Improve TeSS scraping

FUNDERS.yaml

@@ -45,6 +45,7 @@ CINECA-Project:
     funder: true
     funding_id: "825775"
     funding_system: cordis
+    funder_name: Horizon 2020
     url: https://www.cineca-project.eu
 
 epsrc-training-grant:

_layouts/base.html

@@ -22,7 +22,6 @@
         <link rel="manifest" href="{{ site.baseurl }}/manifest.json">
         <meta name="theme-color" content="#2c3143"/>
 
-
         {{ page | generate_dublin_core: site }}
 
         {% assign topic = site.data[page.topic_name] %}

_layouts/faq.html

@@ -1,9 +1,9 @@
 ---
 layout: base
 ---
-
+<div itemscope itemtype="https://schema.org/FAQPage">
 {% assign contributors = site.data['contributors'] %}
-<section class="tutorial">
+<section class="tutorial" itemscope itemprop="mainEntity" itemtype="https://schema.org/Question">
 
   <nav aria-label="breadcrumb">
     <ol class="breadcrumb">
@@ -37,8 +37,7 @@
 
     </ol>
   </nav>
-
-  <h1>{{page.title}}</h1>
+  <h1 itemprop="name">{{page.title}}</h1>
 
     {% if page.contributors %}
     <div class="contributors-line" style="font-size:1rem;">{{ locale['authors'] | default: "Authors" }}: {% include
@@ -48,8 +47,8 @@ <h1>{{page.title}}</h1>
     {% endif %}
 
     <div class="container">
-      <div class="row">
-        <div class="col-md-8">
+      <div class="row" itemscope itemprop="acceptedAnswer" itemtype="https://schema.org/Answer">
+        <div class="col-md-8" itemprop="text">
            {{ page.description }}
 
 
@@ -85,3 +84,4 @@ <h2 id="bibliography">{{locale['references']| default: "References" }}</h2>
     {% endif %}
 
 </section>
+</div>

_layouts/learning-pathway.html

@@ -11,6 +11,24 @@
   {% assign gitter = site.gitter_url %}
 {% endif %}
 
+<script type="application/ld+json">
+{
+	"@context": "https://schema.org",
+	"@type": "Course",
+	"name": "{{ pathway.title }}",
+	"description": "{{ pathway.description }}",
+	"provider": {
+		"@type": "Organization",
+		"email": "[email protected]",
+		"name": "Galaxy Training Network",
+		"url": "https://training.galaxyproject.org",
+		"logo": "https://training.galaxyproject.org/training-material/assets/images/GTNLogo1000.png",
+		"sameAs": "https://training.galaxyproject.org"
+	}
+}
+</script>
+
+
 
 <section class="tutorials-list {{ page.topic_name }} topic-type-{{ topic.type }}">
     <hgroup>

_plugins/gtn/boxify.rb

@@ -155,7 +155,7 @@ def self.generate_collapsible_title(box_type, title, lang = 'en', key, contents:
       [box_id, %(
         <div class="box-title #{box_type}-title" id="#{box_id}">
         <button class="gtn-boxify-button #{box_type}" type="button" aria-controls="#{box_id}#{refers_to_contents}" aria-expanded="true">
-          #{get_icon(box_type)} #{box_title}
+          #{get_icon(box_type)} <span>#{box_title}</span>
           <span class="fold-unfold fa fa-minus-square"></span>
         </button>
         </div>

_plugins/gtn/contributors.rb

@@ -38,6 +38,25 @@ def self.get_authors(data)
       end
     end
 
+    ##
+    # Get the non-author contributors of a material.
+    # Params:
+    # +data+:: +Hash+ of the YAML frontmatter from a material
+    # Returns:
+    # +Array+ of contributor IDs
+    def self.get_non_authors(material)
+      if material.key?('contributors')
+        []
+      elsif material.key?('contributions')
+        material['contributions']
+          .reject { |k| k == 'funding' }
+          .reject { |k| k == 'authorship' }
+          .values.flatten.uniq
+      else
+        []
+      end
+    end
+
     # Convenience method to allow us to handle nil sites, and load directly
     # from disk ourselves.
     def self._load_file(site, category)

_plugins/jekyll-jsonld.rb

@@ -10,7 +10,8 @@ module JsonldFilter
       '@type': 'Organization',
       email: '[email protected]',
       name: 'Galaxy Training Network',
-      url: 'https://galaxyproject.org/teach/gtn/'
+      url: 'https://training.galaxyproject.org',
+      logo: 'https://training.galaxyproject.org/training-material/assets/images/GTNLogo1000.png',
     }.freeze
 
     A11Y = {
@@ -25,6 +26,14 @@ module JsonldFilter
                             'for non-visual users.',
     }.freeze
 
