bin/update_all

@@ -2,15 +2,14 @@
 
 bin/update_representatives
 
-# grace time for pg
-sleep 120
+sleep 10
 
 bin/update_dossiers
 
-sleep 120
+sleep 10
 
-bin/update_votes
+bin/update_proposals
 
-sleep 120
+sleep 10
 
-bin/update_scores
+bin/update_votes

bin/update_dossiers

 #!/bin/bash
-set -ex
+set -x
 
-source ${OPENSHIFT_REPO_DIR}bin/lib.sh
+source bin/lib.sh
 
-pipe_download_to_command ep_dossiers.json.xz parltrack_import_dossiers
+parltrack_download_pipe ep_dossiers.json.xz parltrack_import_dossiers
+
+sleep 10
+
+francedata_download_pipe dossiers.json.gz francedata_import_dossiers

bin/update_proposals

+#!/bin/bash
+set -x
+
+source bin/lib.sh
+
+# Nothing for parltrack - proposals imported with votes
+
+francedata_download_pipe scrutins.json.gz francedata_import_scrutins

bin/update_representatives

 #!/bin/bash
-set -ex
+set -x
 
-source ${OPENSHIFT_REPO_DIR}bin/lib.sh
+source bin/lib.sh
 
-pipe_download_to_command ep_meps_current.json.xz parltrack_import_representatives
+parltrack_download_pipe ep_meps_current.json.xz parltrack_import_representatives
+
+sleep 10
+
+francedata_download_pipe parlementaires.json.gz francedata_import_representatives

bin/update_scores

-#!/bin/bash
-set -ex
-
-source ${OPENSHIFT_REPO_DIR}bin/lib.sh
-
-[ -n "$OPENSHIFT_REPO_DIR" ] && cd $OPENSHIFT_REPO_DIR
-./manage.py update_score

bin/update_votes

 #!/bin/bash
-set -ex
+set -x
 
-source ${OPENSHIFT_REPO_DIR}bin/lib.sh
+source bin/lib.sh
 
-pipe_download_to_command ep_votes.json.xz parltrack_import_votes
+parltrack_download_pipe ep_votes.json.xz parltrack_import_votes
+
+sleep 10
+
+parltrack_download_pipe ep_com_votes.json.xz parltrack_import_votes
+
+sleep 10
+
+francedata_download_pipe votes.json.gz francedata_import_votes
+
+if [ "x$1" != "xnoscores" ]; then
+	sleep 10
+	refresh_scores
+fi

core/templates/core/blocks/pagination.haml

-.pagination-block
-
-  %nav
-    %ul.pagination.pagination-sm
-      - if page_obj.has_previous
-        %li
-          %a{'href': '?={queries.urlencode}&page=1',
-             'aria-label': 'First'}
-            <i aria-hidden="true" class="fa fa-chevron-circle-left"></i>
-
-        %li
-          %a{'href': '?={queries.urlencode}&page=={page_obj.previous_page_number}',
-             'aria-label': 'Previous'}
-            <i aria-hidden="true" class="fa fa-chevron-left"></i>
-
-      - for p in page_range
-        - if p
-          - if p == page_obj.number
-            %li.active
-              %a{'href': ''}
-                {{ p }}
-          - else
-            %li
-              %a{'href': '?={queries.urlencode}&page=={p}'}
-                {{ p }}
-
-      - if page_obj.has_next
-        %li
-          %a{'href': '?={queries.urlencode}&page=={page_obj.next_page_number}',
-             'aria-label': 'Next'}
-            <i aria-hidden="true" class="fa fa-chevron-right"></i>
-
-        %li
-          %a{'href': '?={queries.urlencode}&page=={paginator.num_pages}',
-             'aria-label': 'Last'}
-            <i aria-hidden="true" class="fa fa-chevron-circle-right"></i>
-
-  %div.count
-    Number of results : {{ paginator.count }}
-    %br
-    Number of displayed results :
-    {{ paginator.per_page }}
-    (
-    - for limit in pagination_limits
-      %a{'href': '?paginate_by={{ limit }}'}
-        {{ limit }}
-      - if not forloop.last
-        \/
-    )
-    - if grid_list
-      - include 'core/blocks/grid-list.html'

