0.8.3 (2021-09-12)

Major internal refactoring

• output_vectors becomes internal (_output_vectors). The recommended ways to access data is now Parser.vectors() or Parser.get() at a higher level.

• no more autotranslate initialisation parameter. Everything is harmonized under the hood.

0.6.0 (2021-07-12)

• add engineering shortcuts

• harmonize NodeID and SubcaseID with pynastran

• vectors are now case insensible

0.5.0 (2021-04-16)

• improve parsing speed on large files

0.4.0 (2021-04-16)

add:

• Parser.info()

• Parser.vectors()

0.3.0 (2021-04-16)

• add Parser.info() and Parser.vectors()

• update documentation

• add info() doprint option

0.2.0 - 0.3.0 (2021-04-16)

• Update Documentation

0.1.0 (2021-03-24)

First release on PyPI.