pyaxis

pyaxis is a python library for PC-Axis (or PX) formatted data manipulation which allows reading and writing PC-Axis [1] format with python, using the DataFrame structures provided by the widely accepted pandas library [2]. PX is a standard format for statistical files used by a large number of statistical offices (Spain, Switzerland, Sweden). A package of software has been developed for this format: PX-Win, PX-Web and PX-Edit. pyaxis eases reading and parsing of PX files data and metadata into a pandas Dataframe and a dict structure, allowing their manipulation in a tabular and pythonic manner. It supports both monolingual and multilingual PX files. The package was initially developped by the [Instituto Cántabro de Estadistica](https://www.icane.es) and is now jointly maintained by ICANE and the [Swiss Data Science Center](https://datascience.ch). pyaxis is provided under the Apache License 2.0.

[1]	https://www.scb.se/px-en for PC-Axis information

[2]	https://pandas.pydata.org for Python Data Analysis Library information

Installation

For installation:

pip install pyaxis

Usage

From PX to pandas DataFrame

Typical usage often looks like this:

from pyaxis import pyaxis

EXAMPLE_URL = 'https://www.ine.es/jaxiT3/files/t/es/px/2184.px'

px = pyaxis.parse(EXAMPLE_URL, encoding='ISO-8859-2')
print(px['DATA'])
print(px['METADATA'])

For Multilingual PX files

Typical usage often looks like this:

from pyaxis import pyaxis

EXAMPLE_URL = 'https://www.pxweb.bfs.admin.ch/DownloadFile.aspx?file=px-x-0602000000_107'

# Swiss PX files are in German(de), French(fr), Italian(it), and English(en)
px = pyaxis.parse(EXAMPLE_URL, encoding='ISO-8859-2', lang='fr')
print(px['DATA'])
print(px['METADATA'])
print(px['TRANSLATION'])