core/views.py

-# coding: utf-8
-
-
-class PaginationMixin(object):
-    pagination_limits = (12, 24, 48, 96)
-
-    def get(self, *args, **kwargs):
-        self.set_paginate_by()
-        return super(PaginationMixin, self).get(*args, **kwargs)
-
-    def set_paginate_by(self):
-        if 'paginate_by' in self.request.GET:
-            self.request.session['paginate_by'] = \
-                self.request.GET['paginate_by']
-
-        elif 'paginate_by' not in self.request.session:
-            self.request.session['paginate_by'] = 12
-
-    def get_paginate_by(self, queryset):
-        return self.request.session['paginate_by']
-
-    def get_page_range(self, page):
-        pages = []
-
-        if page.paginator.num_pages != 1:
-            for i in page.paginator.page_range:
-                if page.number - 4 < i < page.number + 4:
-                    pages.append(i)
-
-        return pages
-
-    def get_context_data(self, **kwargs):
-        c = super(PaginationMixin, self).get_context_data(**kwargs)
-        c['pagination_limits'] = self.pagination_limits
-        c['paginate_by'] = self.request.session['paginate_by']
-        c['page_range'] = self.get_page_range(c['page_obj'])
-        return c
-
-
-class GridListMixin(object):
-    def set_session_display(self):
-        if self.request.GET.get('display') in ('grid', 'list'):
-            self.request.session['display'] = self.request.GET.get('display')
-
-        if 'display' not in self.request.session:
-            self.request.session['display'] = 'grid'
-
-    def get(self, *args, **kwargs):
-        self.set_session_display()
-        return super(GridListMixin, self).get(*args, **kwargs)
-
-    def get_template_names(self):
-        return [t.replace('_list', '_%s' % self.request.session['display'])
-                for t in super(GridListMixin, self).get_template_names()]
-
-    def get_context_data(self, **kwargs):
-        c = super(GridListMixin, self).get_context_data(**kwargs)
-        c['grid_list'] = True
-        return c

docs/administration.rst

@@ -10,7 +10,7 @@ Authentication in the admin backend
 If you haven't already, create a super-administrator
 account with command::
 
-    ./manage.py createsuperuser
+    memopol createsuperuser
 
 Then use this account to authenticate in the
 administration backend located in ``/admin``.

