libcnml 0.9.5

LibCNML

libcnml is a CNML parser library for Python.

[](https://travis-ci.org/PabloCastellano/libcnml) []( https://pypi.python.org/pypi/libcnml) [](https://pypi.python.org/pypi/libcnml)

It was part of the [Guifi.net Studio](https://github.com/PabloCastellano/guifinetstudio) project developed by Pablo Castellano during Google Summer of Code 2012.

What is CNML

Community Network Mark Up Language (CNML) is a project that aims to define an open ISO standard and scalable for describing mesh clouds, though it’s not limited to this kind of networks and nowadays it’s being used also in point to point infrastructure networks.

CNML is a specification based on XML, which makes it easily extendible and readable for humans besides of computers. It includes some ideas from other implementations and previous concepts like nodeXchange and SNDX.

Some advantages of using CNML is that it allows to uncouple different functionalities independently of the web application used to show the data, reducing dependence from it and its internal tables of the database.

You can read more about it on: - http://cnml.info - http://en.wiki.guifi.net/wiki/CNML

Install

You can install it by typing:

python setup.py install

or you can get it from PYPI by using pip:

pip install libcnml

Optionally you can also install lxml (read the note below):

sudo apt-get install libxml2-dev libxslt1-dev pip install lxml

libcnml is also available in the Fedora/RHEL/CentOS/Scientific Linux [repositories](https://admin.fedoraproject.org/pkgdb/package/rpms/python-libcnml/).

lxml

lxml Python library does a better memory management and is faster than minidom (default XML library in Python). If you want to manage big sets of nodes like Guifi.net World zone this definitely makes the difference.

For example, these are the results opening a Guifi.net World zone with more than 17.000 nodes: Minidom took ~23 seconds and 1,4GB RAM. Guifinetstudio window didn’t even appear. I had to reboot my laptop. Lxml took ~4s and 284MB RAM. Guifinetstudio worked, moving through the map is difficult but possible.

You can test it by your own:

$ cat cnml1.py from libcnml import * c = CNMLParser(‘tests/detail’)

$ time python cnml1.py Using lxml which is more efficient Loaded OK

real 0m3.974s user 0m3.728s sys 0m0.188s

$ time python cnml1.py lxml module not found. Falling back to minidom Loaded OK

real 0m22.984s user 0m21.997s sys 0m0.868s

License

The code license is GPLv3+