Feature: Sync annotator code with NCBO code

.dockerignore

@@ -1,6 +1,9 @@
-# Git
+.bundle
 .git
 .gitignore
+.github
+.DS_Store
+vendor/bundle
 # Logs
 log/*
 # Temp files

.github/workflows/ruby-unit-tests.yml

@@ -7,11 +7,25 @@ on:
 jobs:
   test:
     runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        backend: ['ruby', 'ruby-agraph'] # api runs tests with 4store backend and api-agraph runs with AllegroGraph backend
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: copy config file
-      run: cp config/config.rb.test config/config.rb
+      run: cp config/config.test.rb config/config.rb
     - name: Build docker-compose
       run: docker-compose build
     - name: Run unit tests
-      run: docker-compose up --exit-code-from unit-test
+      # unit tests are run inside a container
+      # http://docs.codecov.io/docs/testing-with-docker
+      run: |
+        ci_env=`bash <(curl -s https://codecov.io/env)`
+        docker-compose run $ci_env -e CI ${{ matrix.backend }} bundle exec rake test TESTOPTS='-v'
+    - name: Upload coverage reports to Codecov
+      uses: codecov/codecov-action@v3
+      with:
+        flags: unittests
+        verbose: true
+        fail_ci_if_error: false # optional (default = false)

Dockerfile

@@ -1,10 +1,22 @@
-FROM ruby:2.6
+ARG RUBY_VERSION
+ARG DISTRO_NAME=bullseye
+
+FROM ruby:$RUBY_VERSION-$DISTRO_NAME
+
+RUN apt-get update -yqq && apt-get install -yqq --no-install-recommends \
+  openjdk-11-jre-headless \
+  raptor2-utils \
+  && rm -rf /var/lib/apt/lists/*
 
-RUN apt-get update -yqq && apt-get install -yqq --no-install-recommends openjdk-11-jre-headless raptor2-utils
 # The Gemfile Caching Trick
 RUN mkdir -p /srv/ontoportal/ncbo_annotator
-COPY Gemfile* /srv/ontoportal/ncbo_annotator/
+COPY *.gemspec Gemfile* /srv/ontoportal/ncbo_annotator/
+
 WORKDIR /srv/ontoportal/ncbo_annotator
-RUN gem install bundler -v "$(grep -A 1 "BUNDLED WITH" Gemfile.lock | tail -n 1)"
+
+# set bundler to v2.4.22 which is the last version supported by ruby 2.7
+RUN gem install bundler -v 2.4.22
+ENV BUNDLE_PATH /bundle
 RUN bundle install
 COPY . /srv/ontoportal/ncbo_annotator
+CMD ["/bin/bash"]

Gemfile

@@ -1,19 +1,21 @@
 source 'https://rubygems.org'
 
-gem 'cube-ruby', require: 'cube'
+gem 'cube-ruby'
 gem 'faraday', '~> 1.9'
 gem 'ffi'
-gem 'minitest', '~> 4.0'
-gem 'oj', '~> 2.0'
+gem 'oj', '~> 3.0'
 gem 'rake', '~> 10.0'
 gem 'redis'
 gem 'ruby-xxHash'
 
-# Development
-gem 'pry', group: :development
+group :development do
+  gem 'minitest', '~> 4.0'
+  gem 'pry'
+  gem 'simplecov'
+  gem 'simplecov-cobertura' # for codecov.io
+end
 
 # NCBO gems (can be from a local dev path or from rubygems/git)
 gem 'goo', github: 'ontoportal-lirmm/goo', branch: 'development'
 gem 'sparql-client', github: 'ncbo/sparql-client', branch: 'master'
 gem 'ontologies_linked_data', github: 'ontoportal-lirmm/ontologies_linked_data', branch: 'development'
-gem 'ncbo_resource_index', github: 'ncbo/resource_index'

Gemfile.lock

@@ -1,18 +1,5 @@
 GIT
-  remote: https://github.com/ncbo/resource_index.git
-  revision: 24a7f14a6da4f4a0eaba1016ca5a378dfccd7441
-  specs:
-    ncbo_resource_index (0.0.1)
-      elasticsearch (= 2.0.0)
-      mysql2 (= 0.5.2)
-      pony
-      ref
-      ruby-xxHash
-      sequel
-      typhoeus
-
-GIT
-  remote: https://github.com/ncbo/sparql-client.git
+  remote: https://github.com/ontoportal-lirmm/sparql-client.git
   revision: fb4a89b420f8eb6dda5190a126b6c62e32c4c0c9
   branch: master
   specs:
@@ -27,14 +14,13 @@ GIT
   branch: development
   specs:
     goo (0.0.2)
-      addressable (= 2.3.5)
+      addressable (~> 2.8)
       pry
       rdf (= 1.0.8)
       redis
       rest-client
       rsolr
       sparql-client
-      systemu
       uuid
 