docs/api.rst

+API Documentation
+~~~~~~~~~~~~~~~~~
+
+Memopol publishes its data with a read-only REST API using `Django REST
+Framework <http://www.django-rest-framework.org/>`_.  The API is browseable and
+partially self-documented, from the ``/api`` URL.
+
+The goal of this documentation is to plug the holes in the
+automatically-generated documentation.
+
+Overview
+========
+
+The API homepage resides at the ``/api/`` URL. Each model is accessible from two
+URLs:
+
+* ``/api/<model>/`` for a (filterable) list of objects
+* ``/api/<model>/<pk>/`` to get a detailed view of a specific object
+
+When accessing the API from a web browser, data should be displayed in a
+user-friendly paginated interface, and relation fields are links to the
+corresponding API URLs (if that's not the case, you can force it by appending
+``format=api`` to the querystring).  To get raw JSON data instead, either pass
+an ``Accept: application/json`` request header or append ``format=json`` to the
+querystring.
+
+Object lists may be filtered using ``field=value`` in the querystring ; related
+fields may be used in filters with double underscores (for example
+``representative__name=John``).  Alternative operators may be available for
+some fields (for example ``score__gte=100``), amongst which:
+
+* ``lte``: matches values that are lower than or equal to the specified value
+* ``gte``: matches values greater than equal to or the specified value
+* ``icontains``: matches values that contain the specified value, ignoring case
+
+Note that fields that are made available for filtering depend on each model;
+the same goes for alternative operators (by default only exact match is
+available).
+
+Finally, note that this documentation does not show the data schema.  Feel free
+to browse the API yourself to find out how data is published.
+
+Representatives API
+===================
+
+This API provides views on representatives and mandates.
+
+Representatives
+---------------
+
+The ``/api/representatives/[<pk>/]`` endpoints give access to representatives.
+The following fields are available for filtering:
+
+* ``active``: True or False
+* ``slug`` (alt.: icontains)
+* ``id``
+* ``remote_id``
+* ``first_name``
+* ``last_name`` (alt.: icontains)
+* ``gender``: M or F
+* ``birth_place``
+* ``birth_date`` (alt.: gte, lte)
+* ``search``: searches in the ``first_name``, ``last_name`` and ``slug`` fields
+
+Mandates
+--------
+
+The ``/api/mandates/[<pk>/]`` endpoints give access to mandates. The following
+fields are available for filtering:
+
+* ``id``
+* ``group__name`` (alt.: icontains)
+* ``group__abbreviation``
+* ``search``: searches in the ``group__name`` and ``group_abbreviation`` fields
+
+Constituency
+------------
+
+The ``/api/constituencies/[<pk>/]`` endpoints give access to constituencies.
+
+Groups
+------
+
+The ``/api/groups/[<pk>/]`` endpoints give access to groups.
+
+Representatives-votes API
+=========================
+
+This API provides views on dossiers, proposals and votes.
+
+Dossiers
+--------
+
+The ``/api/dossiers/[<pk>/]`` endpoints give access to dossiers. The following
+fields are available for filtering:
+
+* ``fingerprint``
+* ``title`` (alt.: icontains)
+* ``reference`` (alt.: icontains)
+* ``search``: searches in the ``fingerprint``, ``title``, ``reference``,
+  ``text`` and ``proposals__title`` fields
+
+Proposals
+---------
+
+The ``/api/proposals/[<pk>/]`` endpoints give access to proposals. The following
+fields are available for filtering:
+
+* ``fingerprint``
+* ``dossier__fingerprint``
+* ``title`` (alt.: icontains)
+* ``description`` (alt.: icontains)
+* ``reference`` (alt.: icontains)
+* ``datetime`` (alt.: gte, lte)
+* ``kind``
+* ``search``: searches in the ``fingerprint``, ``title``,
+  ``dossier__fingerprint``, ``dossier__title`` and ``dossier__reference`` fields
+
+Votes
+-----
+
+The ``/api/votes/[<pk>/]`` endpoints give access to votes. The following fields
+are available for filtering:
+
+* ``proposal__fingerprint``
+* ``position``
+* ``representative_name`` (alt.: icontains)
+* ``representative``
+
+
+Memopol API
+===========
+
+This API provides views on recommendations and representative scores.
+
+Recommendations
+---------------
+
+The ``/api/recommendations/[<pk>/]`` endpoints give access to recommendations.
+The following fields are available for filtering:
+
+* ``id``
+* ``recommendation``
+* ``title`` (alt.: icontains)
+* ``description`` (alt.: icontains)
+* ``weight`` (alt.: gte, lte)
+* ``search``: searches in the ``title`` and ``description`` fields
+
+Vote Scores
+-----------
+
+The ``/api/vote_scores/[<pk>/]`` endpoints give access to scored votes; that is,
+representative votes with their contribution to the representative score. Only
+votes that match a recommendation are visible using this endpoint.  The
+following fields are available for filtering:
+
+* ``representative``
+* ``proposal``
+* ``proposal__dossier``
+
+Dossier Scores
+--------------
+
+The ``/api/dossier_scores/[<pk>/]`` endpoints give access to dossier scores;
+that is, the contribution of each dossier on a representative score. Only
+dossiers with recommendations are visible using this endpoint.  The following
+fields are available for filtering:
+
+* ``dossier``
+* ``representative``
+* ``score`` (alt.: gte, lte)
+
+Representative scores
+---------------------
+
+The ``/api/scores/[<pk>/]`` endpoints give access to total scores for each
+representative. The following fields are available for filtering:
+
+* ``representative``
+* ``score`` (alt.: gte, lte)

