Initializing Feel++

The core module provides the basic data structures to

  • setup and run Feel++ application in a parallel environment.

  • handle the command line options

  • download and upload data from and to GitHub and/or Girder

1. Setting up the Feel++ Environment

To set Feel++ environment, we create an environment and set the associated repository for the results. A Feel++ environment can be created only once. The repository can be global with respect to $HOME/.feelppconfig globalroot setting or local with respect to the current directory.

Set the Feel++ environment with local repository
import feelpp.core as fppc
import sys
app = fppc.Environment(["myapp"],config=fppc.localRepository(""))
Results 
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
File /usr/lib/petsc/lib/python3/dist-packages/petsc4py/PETSc.py:3
      2 from petsc4py.lib import ImportPETSc  # noqa: E402
----> 3 PETSc = ImportPETSc(ARCH)
      4 PETSc._initialize()

File /usr/lib/petsc/lib/python3/dist-packages/petsc4py/lib/__init__.py:29, in ImportPETSc(arch)
     28 path, arch = getPathArchPETSc(arch)
---> 29 return Import('petsc4py', 'PETSc', path, arch)

File /usr/lib/petsc/lib/python3/dist-packages/petsc4py/lib/__init__.py:95, in Import(pkg, name, path, arch)
     94 # import extension module from 'path/arch' directory
---> 95 module = import_module(pkg, name, path, arch)
     96 module.__arch__ = arch  # save arch value

File /usr/lib/petsc/lib/python3/dist-packages/petsc4py/lib/__init__.py:73, in Import..import_module(pkg, name, path, arch)
     72 sys.modules[fullname] = module
---> 73 spec.loader.exec_module(module)
     74 return module

File petsc4py/PETSc.pyx:1, in init petsc4py.PETSc()

ValueError: numpy.dtype size changed, may indicate binary incompatibility. Expected 96 from C header, got 88 from PyObject

The above exception was the direct cause of the following exception:

SystemError                               Traceback (most recent call last)
File :1
----> 1 import feelpp.core as fppc
      2 import sys
      3 app = fppc.Environment(["myapp"],config=fppc.localRepository(""))

File /usr/lib/python3/dist-packages/feelpp/core/__init__.py:12
     10 from ._feelpp import  *
     11 from ._core import *
---> 12 from ._alg import *
     13 from ._mesh import *
     14 from ._plot import *

SystemError: initialization of _alg raised unreported exception
Query the app about the current environment
print("pid:",app.worldComm().localRank() )
print("isMasterRank:",app.isMasterRank() )
print("is parallel: ",app.isParallel() )
Results 
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
File :1
----> 1 print("pid:",app.worldComm().localRank() )
      2 print("isMasterRank:",app.isMasterRank() )
      3 print("is parallel: ",app.isParallel() )

NameError: name 'app' is not defined

2. Downloading data

Feel++ can query data on GitHub and Girder.

readme=fppc.download( "github:{repo:feelpp,path:README.adoc}", worldComm=app.worldCommPtr() )[0]

print("downloaded Feel++ README.adoc from Github: ",readme)
Results 
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
File :1
----> 1 readme=fppc.download( "github:{repo:feelpp,path:README.adoc}", worldComm=app.worldCommPtr() )[0]
      3 print("downloaded Feel++ README.adoc from Github: ",readme)

NameError: name 'fppc' is not defined

The code will get the file README.adoc from the toplevel Feel++ github directory downloaded

A bit more interesting example: the following code will download a csv file from the Feel++ github repository and plot the data using the plotly library.

acsv=fppc.download( "github:{repo:feelpp,path:toolboxes/fluid/cases/moving_body/gravity/cylinder_under_gravity/curve_comparison.csv}", worldComm=app.worldCommPtr() )[0] (1)
import pandas as pd (2)
df = pd.read_csv(acsv, sep=",") (3)
df.columns = df.columns.str.replace(' ', '')
print(df.head())
Results 
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
File :1
----> 1 acsv=fppc.download( "github:{repo:feelpp,path:toolboxes/fluid/cases/moving_body/gravity/cylinder_under_gravity/curve_comparison.csv}", worldComm=app.worldCommPtr() )[0] #
      2 import pandas as pd #
      3 df = pd.read_csv(acsv, sep=",") #

NameError: name 'fppc' is not defined
1 download the file curve_comparison.csv from the Feel++ github repository toolboxes/fluid/cases/moving_body/gravity/cylinder_under_gravity/curve_comparison.csv
2 use the pandas library to read the csv file
3 read the csv file and remove the spaces in the column names

We can now use plotly to plot the data

import plotly.express as px
fig = px.scatter(df,x="TIME", y="Y_CM", title="y-displacement of the center of mass(CM) of the cylinder",labels={"TIME":"t (s)","Y_CM":r'y-displacement (m)'})
fig.show()
Results

2.1. Next steps