How to run a development environment for the backend

This page describes a list of steps to follow to run a development environment for the backend locally

We assume in this guide:

That you have access to a shell Terminal, either because you are running on a Linux/Mac operating system or Windows wth WSL 2 installed

You will use this same backend environment for ontologies_api, ontologies_linked_data, and goo projects

Steps to follow

Run the backend infrastructure using VirtualBox
1. Install the appliance on VirtualBox
2. Configure port forwarding
Run the code locally

Run the backend infrastructure using VirtualBox

We will use a virtual machine with a nat port forwarding of all ports needed to run the ontologies_api project locally. Port forwarding will make them all accessible from the localhost

Install the appliance on VirtualBox

First get the appliance and run it on VirtualBox

Configure port forwarding

Then on VirtualBox go to Settings > Network, and change the adapter to NAT and click on Advanced then Port Forwarding. Where you add the following ports to forward:

Backend infrastructure port forwarding

If mapping SSH to port 22 didn’t work, you can use a different port port (e.g 2222)

Finally start the appliance and connect to the appliance with this command line: ssh centos@127.0.0.1

Important: if you can’t establish an ssh connection with the appliance ,make sure to run first the SSH service in the appliance, using this command: sudo systemctl enable sshd

Important: to make the port forwarding of redis and 4store work, be sure to disable first the firewall of the appliance, after connecting to the appliance via ssh using the command: sudo systemctl stop iptables

Run the code locally

Fork and clone backend source codes

First fork the ontologies_api repository. More details about how to can be found in the Github official doc

Then clone your forked repository locally

  git clone https://github.com/your_organization/ontologies_api.git

The master branch represent the Bioportal version, if you want to have the Agroportal version. You have to merge the ontoportal-lirmm branch to your master, by doing so git merge ontoportal-lirmm master

Create the config file `config/environments/development.rb`

A sample file of config/environments/development.rb can be found here: https://github.com/ontoportal/ontologies_api/blob/master/config/environments/config.rb.sample

In command line do

cp config/environments/config.rb.sample config/environments/development.rb

Copy the content bellow in the config/environments/development.rb

$HOSTNAME = $UI_HOSTNAME = "ontoportal.local"
$REST_HOSTNAME = "localhost"
$REST_PORT = 9393
$REST_URL_PREFIX = "http://#{[$REST_HOSTNAME, $REST_PORT].compact.join(':')}/"
$DATADIR ||= '/srv/ontoportal/data'

GOO_BACKEND_NAME = '4store'
GOO_PORT        = GOO_BACKEND_NAME.include?('AG') ? 10035                                 : 8081
GOO_PATH_QUERY  = GOO_BACKEND_NAME.include?('AG') ? '/repositories/ontoportal'            : '/sparql/'
GOO_PATH_DATA   = GOO_BACKEND_NAME.include?('AG') ? '/repositories/ontoportal/statements' : '/data/'
GOO_PATH_UPDATE = GOO_BACKEND_NAME.include?('AG') ? '/repositories/ontoportal/statements' : '/update/'

begin
  # For prefLabel extract main_lang first, or anything if no main found.
  # For other properties only properties with a lang that is included in main_lang are used
  Goo.main_lang = ["en", "eng", "fr", "fre"]
  Goo.use_cache = false
rescue NoMethodError
  puts "(CNFG) >> Goo.main_lang not available"
end

begin
  LinkedData.config do |config|
    config.goo_host                   = 'localhost'
    config.goo_port                   = "#{GOO_PORT}"
    config.goo_backend_name           = "#{GOO_BACKEND_NAME}"
    config.goo_path_query             = "#{GOO_PATH_QUERY}"
    config.goo_path_data              = "#{GOO_PATH_DATA}"
    config.goo_path_update            = "#{GOO_PATH_UPDATE}"

    config.rest_url_prefix            = "#{$REST_URL_PREFIX}"
    config.ui_host                    = "#{$UI_HOSTNAME}"
    config.search_server_url          = 'http://localhost:8983/solr/term_search_core1'
    config.property_search_server_url = 'http://localhost:8983/solr/prop_search_core1'
    config.repository_folder          = "#{$DATADIR}/repository"
    config.replace_url_prefix         = true
    config.enable_security            = true
    config.enable_slices              = true

    # Caches
    Goo.use_cache             = false
    config.goo_redis_host     = "localhost"
    config.goo_redis_port     = 6381
    config.enable_http_cache  = true
    config.http_redis_host    = "localhost"
    config.http_redis_port    = 6380

    # PURL server config parameters
    config.enable_purl            = false
    config.purl_host              = "purl.example.org"
    config.purl_port              = 80
    config.purl_username          = "admin"
    config.purl_password          = "password"
    config.purl_maintainers       = "admin"
    config.purl_target_url_prefix = "http://example.org"