docs/conf.py

@@ -113,7 +113,7 @@ html_theme = 'alabaster'
 #html_theme_options = {}
 
 # Add any paths that contain custom themes here, relative to this directory.
-#html_theme_path = []
+html_theme_path = ['../memopol_env/local/lib/python2.7/site-packages']
 
 # The name for this set of Sphinx documents.  If None, it defaults to
 # "<project> v<release> documentation".

docs/deploy-custom.rst

+Deployment on a custom machine
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Prerequisites
+=============
+
+You will need the following:
+
+* A clone of the git repository at ``https://git.laquadrature.net/memopol/memopol.git``
+* A dedicated python virtualenv
+* a PostgreSQL>=9.2 database (not necessarily on the same machine)
+* a WSGI-capable web server
+
+Setup environment
+=================
+
+Set the following environment variables::
+
+    DJANGO_SETTINGS_MODULE=memopol.settings
+
+Customize settings
+==================
+
+Create a copy of the example local settings file::
+
+    $ cp src/memopol/local_settings.py.example src/memopol/local_settings.py
+
+Edit ``src/memopol/local_settings.py`` to set directories, database settings and
+allowed hosts.  Setup your WSGI server to serve:
+
+* Static files from the directory specified in the ``PUBLIC_DIR`` setting to the
+  ``/static`` URL
+* The ``src/memopol/wsgi.py`` WSGI application
+
+Initial memopol setup
+=====================
+
+From the repository root, install python dependencies (you may want to do that
+in a virtualenv)::
+
+    $ pip install -U pip setuptools
+    $ pip install -Ue .
+
+Install client libraries::
+
+    $ src/memopol/bin/install_client_deps.sh
+
+Setup the database schema::
+
+    $ memopol migrate --noinput
+
+Collect static files::
+
+    $ memopol collectstatic --noinput
+
+Memopol should be ready to go.
+
+Updating
+========
+
+To update simply pull the repository and run setup commands again::
+
+    $ git pull
+    $ pip install -Ue .
+    $ src/memopol/bin/install_client_deps.sh
+    $ memopol migrate --noinput
+    $ memopol collectstatic --noinput
+
+Data provisionning
+==================
+
+Set up two cron jobs:
+
+* One to update data from parliaments, that runs ``bin/update_all``.  This
+  script takes quite some time to run, so you should schedule it once every
+  night for example
+* One to refresh scores,  that runs ``memopol refresh_scores``.  This one
+  runs quite quickly (a few seconds), you may want to run it after the update
+  job has completed (but you can run it more often).
+
+Ensure that cron jobs get the same environment as the application.
+
+
+Continue to :doc:`administration`.