+    EDU_ROLES = {
+      'use' => 'Students',
+      'admin-dev' => 'Galaxy Administrators',
+      'basics' => 'Students',
+      'data-science' => 'Data-Science Students',
+      'instructors' => 'Instructors',
+    }
+
     ##
     # Generate the Dublin Core metadata for a material.
     # Parmaeters:
@@ -131,21 +140,36 @@ def generate_org_jsonld(id, contributor, site)
     end
 
     def generate_funder_jsonld(id, contributor, site)
-      organization = {
-        '@context': 'https://schema.org',
-        '@type': 'Grant',
-        identifier: contributor['funding_id'],
-        url: contributor['url'] || Gtn::Contributors.fetch_funding_url(contributor),
-        funder: {
+      organization = [
+        {
+          '@context': 'https://schema.org',
           '@type': 'Organization',
+          'http://purl.org/dc/terms/conformsTo': {
+            '@id': 'https://bioschemas.org/profiles/Organization/0.3-DRAFT',
+            '@type': 'CreativeWork'
+          },
           name: Gtn::Contributors.fetch_name(site, id),
-          description: contributor.fetch('funding_statement',
-                                         'An organization supporting the Galaxy Training Network'),
-          url: Gtn::Contributors.fetch_funding_url(contributor),
+          description: contributor.fetch('funding_statement', 'An organization supporting the Galaxy Training Network'),
+          url: contributor.fetch('url', "https://training.galaxyproject.org/training-material/hall-of-fame/#{id}/"),
+          logo: contributor.fetch('avatar', "https://github.com/#{id}.png"),
+        },
+        {
+          '@context': 'https://schema.org',
+          '@type': 'Grant',
+          identifier: contributor['funding_id'],
+          url: contributor['url'] || Gtn::Contributors.fetch_funding_url(contributor),
+          funder: {
+            '@type': 'Organization',
+            name: contributor['funder_name'],
+            description: contributor.fetch('funding_statement',
+                                           'An organization supporting the Galaxy Training Network'),
+            url: Gtn::Contributors.fetch_funding_url(contributor),
+          }
         }
-      }
-      organization['startDate'] = contributor['start_date'] if contributor.key?('start_date')
-      organization['endDate'] = contributor['end_date'] if contributor.key?('end_date')
+      ]
+
+      organization[1]['startDate'] = contributor['start_date'] if contributor.key?('start_date')
+      organization[1]['endDate'] = contributor['end_date'] if contributor.key?('end_date')
 
       organization
     end
@@ -260,7 +284,7 @@ def to_jsonld(material, topic, site)
         # "associatedMedia":,
         audience: {
           '@type': 'EducationalAudience',
-          educationalRole: 'Students'
+          educationalRole: EDU_ROLES[topic['type']]
         },
         # "audio":,
         # "award":,
@@ -344,7 +368,6 @@ def to_jsonld(material, topic, site)
         # "alternateName":,
         # "description" described below
         # "disambiguatingDescription":,
-        identifier: site['github_repository'],
         # "image":,
         # "mainEntityOfPage":,
         # "name" described below
@@ -353,8 +376,11 @@ def to_jsonld(material, topic, site)
         # "subjectOf":,
         # "url" described below
         workTranslation: [],
-        creativeWorkStatus: material['draft'] ? 'Under development' : 'Active',
+        creativeWorkStatus: material['draft'] ? 'Draft' : 'Active',
       }
+
+      data['identifier'] = "https://gxy.io/GTN:#{material['short_id']}" if material.key?('short_id')
+
       data.update(A11Y)
 
       # info depending if tutorial, hands-on or slide level
@@ -367,15 +393,17 @@ def to_jsonld(material, topic, site)
       data['isPartOf'] = topic_desc
 
       if (material['name'] == 'tutorial.md') || (material['name'] == 'slides.html')
-        if material['name'] == 'tutorial.md'
-          data['learningResourceType'] = 'hands-on tutorial'
-          data['name'] = "Hands-on for '#{material['title']}' tutorial"
-        else
-          data['learningResourceType'] = 'slides'
-          data['name'] = "Slides for '#{material['title']}' tutorial"
-        end
+        data['learningResourceType'] = if material['name'] == 'tutorial.md'
+                                         'hands-on tutorial'
+                                       else
+                                         'slides'
+                                       end
+        data['name'] = material['title']
         data['url'] = "#{site['url']}#{site['baseurl']}#{material['url']}"
 
+        # Requires https://github.com/galaxyproject/training-material/pull/4271
+        data['version'] = Gtn::ModificationTimes.obtain_modification_count(material['path'])
+
         # Time required
         if material.key?('time_estimation') && !material['time_estimation'].nil?
           data['timeRequired'] = "PT#{material['time_estimation'].upcase}"
@@ -397,8 +425,7 @@ def to_jsonld(material, topic, site)
         end
 
         # Keywords
-        data['keywords'] = [topic['name']] + (material['tags'] || [])
-        data['keywords'] = data['keywords'].join(', ')
+        data['keywords'] = [topic['title']] + (material['tags'] || [])
         # Zenodo links
         if material.key?('zenodo_link')
           mentions.push({
@@ -517,6 +544,13 @@ def to_jsonld(material, topic, site)
         data['author'] = authors
       end
 
+      # Add non-author contributors
+      if material.key?('contributions')
+        data['contributor'] = Gtn::Contributors.get_non_authors(material).map do |x|
+          generate_person_jsonld(x, site['data']['contributors'][x], site)
+        end
+      end
+
       about = []
       about.push(topic_desc)
       edam_terms = topic.fetch('edam_ontology', []) | material.fetch('edam_ontology', [])

bin/workflows-fetch.rb

@@ -15,9 +15,13 @@ def fetch_workflows(server)
     http.request(request)
   end
 
-  JSON.parse(response.body).map do |w|
-    w['server'] = server
-    w
+  begin
+    JSON.parse(response.body).map do |w|
+      w['server'] = server
+      w
+    end
+  rescue StandardError
+    []
   end
 end