helerm - Helsinki Electronic Records Management Classification System

Prerequisites

• Python 3.4+
• PostgreSQL 9.1+

Installation

• Setup and activate a virtualenv (virtualenvwrapper is a nice tool to handle virtualenvs)

• Install required Python packages

pip install -r requirements.txt

• Create local_settings.py file in the project root and use it to override settings as needed.

• Create a database

sudo -u postgres createuser -L -R -S helerm
sudo -u postgres createdb -Ohelerm helerm

• Create database tables etc.

python manage.py migrate

• To enable Finnish translations, run

python manage.py compilemessages

• You will probably need a superuser as well

python manage.py createsuperuser

Development

• pip-tools is used to ease requirement handling. To install development packages, run

pip-sync requirements.txt dev-requirements.txt

• To start the development server, run

python manage.py runserver 127.0.0.1:8000

Admin ui will be located at http://127.0.0.1:8000/admin/

API root will be located at http://127.0.0.1:8000/v1/

Import

• First you need to import classifications and create attributes

python manage.py import_classifications data/helsinki-functions.csv
python manage.py create_attributes

• It is also possible to import attributes from an excel file

python manage.py import_attributes <excel file>

• Temporary step: the old data model requires a function object for every available function code even when there is no actual data for the function. Those initial functions can be created based on current classification by running

python manage.py create_initial_functions

• Actual data can then be imported by running

python manage.py import_data <data excel file>

• To import a function template run

python manage.py import_template <data excel file> <sheet name> [template name]

Export

• All data can be exported to a XML file by running

python manage.py export_data <xml file>

• Or using the API http://127.0.0.1:8000/export/

Python XML Schema files and Python bindings

The export uses pyxb library and needs Python bindings to be generated from XSD schema files.

The repo contains two sets of JHS XML Schema files located in data directory. In addition to original ones, there are also HKI customized versions, which are in use at least for now.

Generated bindings are included in metarecord/binding/ so the export should work out of the box.

To generate new bindings from (HKI customized) JHS schema files run

pyxbgen -u Skeema_TOS_kooste_HKI_custom.xsd --module-prefix=metarecord.binding --schema-root=data -m jhs

By default the generated bindings contain unnecessary references to local files. If the bindings are shared to somewhere, put into the repo for example, it is a good idea to remove unnecessary references by

sed -i '' 's/pyxb.utils.utility.Location([^)]*)/None/' metarecord/binding/*jhs.py