docs/deployment.rst

-Deployment on OpenShift
-~~~~~~~~~~~~~~~~~~~~~~~
-
-OpenShift is an Open-Source Platform-as-a-Service software by Red Hat. It is
-also available in its hosted version known as "OpenShift Online" and the first
-three websites ("gears") are free.
-
-Clone the repository
-====================
-
-You should fork the project on github and use the fork's clone url. For the
-sake of the demo, we'll use the main repository URL::
-
-    $ git clone https://github.com/political-memory/political_memory.git
-    Cloning into 'political_memory'...
-    remote: Counting objects: 2516, done.
-    remote: Compressing objects: 100% (109/109), done.
-    remote: Total 2516 (delta 44), reused 0 (delta 0), pack-reused 2402
-    Receiving objects: 100% (2516/2516), 4.40 MiB | 79.00 KiB/s, done.
-    Resolving deltas: 100% (1103/1103), done.
-    Checking connectivity... done.
-
-    $ cd political_memory/
-
-Create your own branch, ie::
-
-    $ git checkout -b yourbranch origin/pr
-    Branch yourbranch set up to track remote branch pr from origin.
-    Switched to a new branch 'yourbranch'
-
-Create an app on OpenShift
-==========================
-
-To deploy the website, use a command like::
-
-    $ rhc app-create \
-        python-2.7 \
-        cron-1.4 \
-        postgresql-9.2 \
-        -a yourappname \
-        -e OPENSHIFT_PYTHON_WSGI_APPLICATION=memopol/wsgi.py \
-        --from-code https://github.com/political-memory/political_memory.git \
-        --no-git
-
-This should create an app on openshift. Other commands would deploy it at once
-but in this tutorial we're going to see how to manage it partly manually for
-development.
-
-Add the git remote created by OpenShift
-=======================================
-
-Add the git remote openshift created for you, you can see it with
-``rhc app-show``, ie.::
-
-    $ rhc app-show -a yourappname
-    [snip]
-    Git URL:         ssh://569f5cf500045f6a1839a0a4@yourappname-yourdomain.rhcloud.com/~/git/yourappname.git/
-    Initial Git URL: https://github.com/political-memory/political_memory.git
-    SSH:             569f5cf500045f6a1839a0a4@yourappname-yourdomain.rhcloud.com
-    [snip]
-
-    $ git remote add oo_yourappname ssh://569f5cf500045f6a1839a0a4@yourappname-yourdomain.rhcloud.com/~/git/yourappname.git/
-
-Activate OpenShift's git post-recieve hook
-==========================================
-
-Activate OpenShift's post-receive hook on your branch::
-
-    $ rhc app-configure -a yourappname --deployment-branch yourbranch
-
-Deploy your branch
-==================
-
-OpenShift will deploy when it receives commits on the deployment branch, to
-deploy just do::
-
-    $ git push oo_yourappname yourbranch
-
-If something goes wrong and you want to retry, use the ``rhc app-deploy``
-command, ie::
-
-    $ rhc app-deploy yourbranch -a yourappname
-
-Data provisionning
-==================
-
-To fill up the representatives database table, either wait for the cron script
-to be executed, either do it manually::
-
-    $ rhc ssh -a yourappname 'cd app-root/repo/ && bin/update_all'
-
-OpenShift is fun, login with ssh and look around if you're curious, you'll be
-able to recreate your app without much effort if you break it anyway.
-
-Continue to :doc:`administration`.