    # Email notifications
    config.enable_notifications   = true
    config.email_sender           = "notifications@test.com" # Default sender for emails
    config.email_override         = "notifications@test.com" # all email gets sent here. Disable with email_override_disable.
    config.email_disable_override = true
    config.smtp_host              = "smtp.lirmm.fr"
    config.smtp_port              = 25
    config.smtp_auth_type         = :none # :none, :plain, :login, :cram_md5
    config.smtp_domain            = "lirmm.fr"
    # Emails of the instance administrators to get mail notifications when new user or new ontology
    config.admin_emails           = []

    # Ontology Google Analytics Redis
    # disabled
    config.ontology_analytics_redis_host = "localhost"
    config.ontology_analytics_redis_port = 6379

    # Used to define other bioportal that can be mapped to
    # Example to map to ncbo bioportal : {"ncbo" => {"api" => "http://data.bioontology.org", "ui" => "http://bioportal.bioontology.org", "apikey" => ""}
    # Then create the mapping using the following class in JSON : "http://purl.bioontology.org/ontology/MESH/C585345": "ncbo:MESH"
    # Where "ncbo" is the namespace used as key in the interportal_hash
    config.interportal_hash   = {
      "agroportal" => {
        "api" => "http://data.agroportal.lirmm.fr",
        "ui" => "http://agroportal.lirmm.fr",
        "apikey" => "1cfae05f-9e67-486f-820b-b393dec5764b"
      },
      "ncbo" => {
        "api" => "http://data.bioontology.org",
        "ui" => "http://bioportal.bioontology.org",
        "apikey" => "4a5011ea-75fa-4be6-8e89-f45c8c84844e"
      },
      "sifr" => {
        "api" => "http://data.bioportal.lirmm.fr",
        "ui" => "http://bioportal.lirmm.fr",
        "apikey" => "1cfae05f-9e67-486f-820b-b393dec5764b"
      }
    }
  end
rescue NameError
  #  puts '(CNFG) >> LinkedData not available, cannot load config'
end
begin
  Annotator.config do |config|
    config.mgrep_dictionary_file   = "#{$DATADIR}/mgrep/dictionary/dictionary.txt"
    config.stop_words_default_file = './config/default_stop_words.txt'
    config.mgrep_host              = 'localhost'
    config.mgrep_port              = 55555
    config.mgrep_alt_host          = 'localhost'
    # secondary mgrep instance is not configured for appliance. routing all requestes to the primary mgrep
    config.mgrep_alt_port          = 55555
    config.annotator_redis_host    = 'localhost'
    config.annotator_redis_port    = 6379
  end
rescue NameError
  #  puts '(CNFG) >> Annotator not available, cannot load config'
end

begin
  OntologyRecommender.config do |config|
  end
rescue NameError
  #  puts '(CNFG) >> OntologyRecommender not available, cannot load config'
end

begin
  LinkedData::OntologiesAPI.config do |config|
    config.enable_unicorn_workerkiller = true
    config.enable_throttling           = false
    config.http_redis_host             = LinkedData.settings.http_redis_host
    config.http_redis_port             = LinkedData.settings.http_redis_port
    config.restrict_download           = []
    #config.ontology_rank               = ""
  end
rescue NameError
  #  puts '(CNFG) >> OntologiesAPI not available, cannot load config'
end

begin
  NcboCron.config do |config|
    config.redis_host           = Annotator.settings.annotator_redis_host
    config.redis_port           = Annotator.settings.annotator_redis_port
    config.enable_ontology_analytics = false
    config.search_index_all_url = 'http://localhost:8983/solr/term_search_core2'
    config.property_search_server_index_all_url = 'http://localhost:8983/solr/prop_search_core2'
    config.ontology_report_path = "#{$DATADIR}/reports/ontologies_report.json"
    config.enable_spam_deletion = false
  end
rescue NameError
  puts '(CNFG) >> NcboCron not available, cannot load config'
end

Install dependencies

You will need to install Ruby, for the version see the RUBY_VERSION argument

Link your local `ontologies_linked_data` and `goo` code to `ontologies_api` project

If you want to use your local code for ontologies_linked_data and goo you will need also to fork and clone them from the Ontoportal repositories

In addition, you will need to configure the bundler to use them, by doing like below

cd put_here_the_path_to_your_local_ontologies_api_code
bundle config local.goo put_here_the_path_to_your_local_goo_code
bundle config local.ontologies_linked_data put_here_the_path_to_your_local_ontologies_linked_data_code

Run the ontologies_api server

bundle exec shotgun --env=development

Opening in browser

See http://localhost:9393