Add Solr and basic search page

Gemfile

@@ -29,6 +29,7 @@ gem "plek", "~> 4.1" # TODO: unconstrain once govuk_pub_components up-to-date.
 gem "puma"
 gem "redcarpet"
 gem "rest-client"
+gem "rsolr"
 gem "sass-rails"
 gem "secure_headers"
 gem "sentry-raven"

Gemfile.lock

@@ -528,6 +528,9 @@ GEM
     rexml (3.2.8)
       strscan (>= 3.0.9)
     rouge (4.2.0)
+    rsolr (2.6.0)
+      builder (>= 2.1.2)
+      faraday (>= 0.9, < 3, != 2.0.0)
     rspec (3.13.0)
       rspec-core (~> 3.13.0)
       rspec-expectations (~> 3.13.0)
@@ -697,6 +700,7 @@ DEPENDENCIES
   rails (= 7.0.5.1)
   redcarpet
   rest-client
+  rsolr
   rspec
   rspec-rails
   rubocop-govuk

app/controllers/solr_search_controller.rb

@@ -0,0 +1,14 @@
+class SolrSearchController < ApplicationController
+  before_action :search_for_dataset, only: [:search]
+
+  def search; end
+
+private
+
+  def search_for_dataset
+    query = Search::Solr.search(params)
+
+    @datasets = query["response"]["docs"]
+    @num_results = query["response"]["numFound"]
+  end
+end

app/services/search/solr.rb

@@ -0,0 +1,37 @@
+module Search
+  class Solr
+    def self.search(params)
+      query_param = params.fetch("q", "").squish
+      page = params["page"]
+      page && page.to_i.positive? ? page.to_i : 1
+
+      solr_client = client
+
+      query = "*:*" if query_param.empty?
+
+      solr_client.get "select", params: {
+        q: query,
+        start: page,
+        rows: 20,
+        fl: field_list,
+      }
+    end
+
+    def self.field_list
+      %w[
+        id
+        name
+        title
+        organization
+        notes
+        metadata_modified
+        extras_theme-primary
+        validated_data_dict
+      ].freeze
+    end
+
+    def self.client
+      @client ||= RSolr.connect(url: ENV["SOLR_URL"])
+    end
+  end
+end

app/views/solr_search/search.html.erb

@@ -0,0 +1,55 @@
+<% content_for :page_title do %><%= search_term.present? ?
+  t('.results_page_title', query: search_term) :
+  t('.search_results') %>
+<% end %>
+
+<% content_for :head do %>
+  <meta name="robots" content="noindex">
+<% end %>
+<form action="/search/solr" method="GET">
+  <main role="main" id="main-content" class="govuk-main-wrapper">
+    <div class="govuk-width-container">
+      <%= render "govuk_publishing_components/components/heading", {
+        text: t(".search_results_heading"),
+        margin_bottom: 6,
+        heading_level: 1,
+        font_size: "l",
+      } %>
+      <div class="govuk-grid-row">
+        <div class="govuk-grid-column-two-thirds" id="search-box">
+          <%= render "govuk_publishing_components/components/search", {
+            label_text: t('.search_data_gov_uk'),
+          } %>
+        </div>
+      </div>
+
+      <div class="govuk-grid-row dgu-filters">
+        <div class="govuk-grid-column-two-thirds dgu-results">
+          <span class="dgu-results__summary">
+            <span class="govuk-body-s govuk-!-font-weight-bold"><%= number_with_delimiter(@num_results) %></span>
+            <%= t('.result').pluralize(@num_results) %> <%= t('.found') %>
+          </span>
+
+          <% @datasets.each do |dataset| %>
+            <div class="dgu-results__result">
+              <h2 class="govuk-heading-m">
+                <%= dataset["title"] %>
+              </h2>
+              <dl class="dgu-metadata__box">
+                <dt><%= t('.meta_data_box.published_by') %>:</dt>
+                <dd class="published_by"><%= JSON.parse(dataset["validated_data_dict"])["organization"]["title"] %></dd>
+                <dt><%= t('.meta_data_box.last_updated') %>:</dt>
+                <% if dataset["metadata_modified"].present? %>
+                  <dd><%= format_timestamp(dataset["metadata_modified"]) %></dd>
+                <% else %>
+                  <dd class="dgu-secondary-text"> <%= t('.meta_data_box.not_applicable') %></dd>
+                <% end %>
+              </dl>
+              <p><%= truncate(strip_tags(dataset["notes"]), length: 200, separator: ' ') %></p>
+            </div>
+          <% end %>
+        </div>
+      </div>
+    </div>
+  </main>
+</form>

config/locales/views/search/en.yml

@@ -42,3 +42,16 @@ en:
       sort_by: "Sort by"
       accessibility:
         submit_button: "Apply sorting"
+  solr_search:
+    search:
+      results_page_title: 'Results for "%{query}"'
+      search_results: 'Search Results'
+      search_results_heading: "Search results"
+      search_data_gov_uk: "Search data.gov.uk"
+      result: "result"
+      found: "found"
+      meta_data_box:
+        availability: "Availability"
+        published_by: "Published by"
+        last_updated: "Last updated"
+        not_applicable: "Not applicable"

config/routes.rb

@@ -50,6 +50,9 @@
   get "dataset/:uuid/:name", to: "datasets#show", as: "dataset"
   get "dataset/:dataset_uuid/:name/datafile/:datafile_uuid/preview", to: "previews#show", as: "datafile_preview"
 
