README.md

PeerScout

Configuration

AWS Credentials and Configuration

Create ~/.aws/credentials and populate the credentials, e.g.:

[default]
aws_access_key_id=<key id>
aws_secret_access_key=<access key>

You may also create ~/.aws/config:

[default]
region=<region>

App Configuration

Copy app-example.cfg to app.cfg and make the necessary configuration changes.

When using Docker for development, the configuration file is app-dev.cfg instead. It would very much have the same config but the database should point to the db service (defined in docker-compose.override.yml).

Development with Docker

Pre-requisites (with Docker)

• Docker and Docker Compose

Create / Migrate Schema (with Docker)

make migrate-schema

Populate Database (with Docker)

This will download updates from AWS and populate the database with it. This may take a while. It will also download data from Crossref, which might fail, which is 'okay'.

make update-data-and-reload

Data Directory

Downloaded files and cache are stored to the .data directory. Which in development is mounted to the host. To make the directory writable by the Docker user (elife) call (writable to all users):

make fix-data-permissions

Start

make start

Development without Docker

Pre-requisites (without Docker)

• Python 3 (including dev dependencies, e.g. python3-dev on Ubunutu / Debian)
• system dev dependencies (e.g. build-essential)
• PostgreSQL (although SQLite may work too)
• Node.js and npm

Setup (without Docker)

Python Dependencies

A) Using Local Virtual Environment

If you want to use a separate virtual environment (venv) for this project.

./install.sh
source venv/bin/activate

B) Using Your Own Virtual Environment

Assuming you already created and switched to your own virtual environment.

for line in $(cat requirements.txt); do pip install $line; done

Download SpaCy models (~1 GB).

python -m spacy.en.download all

Database

The recommended database is PostgreSQL.

Assuming you want to create a database reviewer_suggestions_db with the user reviewer_suggestions_user and password 'password', you may use the following commands as a guidance:

sudo -u postgres psql -c "create user reviewer_suggestions_user;"
sudo -u postgres psql -c "create database reviewer_suggestions_db;"
sudo -u postgres psql -c "alter user reviewer_suggestions_user with encrypted password 'password';"
sudo -u postgres psql -c "grant all privileges on database reviewer_suggestions_db to reviewer_suggestions_user;"

Create / Update Database Schema

python -m peerscout.preprocessing.migrateSchema

Populate Database

This will download updates from AWS and populate the database with it. This may take a while. It will also download data from Crossref, which might fail, which is 'okay'.

python -m peerscout.preprocessing.updateDataAndReload

Compile Client

cd client
npm install
npm run bundle

Start Server

python -m peerscout.server

Then go to http://localhost:8080/. (If you are getting a 'Not Found' error you may not have compiled the client)

The server will provide the REST API http://localhost:8080/api/ and serve the static client bundle.

Start Client Dev Server

Use this option to develop the client, in addition to the python server (which will still provide the API).

cd client
npm start

Then server will be availabe under http://localhost:8081/.

Tests

Run All Tests

./project_tests.sh

Python Tests

pytest

To watch tests, run pytest-watch instead.

JavaScript Tests

cd client
npm test

Watch:

cd client
npm run ~test