The Herd Book Is a NBIS project Supporting the Swedish Board of Agriculture (Jordbruksverket) Action Plan for the Long-term Sustainable Management of Swedish Animal Genetic Resources 2010–2020.
Mainly measure 5 “Keeping herd books and planning genetic conservation breeding”. Found in this report
The aim here is to set up a online web-service for keeping a herd book and spreading admin task to all local genebank holders.
A site participating in the preservation program. These will typically have signed an agreement and have efforts in place to make sure there is no accidental breeding with other races. Have a local pool of animals that is the actual in situ genebank.
Several different Herds of the same race will form a Genebank
A user tied to one or more herds. The owner or owners of a herd of animals
A user tasked with overall maintenance of the herd book and managing the breeding/preservation effort for a genebank. Usually responsible for a specific animal breed.
First set up your secure environment variables by
copy or rename the default .env
files found in the .docker
folder so there are files without the
.default
suffix.
Remember to update your secrets before going into production.
The following files should exist:
.docker/database-variables.env
.docker/r-api-variables.env
It is also recommended to speed up startup by generating persistent forward-secrecy parameters:
openssl dhparam -out config/dhparam.pem 2048
To set up the developer environment, you should be able to run
docker-compose build
docker-compose up
You can also use the script run-with-prebuilt-images.sh
, that uses the precompiled images generated
in the production server. This option is much faster and therefore more recommended.
After a few minutes you should have your services up and running. If you run docker ps
you will see an output similiar to this one:
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
f62cebbc2e43 herdbook_main "/bin/sh -c /entrypo…" 3 hours ago Up 3 hours 0.0.0.0:2345->2345/tcp, 0.0.0.0:4200->8080/tcp, 0.0.0.0:4443->8443/tcp herdbook-main
7385113e91c0 postgres:12 "docker-entrypoint.s…" 3 hours ago Up 3 hours 0.0.0.0:5432->5432/tcp herdbook-db
5c9193fc774c herdbook_frontend:latest "docker-entrypoint.s…" 20 hours ago Up 3 hours theherdbook_herdbook-frontend-devel_1
To access the local server deployed open the encryption-enabled interface in your browser. You will need to
configure the browser to allow self-signed localhost certificates. In Chrome, this can be done by accessing
this property from the browser: chrome://flags/#allow-insecure-localhost
and setting its vale to Enabled.
In Firefox, when loading the url you can click on Advanced and add the exception suggested from the browser.
You can also use the interface without encryption to access the frontend.
To be able to login in the website and play with it you will need to create an user with admin privileges. This can be done by executing register_user.sh, providing your email and password:
./register_user.sh 'username' 'userpassword' '[email protected]'
All branches that are pushed to github has prebuilt images. To use the prebuilt images instead of building locally, use:
./run-with-prebuilt-images.sh
You also need to configure a user for the R server. This is done by providing an Authorized
header to use as API_AUTH
in .docker/r-api-variables.env
. This header must provide credentials to see all individuals for which the R service
will be used.
You typically create a specific user for this:
./register_user.sh 'rapiuser' 'rapipassword' '[email protected]'
You can then create the corresponding header as Authorization: Basic $(echo -n rapiuser:rapipassword |base64)
so the line in .docker/r-api-variables.env
might look something like:
API_AUTH=Authorization: Basic dGVzdDp0ZXN0
The default file contains the credentials for the user test
, password test
.
Data files are delivered out of band. Instructions for inital importing of data are available in scripts/README.docker
.
There are a number of tests written, which can be run using ./run_tests.sh
.
Currently the test coverage is lacking, and frontend testing is missing. Given
time, this will be resolved before version 1.0 is finished.
The tests will also run through pytest
using github actions when pushing new
code to github.
To set up Google authentication, visit the developer console and select/create a project (dropdown next to the Google cloud platform in the upper left corner).
When you have a project, you can select "Credentials" and use the "Create credentials" to create an OAuth client ID.
Once selected, you will be told you need to configure the consent screen. You can choose "Internal" to only provide login for users within your organization or external, potentially allowing for access from any google account (but which may require verification).
Once selected, you'll need to supply an application name (e.g. Herdbook) and a support contact. You also need to supply a developer contact (for Google) at the bottom.
Once filled in, you can save and continue.
Next, you'll need to select the scopes provided. Select "Add and remove scopes" and pick openid
.
If you want to support automatic account creation, you also need to select ".../auth/userinfo.email"
Once done, select save and continue to finish configuration of the consent screen.
As you now have the consent screen, you can again select Credentials and use Create new credential -> OAuth client ID.
When asked for the application type, select "Web application".
You will be asked for URLs. Under "Authorized redirect URI" you should provide
https://YOURSITE/api/login/google/back/google/authorized
where YOURSITE is any address your service can be accessed at.
Once you've added all such URIs, you can select "Create".
You will be presented with a box with "Your client ID" and "Your client secret", copy these.
Create/edit config/auth.ini
and add a [google]
section where you provide your client ID as key
and your secret as secret
, e.g.
[google]
key=YourIDHereLikelyEndsWith.apps.googleusercontent.com
secret=YourClientSecretGoesHere
and restart the server, once this is done - you should see the option to login with Google.
If you want to support automatic creation of accounts from Google, also include
autocreate = yes
in the [google]
section.
In short, you need to set up OAuth 2.0 with the openid
scope (and optionally the .../userinfo.email
scope to allow automatic creation). Once you've done that, provide the Client secret
as secret
and Client ID
as key
in the [google]
section in auth.ini
.
The google authenticator driver also supports automatically creating herd ownership links at the time of account creation. This requires using a custom domain / Google workspace.
To use that, you need to enable the Admin SDK API
in the google developer
console. Next, create a service account under "Credentials" in your project and
create and download keys for the service account in JSON format.
Note the Unique ID
for the service account. Head over to the
workspace console and add a domain wide delegation
(under Security -> API controls).
Create a domain wide delegation for the unique id from the service account with
the scope https://www.googleapis.com/auth/admin.directory.user.readonly
.
Once this is done, you can add the key JSON file to the config
directory.
Also modify the auth.ini
in the [google]
section, adding herdattribute
describing what attribute to be used for tying the user to herd(s). Other keys
needed are credentialsfile
which should be the name of the key file for the
servicea count as well as lookupuser
which should be set to the e-mail of an
administrative user for the domain.