docs/development.rst

 Local development tutorial
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-.. warn:: I reverse-engineered this from the source code I inherited, I might
-          not be doing the right way nor be able to defend all of technical
-          decisions.
-
 This tutorial drives through a local installation of the project for
-development on Linux. It requires git, a fairly recent version of nodejs (see
-:file:`.openshift/action_hooks/deploy` for a way to install it), python2 and
-virtualenv.
+development on Linux. It requires git, a fairly recent version of python2,
+virtualenv and PostgreSQL.
+
+Setup the database
+==================
+
+Memopol requires PostgreSQL 9.1 or higher.  It used to run with SQLite, too, but
+that is no longer the case. It is better to install and configure you're local
+PostgreSQL server before starting to install Memopol.
+
+On Debian
+---------
+
+To setup you're PostgreSQL database on a debian stable distribution, you can use
+the package manager apt::
+
+  $ apt install postgresql postgresql-server-dev-9.X
+
+Then you need to create the 'memopol'  user and the 'memopol' database::
+
+  # To have a root access on Postgres, you need to connect as user 'postgres'
+  $ su - postgres
+  $ psql -c "create user memopol with password 'memopol';"
+  $ psql -c "alter role memopol with createdb;"
+  $ psql -c "create database memopol with owner memopol;"
+  $ exit
+
+You're database is now setup for Memopol. You can now launch the 'quickstart.sh'
+script to automatically install all the components or do it manually.
+
+In General
+----------
+
+Make sure the corresponding user and database exist on your system; the user
+will need the 'createdb' permission in order to be able to run tests.  To create
+them, you may use the following commands::
+
+    $ psql -c "create user memopol with password 'memopol';" -U postgres
+    $ psql -c "alter role memopol with createdb;" -U postgres
+    $ psql -c "create database memopol with owner memopol;" -U postgres
+
+
+Automatic Install
+=================
+
+There is a quickstart script used and tested manually by some of the
+developers. Feel free to try it, but don't worry if it doesn't work for you
+then you can do each install step manually, which is recommended because well
+that's how you will learn most.
+
+Here's how to try it::
+
+    $ git clone gitlab@git.laquadrature.net:memopol/memopol.git
+    $ cd memopol
+    $ source bin/quickstart.sh
+
+At this point, you should now run the development server and access to Memopol::
+
+  $ memopol runserver
+
+If you want more control or if it doesn't work for you, then follow the steps
+below or have a look at what the quickstart script does.
+
+Development helper
+===================
+
+You can run the script 'bin/dev.sh' to automaticaly setup some aliases. It works
+only with Bash and Zsh.
+
+The script build a custom file named '.memopol.alias' at the root of the project
+containing all the aliases for memopol. All the path to the project are build
+automatically. A single line is added to your '$HOME/.bashrc' or '$HOME/.zshrc'
+to source the aliases.
+
+After execute 'bin/dev.sh' you should close the current terminal and open
+another one to have access to the aliases.
+
+There is a quick list of available aliases::
+
+  memopol-code : Go into the repository and activate virtualenv and
+  set Django in debug mode
+  memopol-launch : Run the development server echo
+  memopol-update-all : Get all the production data
+  memopol-refresh-scores : Refresh all scores
+
+.. warning:: If you are using multiple setup of Memopol, it is not recommended to
+  use this script.
+
+If you need to change the location of the project, you should remove this line
+from your .bashrc or .zshrc::
+
+  source $PATH_TO_THE_PROJECT/.memopol.alias
 
 Make a virtual environment
 ==========================
@@ -28,26 +113,32 @@ Create a python virtual environment and activate it::
 
     $ source memopol_env/bin/activate
 
+Alternatively, use the tox command::
+
+    $ tox -e py27
+    $ source .tox/py27/bin/activate
+
 Clone the repository
 ====================
 
-You should fork the project on github and use the fork's clone url. For the
-sake of the demo, we'll use the main repository URL::
+The project is hosted on https://git.laquadrature.net/memopol/memopol
+
+You can get the code with git ::
 
-    $ git clone https://github.com/political-memory/political_memory.git
-    Cloning into 'political_memory'...
-    remote: Counting objects: 2516, done.
-    remote: Compressing objects: 100% (109/109), done.
-    remote: Total 2516 (delta 44), reused 0 (delta 0), pack-reused 2402
-    Receiving objects: 100% (2516/2516), 4.40 MiB | 79.00 KiB/s, done.
-    Resolving deltas: 100% (1103/1103), done.
-    Checking connectivity... done.
+    $ git clone https://git.laquadrature.net/memopol/memopol
+    Clonage dans 'memopol'...
+    remote: Counting objects: 7972, done.
+    remote: Compressing objects: 100% (2668/2668), done.
+    remote: Total 7972 (delta 5203), reused 7830 (delta 5099)
+    Réception d'objets: 100% (7972/7972), 4.88 MiB | 4.73 MiB/s, fait.
+    Résolution des deltas: 100% (5203/5203), fait.
+    Vérification de la connectivité... fait.
 