 GIT
@@ -49,7 +35,6 @@ GIT
       json
       libxml-ruby
       multi_json
-      ncbo_resource_index
       oj
       omni_logger
       pony
@@ -67,25 +52,19 @@ GEM
       multi_json (~> 1.3)
       thread_safe (~> 0.1)
       tzinfo (~> 0.3.37)
-    addressable (2.3.5)
-    bcrypt (3.1.17)
+    addressable (2.8.6)
+      public_suffix (>= 2.0.2, < 6.0)
+    bcrypt (3.1.20)
+    bigdecimal (3.1.5)
     builder (3.2.4)
     coderay (1.1.3)
-    concurrent-ruby (1.1.10)
+    concurrent-ruby (1.2.2)
+    connection_pool (2.4.1)
     cube-ruby (0.0.3)
-    domain_name (0.5.20190701)
-      unf (>= 0.0.5, < 1.0.0)
-    elasticsearch (2.0.0)
-      elasticsearch-api (= 2.0.0)
-      elasticsearch-transport (= 2.0.0)
-    elasticsearch-api (2.0.0)
-      multi_json
-    elasticsearch-transport (2.0.0)
-      faraday
-      multi_json
-    ethon (0.15.0)
-      ffi (>= 1.15.0)
-    faraday (1.10.0)
+    date (3.3.4)
+    docile (1.4.0)
+    domain_name (0.6.20240107)
+    faraday (1.10.3)
       faraday-em_http (~> 1.0)
       faraday-em_synchrony (~> 1.0)
       faraday-excon (~> 1.1)
@@ -101,94 +80,116 @@ GEM
     faraday-em_synchrony (1.0.0)
     faraday-excon (1.1.0)
     faraday-httpclient (1.0.1)
-    faraday-multipart (1.0.3)
-      multipart-post (>= 1.2, < 3)
+    faraday-multipart (1.0.4)
+      multipart-post (~> 2)
     faraday-net_http (1.0.1)
     faraday-net_http_persistent (1.2.0)
     faraday-patron (1.0.0)
     faraday-rack (1.0.0)
     faraday-retry (1.0.3)
-    ffi (1.15.5)
+    ffi (1.16.3)
     http-accept (1.7.0)
-    http-cookie (1.0.4)
+    http-cookie (1.0.5)
       domain_name (~> 0.5)
     i18n (0.9.5)
       concurrent-ruby (~> 1.0)
-    json (2.6.1)
-    json_pure (2.6.1)
-    libxml-ruby (3.2.2)
-    logger (1.5.1)
+    json (2.7.1)
+    json_pure (2.7.1)
+    libxml-ruby (5.0.2)
+    logger (1.6.0)
     macaddr (1.7.2)
       systemu (~> 2.6.5)
-    mail (2.7.1)
+    mail (2.8.1)
       mini_mime (>= 0.1.1)
+      net-imap
+      net-pop
+      net-smtp
     method_source (1.0.0)
-    mime-types (3.4.1)
+    mime-types (3.5.2)
       mime-types-data (~> 3.2015)
-    mime-types-data (3.2022.0105)
-    mini_mime (1.1.2)
+    mime-types-data (3.2023.1205)
+    mini_mime (1.1.5)
     minitest (4.7.5)
     multi_json (1.15.0)
-    multipart-post (2.1.1)
-    mysql2 (0.5.2)
+    multipart-post (2.3.0)
     net-http-persistent (2.9.4)
+    net-imap (0.4.9.1)
+      date
+      net-protocol
+    net-pop (0.1.2)
+      net-protocol
+    net-protocol (0.2.2)
+      timeout
+    net-smtp (0.4.0.1)
+      net-protocol
     netrc (0.11.0)
-    oj (2.18.5)
+    oj (3.16.3)
+      bigdecimal (>= 3.0)
     omni_logger (0.1.4)
       logger
     pony (1.13.1)
       mail (>= 2.0)
-    pry (0.14.1)
+    pry (0.14.2)
       coderay (~> 1.1)
       method_source (~> 1.0)
-    rack (2.2.3)
-    rack-test (1.1.0)
-      rack (>= 1.0, < 3)
+    public_suffix (5.0.4)
+    rack (3.0.8)
+    rack-test (2.1.0)
+      rack (>= 1.3)
     rake (10.5.0)
     rdf (1.0.8)
       addressable (>= 2.2)
-    redis (4.6.0)
-    ref (2.0.0)
+    redis (5.0.8)
+      redis-client (>= 0.17.0)
+    redis-client (0.19.1)
+      connection_pool
     rest-client (2.1.0)
       http-accept (>= 1.7.0, < 2.0)
       http-cookie (>= 1.0.2, < 2.0)
       mime-types (>= 1.16, < 4.0)
       netrc (~> 0.8)
+    rexml (3.2.6)
     rsolr (2.5.0)
       builder (>= 2.1.2)
       faraday (>= 0.9, < 3, != 2.0.0)
     ruby-xxHash (0.4.0.2)
     ruby2_keywords (0.0.5)
     rubyzip (2.3.2)
-    sequel (5.56.0)
+    simplecov (0.22.0)
+      docile (~> 1.1)
+      simplecov-html (~> 0.11)
+      simplecov_json_formatter (~> 0.1)
+    simplecov-cobertura (2.1.0)
+      rexml
+      simplecov (~> 0.19)
+    simplecov-html (0.12.3)
+    simplecov_json_formatter (0.1.4)
     systemu (2.6.5)
     thread_safe (0.3.6)
-    typhoeus (1.4.0)
-      ethon (>= 0.9.0)
-    tzinfo (0.3.60)
-    unf (0.1.4)
-      unf_ext
-    unf_ext (0.0.8.1)
+    timeout (0.4.1)
+    tzinfo (0.3.62)
     uuid (2.3.9)
       macaddr (~> 1.0)
 
