git clone [email protected]:stats4sd/ukrainesurvey.git cd ukrainesurvey

Copy the file config.yml.exemple into config.yml, and change the information in this file according to your own connection parameters. Note that this is a YAML file, so indentations and spaces are important!

Two 'blocks' are needed in the config file:

• "mysql" -> containing details of the local mysql connection
• "kobo" -> containing details of the kobotoolbox user account used to host the forms

R also requires a "default" block, even though we don't reference it in the code.

1. Open the ukraine tools.Rproj in R Studio.
2. It will probably give a warning: "Packrat is not installed in the local library -- attempting to bootstrap an installation...". It may take a minute or so to load fully, but once you see the Files, Packages sidebar etc, you're ready to go.
3. run packrat::restore(). This will read the packrat/packrat.lock file and attempt to install the required packages locally. It will likely take several minutes.
4. You should then be able to run the app through RStudio!

We use python to pull data from Kobotools into the database and to generate fake data for testing.

1. Install Python3.7+
2. Run pip3 install -r requirements.txt to make sure you have all the packages needed.

Restore the database in Mysql/ukrainedb.sql to the database specified in yøur config.yml:

mysql -u {username} -p {database-name} < Mysql/ukrainedb.sql

Then run the update_sql_views script to ensure you have the latest version of all MySQL Views:

python3 deployement/update_sql_views.py

Generate buildings with dwellings within each cluster python3 Data/building_generator.py

You should now have a working platform where each cluster has a set of buildings and dwellings ready for sampling.