-    $ cd political_memory/
+    $ cd memopol/
 
 Create your own branch, ie::
 
-    $ git checkout -b yourbranch origin/pr
+    $ git checkout -b yourbranch
     Branch yourbranch set up to track remote branch pr from origin.
     Switched to a new branch 'yourbranch'
 
@@ -57,22 +148,22 @@ Install Python dependencies
 Then, install the package for development::
 
     $ pip install -e .
-    Obtaining file:///tmp/political_memory
-    Collecting django (from political-memory==0.0.1)
+    Obtaining file:///tmp/memopol
+    Collecting django (from memopol==0.0.1)
       Using cached Django-1.9-py2.py3-none-any.whl
 
     [output snipped for readability]
 
-    Installing collected packages: django, sqlparse, django-debug-toolbar, django-pdb, six, django-extensions, werkzeug, south, pygments, markdown, hamlpy, django-coffeescript, ijson, python-dateutil, pytz, political-memory
-      Running setup.py develop for political-memory
-    Successfully installed django-1.9 django-coffeescript-0.7.2 django-debug-toolbar-1.4 django-extensions-1.5.9 django-pdb-0.4.2 hamlpy-0.82.2 ijson-2.2 markdown-2.6.5 political-memory pygments-2.0.2 python-dateutil-2.4.2 pytz-2015.7 six-1.10.0 south-1.0.2 sqlparse-0.1.18 werkzeug-0.11.2
+    Installing collected packages: django, sqlparse, django-debug-toolbar, django-pdb, six, django-extensions, werkzeug, south, pygments, markdown, hamlpy, django-coffeescript, ijson, python-dateutil, pytz, memopol
+      Running setup.py develop for memopol
+    Successfully installed django-1.9 django-coffeescript-0.7.2 django-debug-toolbar-1.4 django-extensions-1.5.9 django-pdb-0.4.2 hamlpy-0.82.2 ijson-2.2 markdown-2.6.5 memopol pygments-2.0.2 python-dateutil-2.4.2 pytz-2015.7 six-1.10.0 south-1.0.2 sqlparse-0.1.18 werkzeug-0.11.2
 
 Install client dependencies
 ===========================
 
 We'll also need to download client libraries::
 
-    $ bin/install_client_deps.sh
+    $ src/memopol/bin/install_client_deps.sh
     * Downloading jquery/jquery (2.1.4) from Github...
     * Downloading FortAwesome/Font-Awesome (v4.3.0) from Github...
     * Downloading lipis/flag-icon-css (0.7.1) from Github...
@@ -83,45 +174,20 @@ Activate ``DJANGO_DEBUG``
 =========================
 
 ``DEBUG`` is disabled by default, the development server
-won't run properly by default thnen, to enable it export
+won't run properly by default then, to enable it export
 the ``DJANGO_DEBUG`` variable in the current shell::
 
     $ export DJANGO_DEBUG=True
 
-Run the development server
-==========================
-
-Run the development server::
-
-    $ ./manage.py runserver
-
-    Performing system checks...
-
-    System check identified no issues (0 silenced).
-    December 09, 2015 - 21:26:47
-    Django version 1.8.7, using settings 'memopol.settings'
-    Starting development server at http://127.0.0.1:8000/
-    Quit the server with CONTROL-C.
-    [09/Dec/2015 21:26:48] "GET / HTTP/1.1" 200 13294
-
-The website is running on ``http://127.0.0.1:8000/``.
 
