Scripts for Decidim

Preparation:

Backup a Decidim Database (ie: Decidim preproduction in Heroku):

heroku pg:backups:capture
heroku pg:backups:download

Import in local for testing:

Create database if needed:

cd decidim-barcelona-preprod
bin/rails db:create

Import (repeat every time to re-test):

pg_restore --verbose --clean --no-acl --no-owner -h localhost -U decidim -d decidim-barcelona_development latest.dump

Change the host for testing in localhost:

bin/rails c
o=Decidim::Organization.first
o.host="localhost"
o.save

Importing massive answers for proposals

This script answers a list of proposals, it requires an excel with headers with this format:

ID, status, text_ca, text_es, ...
123, accepted, una resposta, una respuesta, ... 

or

id, state, answer/ca, answer/es, ...
123, acceptada, una resposta, una respuesta, ... 

NOTE:

• Component setting "Publish proposal answers immediately" is honored, so, if activated, emails and notifications will be sent. Otherwise will just introduce the answers in the database.
• Answers are are parsed to highlight links automatically and convert line breaks to <br>;

Usage:

Copy the script into your lib/tasks of your Decidim installation:

cd lib/tasks
wget -qO- https://github.com/Platoniq/decidim-scripts/archive/0.7.tar.gz | tar --transform 's/^decidim-scripts-0.7//' -xvz

Run in local (as testing):

bin/rails "proposals:batch:answer[admin@example.org,./proposal-answers.csv]"

Massively geolocating proposals

This script geolocates a list of proposals, it requires an excel with headers with this format:

ID, address, ...
123, some place, ... 

or

i, adreça, ...
123, some place, ... 

NOTE:

• Component setting "Geocoding enabled" is honored, so, if not activated, proposals won't be modified.

bin/rails proposals:batch:geoloc[admin@example.org,./proposal-geolocs.csv]

Running in heroku:

In Heroku you need to temporary copy the relevant files into a running container first. As there's no "copy" command to do that you can use the service https://transfer.sh with this trick:

1. From your computer, send the files to transfer.sh:

curl --upload-file proposal-answers.csv https://transfer.sh/proposal-answers.csv -H "Max-Days: 1"
curl --upload-file proposal-geolocs.csv https://transfer.sh/proposal-geolocs.csv -H "Max-Days: 1"

Which will return (for instance) the download addresses:

https://transfer.sh/YwYAN/proposal-answers.csv
https://transfer.sh/x3hUa/proposal-geolocs.csv

2. Login into a shell session in heroku

heroku run bash

3. Download the files inside the dyno:

wget https://transfer.sh/YwYAN/proposal-answers.csv
wget https://transfer.sh/x3hUa/proposal-geolocs.csv

4. Download the scripts into the lib/tasks folder:

wget -qO- https://github.com/Platoniq/decidim-scripts/archive/0.7.tar.gz | tar --transform 's/^decidim-scripts-0.7//' -xvz -C lib/tasks

5. Run the script inside the shell session (2nd terminal):

bin/rails proposals:batch:answer
bin/rails proposals:batch:geoloc

or:

bin/rails proposals:batch:answer[admin@example.org,./proposal-answerss.csv]
bin/rails proposals:batch:geoloc[admin@example.org,./proposal-geolocs.csv]
bin/rails proposals:batch:cost[admin@example.org,./proposal-costs.csv]

NOTE: in case the program runs out of memory in Redis, just execute it again after a minute or so to allow for the queue to empty.