Orange3-HXLvisualETL 0.3.0rc1

Humanitarian Exchange Language (HXL) visual Extract, Transform, Load (ETL)

HXL visual ETL (Orange3 add-on)

This is an early draft of Orange3 add-on with minimal awareness of data labeled with HXL.

Features

Data Vault Conf

[WORKING DRAFT] Configure active local data vault configurations. This allows overriding defaults.

Download Raw File

Download remote resource into a local FileRAW

Unzip Raw File

[WORKING DRAFT] Unzip (zip, gzip, bzip, ...) an FileRAW into an FileRAWCollection

Select Raw File

[DRAFT] From a local FileRAWCollection, select an FileRAW

Load Raw File

Convert a local FileRAW into Orange3 Data / DataFrame. Required to allow use with other widgets.

Supported features (*):

• pandas.read_table
• pandas.read_csv
• pandas.read_excel
• pandas.read_feather
• pandas.read_fwf
• pandas.read_html
• pandas.read_json
• pandas.json_normalize
• pandas.read_orc
• pandas.read_parquet
• pandas.read_sas
• pandas.read_spss
• pandas.read_stata
• pandas.read_xml

(*) Some features will require additional python packages which are not installed by default with this add-on. The user will be warned about this.

Statistical Role

Change statistical role (the "feature", "target", "meta", "ignore") using HXL patterns instead of stric exact names for the data variables.

Data Type

[DRAFT] Change the computational data type (the "numeric", "categorical" "text", "datetime") using HXL patterns instead of stric exact names for the data variables.

HXL short names

[EARLY DRAFT] Make HXLated input data with shorter variable names.

RAW Info

[DRAFT] Inspect a FileRAW or FileRAWCollection

Installation

To install the add-on from source run

pip install .

To register this add-on with Orange, but keep the code in the development directory (do not copy it to Python's site-packages directory), run

pip install -e .

Documentation / widget help can be built by running

make html htmlhelp

from the doc directory.

Usage

After the installation, the widget from this add-on is registered with Orange. To run Orange from the terminal, use

orange-canvas

or

python -m Orange.canvas

The new widget appears in the toolbox bar under the section Example.

TODOs

• https://github.com/fititnt/orange3-hxl/issues/1
• (...)