 PLATFORMS
-  x86_64-darwin-21
+  ruby
+  x86_64-linux
 
 DEPENDENCIES
   cube-ruby
   faraday (~> 1.9)
   ffi
   goo!
   minitest (~> 4.0)
-  ncbo_resource_index!
-  oj (~> 2.0)
+  oj (~> 3.0)
   ontologies_linked_data!
   pry
   rake (~> 10.0)
   redis
   ruby-xxHash
+  simplecov
+  simplecov-cobertura
   sparql-client!
 
 BUNDLED WITH
-   2.3.7
+   2.4.22

config/config.test.rb

@@ -0,0 +1,36 @@
+# frozen_string_literal: true
+
+#
+# configuration for unit testing
+#
+GOO_HOST         = ENV.include?('GOO_HOST')         ? ENV['GOO_HOST']         : 'localhost'
+GOO_PORT         = ENV.include?('GOO_PORT')         ? ENV['GOO_PORT']         : 9000
+GOO_BACKEND_NAME = ENV.include?("GOO_BACKEND_NAME") ? ENV["GOO_BACKEND_NAME"] : "4store"
+GOO_PATH_QUERY   = ENV.include?("GOO_PATH_QUERY")   ? ENV["GOO_PATH_QUERY"]   : "/sparql/"
+GOO_PATH_DATA    = ENV.include?("GOO_PATH_DATA")    ? ENV["GOO_PATH_DATA"]    : "/data/"
+GOO_PATH_UPDATE  = ENV.include?("GOO_PATH_UPDATE")  ? ENV["GOO_PATH_UPDATE"]  : "/update/"
+REDIS_HOST       = ENV.include?('REDIS_HOST')       ? ENV['REDIS_HOST']       : 'localhost'
+REDIS_PORT       = ENV.include?('REDIS_PORT')       ? ENV['REDIS_PORT']       : 6379
+MGREP_HOST       = ENV.include?('MGREP_HOST')       ? ENV['MGREP_HOST']       : 'localhost'
+MGREP_PORT       = ENV.include?('MGREP_PORT')       ? ENV['MGREP_PORT']       : 55556
+SOLR_TERM_SEARCH_URL = ENV.include?("SOLR_TERM_SEARCH_URL") ? ENV["SOLR_TERM_SEARCH_URL"] : "http://localhost:8983/solr/term_search_core1"
+SOLR_PROP_SEARCH_URL = ENV.include?("SOLR_PROP_SEARCH_URL") ? ENV["SOLR_PROP_SEARCH_URL"] : "http://localhost:8983/solr/prop_search_core1"
+
+
+LinkedData.config do |config|
+  config.goo_backend_name           = GOO_BACKEND_NAME.to_s
+  config.goo_host                   = GOO_HOST.to_s
+  config.goo_port                   = GOO_PORT.to_i
+  config.goo_path_query             = GOO_PATH_QUERY.to_s
+  config.goo_path_data              = GOO_PATH_DATA.to_s
+  config.goo_path_update            = GOO_PATH_UPDATE.to_s
+  config.search_server_url          = SOLR_TERM_SEARCH_URL.to_s
+  config.property_search_server_url = SOLR_PROP_SEARCH_URL.to_s
+end
+Annotator.config do |config|
+  config.mgrep_host            = MGREP_HOST.to_s
+  config.mgrep_port            = MGREP_PORT.to_i
+  config.mgrep_dictionary_file = './test/data/dictionary.txt'
+  config.annotator_redis_host  = REDIS_HOST.to_s
+  config.annotator_redis_port  = REDIS_PORT.to_i
+end