 Database migrations
 ===================
 
-The repo comes with a pre-configured SQLite db with sample data so that you can
-start hacking right away. However, if you were to use a local postgresql
-database ie. with this sort of environment::
+Database migrations ensure the database schema is up to date with the project.
+If you're not sure, you can run them anyway, they won't do any harm.  Use the
+following command::
 
-    export DJANGO_DATABASE_DEFAULT_NAME=memopol
-    export DJANGO_DATABASE_DEFAULT_USER=postgres
-    export DJANGO_DATABASE_DEFAULT_ENGINE=django.db.backends.postgresql_psycopg2
-    export DJANGO_DEBUG=1
-    export DJANGO_SETTINGS_MODULE=memopol.settings
-
-Then you could run database migrations::
-
-    $ ./manage.py migrate
+    $ memopol migrate
     Operations to perform:
       Synchronize unmigrated apps: django_filters, staticfiles, datetimewidget, autocomplete_light, messages, adminplus, compressor, humanize, django_extensions, constance, bootstrap3
       Apply all migrations: legislature, votes, database, admin, positions, sessions, representatives, auth, contenttypes, representatives_votes, taggit
@@ -140,13 +206,34 @@ Then you could run database migrations::
 Provision with data
 ===================
 
-Again, the repo comes with a pre-configured SQLite db with sample data so that
-you can start hacking right away. However, you could still reload sample data::
+You can load a small data sample for quick setup:
+
+    $ memopol loaddata small_sample.json
+
+If you launch memopol for the first time, you need to launch this command :
 
-    $ ./manage.py loaddata memopol/fixtures/small_sample.json
+    $ memopol refresh_scores
 
 Or actual data (takes a while)::
 
     $ bin/update_all
 
+Run the development server
+==========================
+
+Run the development server::
+
+    $ memopol runserver
+
+    Performing system checks...
+
+    System check identified no issues (0 silenced).
+    December 09, 2015 - 21:26:47
+    Django version 1.8.7, using settings 'memopol.settings'
+    Starting development server at http://127.0.0.1:8000/
+    Quit the server with CONTROL-C.
+    [09/Dec/2015 21:26:48] "GET / HTTP/1.1" 200 13294
+
+The website is running on ``http://127.0.0.1:8000/``.
+
 Continue to :doc:`administration`.

docs/hacker.rst

 Hacker guide
 ~~~~~~~~~~~~
 
+See a `hacking demo on the Memopol project in some epic
+slides
+<https://slides.com/jamespic/cd-devops/fullscreen#/>`_.
+
+Read about it in `Continuous Delivery and DevOps
+quickstart
+<https://www.packtpub.com/application-development/continuous-delivery-and-devops-%E2%80%93-quickstart-guide-second-edition)>`_,
+and I bet you'll order a paperback edition for reference !
+
+Testing
+=======
+
+Use the ``tox -l`` command to list tests::
+
+    $ pip install tox
+    $ cd memopol/
+    $ tox -l
+
+Use the ``tox -e`` command to execute a particular test suite::
+
+    $ tox -e py27
+
+And use the ``tox`` command without argument to execute all test suites,
+exactly like in CI.
+
 Adding random recommendations
 =============================
 
 ::
 
-    $ ./manage.py shell
+    $ memopol shell
     In [1]: from representatives_votes.models import Proposal
     In [2]: from votes.models import Recommendation
     In [3]: import random
@@ -23,14 +48,14 @@ test run is.
 To create test fixtures for representatives_positions, insert some Position
 objects, and reduce the database with::
 
-    ./manage.py remove_representatives_without_position
-    ./manage.py remove_groups_without_mandate
-    ./manage.py remove_countries_without_group
+    memopol remove_representatives_without_position
+    memopol remove_groups_without_mandate
+    memopol remove_countries_without_group
 
 For representatives_recommendations::
 
-    ./manage.py remove_proposals_without_recommendation
-    ./manage.py remove_dossiers_without_proposal
-    ./manage.py remove_representatives_without_vote
-    ./manage.py remove_groups_without_mandate
-    ./manage.py remove_countries_without_group
+    memopol remove_proposals_without_recommendation
+    memopol remove_dossiers_without_proposal
+    memopol remove_representatives_without_vote
+    memopol remove_groups_without_mandate
+    memopol remove_countries_without_group