+  # /solr feature flag
+  get "search/solr", to: "solr_search#search"
+
   get "acknowledge", to: "messages#acknowledge"
 
   # Route everything else to CKAN

spec/features/solr_search_page_spec.rb

@@ -0,0 +1,45 @@
+require "rails_helper"
+
+RSpec.describe "Solr Search page", type: :feature do
+  let(:results) { File.read(Rails.root.join("spec/fixtures/solr_response.json").to_s) }
+
+  before do
+    allow(Search::Solr).to receive(:search).and_return(JSON.parse(results))
+  end
+
+  scenario "Displays search heading" do
+    visit "/search/solr"
+    expect(page).to have_css("h1", text: "Search results")
+  end
+
+  scenario "Displays search box" do
+    visit "/search/solr"
+    expect(page).to have_content("Search data.gov.uk")
+  end
+
+  scenario "Displays search results" do
+    visit "/search/solr"
+    expect(page).to have_content("2 results found")
+  end
+
+  scenario "Displays the title for each search result" do
+    visit "/search/solr"
+    expect(page).to have_css("h2", text: "A very interesting dataset")
+    expect(page).to have_css("h2", text: "A dataset with additional inspire metadata")
+  end
+
+  scenario "Displays the publisher for each search result" do
+    visit "/search/solr"
+    results = all(".published_by")
+    expect(results.length).to be(2)
+    expect(results[0]).to have_content "Ministry of Housing, Communities and Local Government"
+    expect(results[1]).to have_content "Mole Valley District Council"
+  end
+
+  scenario "Displays the last updated for each search result" do
+    visit "/search/solr"
+    expect(page).to have_content("Last updated", count: 2)
+    expect(page).to have_content("30 June 2017")
+    expect(page).to have_content("17 August 2018")
+  end
+end

spec/services/search/solr_spec.rb

@@ -0,0 +1,79 @@
+require "rails_helper"
+
+RSpec.describe Search::Solr do
+  before do
+    ENV["SOLR_URL"] = "http://localhost:8983/solr"
+  end
+
+  describe "#client" do
+    let(:client) { described_class.client }
+
+    it "is an instance of the RSolr client" do
+      expect(client).to be_an_instance_of(RSolr::Client)
+    end
+
+    it "connects to Solr with a valid URL" do
+      expect(client.options[:url]).to eq(ENV["SOLR_URL"])
+    end
+
+    it "only sets up the connection once" do
+      client1 = described_class.client
+      expect(client1).to eq(client)
+    end
+  end
+
+  describe "#search" do
+    let(:response) { File.read(Rails.root.join("spec/fixtures/solr_response.json").to_s) }
+    let(:results) { described_class.search("q" => "") }
+
+    before do
+      allow_any_instance_of(RSolr::Client).to receive(:get).and_return(JSON.parse(response))
+    end
+
+    it "returns a JSON response" do
+      expect(results).to be_a(Hash)
+    end
+
+    it "includes a count of the results" do
+      expect(results["response"]["numFound"]).to eq(2)
+    end
+
+    it "includes the datasets" do
+      datasets = results["response"]["docs"]
+      expect(datasets.length).to eq(2)
+    end
+
+    context "datasets" do
+      let(:dataset) { results["response"]["docs"].first }
+
+      it "includes the id" do
+        expect(dataset["id"]).to eq("420932c7-e6f8-43ea-adc5-3141f757b5cb")
+      end
+
+      it "includes the name" do
+        expect(dataset["name"]).to eq("a-very-interesting-dataset")
+      end
+
+      it "includes the title" do
+        expect(dataset["title"]).to eq("A very interesting dataset")
+      end
+
+      it "includes the notes" do
+        expect(dataset["notes"]).to eq("Lorem ipsum dolor sit amet, consectetur adipiscing elit. Donec euismod mauris in augue laoreet congue. Phasellus bibendum leo vel magna lacinia eleifend. Nam vitae lectus quis nulla varius faucibus id quis nibh. Nullam auctor ipsum non nunc efficitur bibendum Sed vitae ex nisi. Suspendisse posuere purus ac dui posuere, in interdum risus ornare. \n\nThe following files can be found on the website here:\n\nhttps://www.gov.uk/")
+      end
+
+      it "includes the last updated date" do
+        expect(dataset["metadata_modified"]).to eq("2017-06-30T09:08:37.040Z")
+      end
+
+      it "includes the organization slug" do
+        expect(dataset["organization"]).to eq("department-for-communities-and-local-government")
+      end
+
+      it "includes the organization name" do
+        data_dict = JSON.parse(dataset["validated_data_dict"])
+        expect(data_dict["organization"]["title"]).to eq("Ministry of Housing, Communities and Local Government")
+      end
+    end
+  end
+end