Data generator for the-law-factory project
Python Perl Shell HTML
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.


Build Status Coverage Status

Data generator for the-law-factory project (

Code used to generate the API available at:

Install the dependencies

You should set up a dedicated virtualenv with Python 3.5+:

virtualenv -p $(which python3) venv
source venv/bin/activate

Using Pypy can seriously boost performance. You can easily install it and create a virtualenv with it for instance by installing Pyenv:

pyenv install pypy3.5-6.0.0
pyenv virtualenv pypy3.5-6.0.0 lafabrique
pyenv activate lafabrique

Then with your choice of virtualenv activated, install the dependencies:

pip install --upgrade setuptools pip # not necessary but always a good idea
pip install --process-dependency-links -e .

Generate data for one bill

tlfp-parse <url>

The data is generated in the "data" directory. You can change this default behavior by inputting a data path as extra argument: tlfp-parse <url> <dataDir>.

For example, to generate data about the "Enseignement supérieur et recherche" bill:

ls data/pjl12-614/

You can also use directly Senate's ids such as: tlfp-parse pjl12-614

Development options --debug, --enable-cache and --only-promulgated can also be used.

Generate data for many bills

To generate all bills from 2008, you can pipe a list of ids or urls into tlfp-parse-many.

A convenient way to do so is to use senapy:

senapy-cli doslegs_urls --min-year=2008 | tlfp-parse-many data/

See senapy-cli doslegs_urls help for more options. You can also use anpy with anpy-cli doslegs_urls.

Serve bills locally for The Law Factory website

First, you need to build data for all desired bills.

Then generate the files required by the frontend:

python tlfp/ data/       # generates home.json and dossiers_promulgues.csv used by the searchbar
python tlfp/tools/ data/   # generates dossiers_n.json files used by the Navettes viz

Finally, serve the data directory however you like. For instance, you can serve it on a specific port with a simple http server like nodeJs', in which case, you'll need to enable cors: just install http-server with npm and run it in data directory on a given port (8002 in the example):

npm install -g http-server
cd data & http-server -p 8002 --cors

Generate git version for a bill

Work In Progress

You can export all your bills as git repositories: python tlfp/tools/ git_export

Other things you can do

  • parse a sénat dosleg: senapy-cli parse pjl15-610
  • parse all the sénat doslegs: senapy-cli doslegs_urls | senapy-cli parse_many senat_doslegs/
  • parse all the AN doslegs anpy-cli doslegs_urls | anpy-cli parse_many_doslegs an_doslegs/
  • parse an AN dosleg: anpy-cli show_dossier_like_senapy
  • generate a graph of the steps: python tlfp/tools/ data/ | dot -Tsvg > steps.svg


To run the tests, you can follow the .travis.yml file.

git clone
python tests/ the-law-factory-parser-test-cases

If you modify something, best is to regenerate the test-cases with the --regen flag:

python tests/ the-law-factory-parser-test-cases --regen

To make the tests faster, you can also use the --enable-cache flag.

You can also watch for parts of the code not yet covered by the tests:

  • First, install coverage: pip install coverage
  • Then, you can execute bash
  • Then, the report is in htmlcov/index.html


This work, a collaboration between Regard Citoyens, médialab Sciences Po and CEE Sciences Po, is supported by a public grant overseen by the French National Research Agency (ANR) as part of the "Investissements d'Avenir" program within the framework of the LIEPP center of excellence (ANR11LABX0091, ANR 11 IDEX000502).

More details at