Open Data Census is a web application that supports a submission and reviews workflow to collect information on the state of open data.
Some presentation of data is supported, along with partitioning results by year.
The code base supports multiple censuses in a multi-tenant configuration, where each tenant runs a census from a subdomain.
Tenant administrators can customize parts of the app, from look and feel to key texts on various views.
If you want to check out what an Open Data Census site looks like we have a demo site running at:
Open Data Census is a Node.js app, running Express v4 and Postgres 9.4 for the database.
Get a local server setup with the following steps:
NOTE: If you need to prefix your commands in your local environment with sudo
, then do that.
- Install Postgres 9.4 on your machine.
- Setup to appropriate credentials on Google and Facebook so they are OAuth providers for your app.
- For Google: follow these steps
- The callBack url for Google+ API is:
http://id.dev.census.org:5000/google/callback
- The callBack url for Google+ API is:
- For Facebook: follow these steps
- The callBack url for Facebook is:
http://id.dev.census.org:5000/facebook/callback
- The callBack url for Facebook is:
- For Google: follow these steps
- Ensure you are running the supported version of Node.js, which is declared in the 'engines' section of
package.json
. - Create a database with
createdb opendatacensus
. - Add this line to your hosts file:
127.0.0.1 demo.dev.census.org global.dev.census.org id.dev.census.org system.dev.census.org
. - Create a local directory called
opendatasurvey
and move into it withcd opendatasurvey
. - Clone the code with
git clone https://github.com/okfn/opendatasurvey .
. - Install the dependencies with
npm install
. - Create a copy of
settings.json.example
file and name itsettings.json
changing any values as required.
Now we should be ready to run the server:
- Run the app with
npm start
(the server will be run on the 5000 port) - Log in at
http://id.dev.census.org:5000/login
with your admin account (the same that was setup on the settings.json file) - Load registry and config data at
http://system.dev.census.org:5000/control
- Load the data for a specific site, e.g.:
http://demo.dev.census.org:5000/admin
- Visit the site:
http://demo.dev.census.org:5000/
Other things you can do:
- Run the test suite with
npm test
- Check your code style with
npm run lint
using the eslint config/.eslintrc
.
Most of the site configuration is taken from config sheets in Google Sheets. You can use this registry sheet and its linked sheets as examples and clone them as necessary.
NOTE: Ensure your registry and all other config sheets have been published as CSV in Google Sheets (click File, Publish to the Web).
We run deployments on Heroku. The app should run anywhere that you can run Node.js and Postgres. The important thing to remember for deployments is that the settings.json
file you are using for local development is not available, and therefore you need to configure several settings via environment variables. The key settings you should ensure are set:
SESSION_SECRET
BASE_DOMAIN
DATABASE_URL
SYS_ADMIN
FACEBOOK_APP_ID
FACEBOOK_APP_SECRET
GOOGLE_APP_ID
GOOGLE_APP_SECRET
When templates or strings in core code change, the translations have to be changed. Extract strings to the messages.pot
file by running this command:
./node_modules/.bin/gulp pot
You will need the GNU gettext commands. See here for more information.
To update the existing .po files, run:
./node_modules/.bin/gulp update-po
To add a new language, create the directory locale/[language-code]/LC_MESSAGES
and create the translation files (*.po). Alternatively, you can copy the locale/en
directory to locale/[language-code]
and change existing files.
To update the translations cache, run:
./node_modules/.bin/gulp compile-po
Any column can be internationalised by adding another column with @locale
after it. For example, the description
column can be translated to German by adding a column of description@de
. Only languages which have template translations created for them are valid. The locales
setting in the config document can be used to restrict the number of locales available. The first locale in the list is the default locale.
A script is provided to remove users by provider id. Removing a user will update entries where the user is a submitter or reviewer (changing the user id to the anonymous user), then remove the user.
$ npm run anonymize_user -- <provider-id>
Use -d
to perform a dry run, which will print the user id, and entry ids which will be affected (where the user is either a submitter or reviewer).
createdb opendatacensus_test
npm test