f90nml - A Fortran namelist parser, generator, and editor

A Python module and command line tool for parsing Fortran namelist files

Documentation

The complete documentation for f90nml is available from Read The Docs.

http://f90nml.readthedocs.org/en/latest/

About f90nml

f90nml is a Python module and command line tool that provides a simple interface for the reading, writing, and modifying Fortran namelist files.

A namelist file is parsed and converted into an Namelist object, which behaves like a standard Python dict. Values are converted from Fortran data types to equivalent primitive Python types.

The command line tool f90nml can be used to modify individual values inside of a shell environment. It can also be used to convert the data between namelists and other configuration formats. JSON and YAML formats are currently supported.

Quick usage guide

To read a namelist file sample.nml which contains the following namelists:

&config_nml
   input = 'wind.nc'
   steps = 864
   layout = 8, 16
   visc = 1.0e-4
   use_biharmonic = .false.
/

we would use the following script:

import f90nml
nml = f90nml.read('sample.nml')

which would would point nml to the following dict:

nml = {
    'config_nml': {
        'input': 'wind.nc',
        'steps': 864,
        'layout': [8, 16],
        'visc': 0.0001,
        'use_biharmonic': False
    }
}

File objects can also be used as inputs:

with open('sample.nml') as nml_file:
    nml = f90nml.read(nml_file)

To modify one of the values, say steps, and save the output, just manipulate the nml contents and write to disk using the write function:

nml['config_nml']['steps'] = 432
nml.write('new_sample.nml')

Namelists can also be saved to file objects:

with open('target.nml') as nml_file:
   nml.write(nml_file)

To modify a namelist but preserve its comments and formatting, create a namelist patch and apply it to a target file using the patch function:

patch_nml = {'config_nml': {'visc': 1e-6}}
f90nml.patch('sample.nml', patch_nml, 'new_sample.nml')

Command line interface

A command line tool is provided to manipulate namelist files within the shell:

$ f90nml config.nml -g config_nml -v steps=432

&config_nml
   input = 'wind.nc'
   steps = 432
   layout = 8, 16
   visc = 1.0e-4
   use_biharmonic = .false.
/

See the documentation for details.

Installation

f90nml is available on PyPI and can be installed via pip:

$ pip install f90nml

The latest version of f90nml can be installed from source:

$ git clone https://github.com/marshallward/f90nml.git
$ cd f90nml
$ pip install .

conda

There is a conda-forge feedstock (not maintained, nor supported by, the Author).

Information on supported versions and platforms, and detailed installation instructions using conda and conda-forge is available here:

https://github.com/conda-forge/f90nml-feedstock

but TL;DR:

$ conda install -c conda-forge f90nml

Package distribution

f90nml is not distributed through any official packaging tools, but it is available on Arch Linux via the AUR:

$ git clone https://aur.archlinux.org/python-f90nml.git
$ cd python-f90nml
$ makepkg -sri

Volunteers are welcome to submit and maintain f90nml on other distributions.

Local install

Users without install privileges can append the --user flag to pip from the top f90nml directory:

$ pip install --user .

If pip is not available, then setup.py can still be used:

$ python setup.py install --user

When using setup.py locally, some users have reported that --prefix= may need to be appended to the command:

$ python setup.py install --user --prefix=

YAML support

The command line tool offers support for conversion between namelists and YAML formatted output. If PyYAML is already installed, then no other steps are required. To require YAML support, install the yaml extras package:

$ pip install f90nml[yaml]

To install as a user:

$ pip install --user .[yaml]

Contributing to f90nml

Users are welcome to submit bug reports, feature requests, and code contributions to this project through GitHub. More information is available in the Contributing guidelines.