Class: Yarrow::Content::CollectionExpander

Inherits:
Object
  • Object
show all
Includes:
Tools::FrontMatter
Defined in:
lib/yarrow/content/collection_expander.rb

Instance Method Summary collapse

Methods included from Tools::FrontMatter

#extract_split_content, #read_split_content

Constructor Details

#initialize(object_types = nil) ⇒ CollectionExpander

If a list of object types is not provided, a default pages type is created.



8
9
10
11
12
 
# File 'lib/yarrow/content/collection_expander.rb', line 8

def initialize(object_types=nil)
  @object_types = object_types || [
    Yarrow::Content::ObjectType.from_name(:pages)
  ]
end

Instance Method Details

#build_content_nodes(graph, objects, type, parent_index) ⇒ Object 



163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
 
# File 'lib/yarrow/content/collection_expander.rb', line 163

def build_content_nodes(graph, objects, type, parent_index)
  # TODO: this may need to use a strategy that can be overriden
  content_type = Yarrow::Symbols.to_singular(type)

  # Process collected content objects and generate entity nodes
  objects.each do |name, sources|
    item_node = graph.create_node do |node|
      # TODO: Rename this to :entry and support similar fields to Atom
      node.label = :item
      node.props[:name] = name
      node.props[:type] = content_type

      meta = {}
      content = ""

      sources.each do |source|
        body, data = process_content(source.props[:entry])
        meta.merge!(data) unless data.nil?
        content << body unless body.nil?
      end

      if meta[:url]
        # If a URL is explicitly proided in metadata then use it
        node.props[:url] = meta[:url]
      elsif meta[:permalink]
        # Support for legacy permalink attribute
        node.props[:url] = meta[:permalink]
      else
        # Default URL generation strategy when no explicit URL is provided
        # TODO: collection nodes need URL generation too
        # TODO: replace this with URL generation strategy
        # TODO: slug vs name - why do some nodes have 2 and some 3 props?
        node.props[:url] = if parent_index.props[:name].to_sym == parent_index.props[:type]
          "/#{parent_index.props[:type]}/#{name}"
        else
          "/#{parent_index.props[:type]}/#{parent_index.props[:slug]}/#{name}"
        end
      end

      # For now, we are storing title, url, etc on the top-level item.
      node.props[:title] = meta[:title]

      # TODO: What belongs on the entity and what belongs on the item?
      entity_props = meta.merge(body: content, name: meta[:id], url: node.props[:url])


      # TODO: consider whether to provide `body` on the item/document or at
      # the custom content type level.
      begin
        content_struct = Yarrow::Symbols.to_const(content_type)
      rescue
        # No immutable struct found: fall back to slower dynamically typed open struct
        require "ostruct"
        content_struct = OpenStruct
      end

      node.props[:entity] = content_struct.new(entity_props)
    end

    # Connect entity with source content
    sources.each do |source|
      graph.create_edge do |edge|
        edge.label = :source
        edge.from = item_node
        edge.to = source.id
      end
    end

    # Connect entity with parent collection
    graph.create_edge do |edge|
      edge.label = :child
      edge.from = parent_index
      edge.to = item_node
    end
  end
end

#expand(graph) ⇒ Object 



14
15
16
17
18
 
# File 'lib/yarrow/content/collection_expander.rb', line 14

def expand(graph)
  @object_types.each do |object_type|
    expand_nested(graph, object_type)
  end
end

#expand_nested(graph, content_type) ⇒ Object 



20
21
22
23
 
# File 'lib/yarrow/content/collection_expander.rb', line 20

def expand_nested(graph, content_type)
  strategy = TreeExpansion.new(graph)
  strategy.expand(content_type)
end

#expand_nested_legacy(graph, content_type) ⇒ Object 



25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
 
# File 'lib/yarrow/content/collection_expander.rb', line 25

def expand_nested_legacy(graph, content_type)
  type = content_type.collection
  exts = content_type.extensions

  # If match path represents entire content dir, then include the entire
  # content dir instead of scanning from a subfolder matching the name of
  # the collection.
  start_node = if content_type.match_path == "."
    graph.n(:root)
  else
    graph.n(:root).out(name: type.to_s)
  end

  # Extract metadata from given start node
  data = extract_metadata(start_node, type)

  # Collects all nested collections in the subgraph for this content type
  subcollections = {}
  index = nil

  # Define alias for accessing metadata in the loop
  metadata = data

  # Scan and collect all nested directories under the top level source
  start_node.depth_first.each do |node|
    if node.label == :directory
      # Check if this entry has metadata defined at the top level
      if data[:collections]
        item = data[:collections].find { |c| c[:slug] == node.props[:slug] }
        metadata = item if item
      end

      # Create a collection node representing a collection of documents
      index = graph.create_node do |collection_node|
        collection_node.label = :collection
        collection_node.props[:type] = type
        collection_node.props[:name] = node.props[:name]
        collection_node.props[:slug] = node.props[:slug]
        collection_node.props[:title] = metadata[:title]

        # Override default status so that mapped index collections always show
        # up in the resulting document manifest, when they don’t have
        # associated metadata. This is the opposite of how individual pieces
        # of content behave (default to draft status if one isn’t supplied).
        collection_node.props[:status] = if data[:status]
          data[:status]
        else
          "published"
        end

        # TODO: URL generation might need to happen elsewhere
        collection_node.props[:url] = if data[:url]
          data[:url]
        else
          "#{node.props[:path].split('./content').last}/"
        end
      end

      # Add this collection id to the lookup table for edge construction
      subcollections[node.props[:path]] = index

      # Join the collection to its parent
      unless node.props[:slug] == type.to_s || !subcollections.key?(node.props[:entry].parent.to_s)
        graph.create_edge do |edge|
          edge.label = :child
          edge.from = subcollections[node.props[:entry].parent.to_s].id
          edge.to = index.id
        end
      end
    end
  end

  # If there are no subcollections then we need to look at the start node
  # TODO: test to verify if this could be used in all cases, not just
  # the situation where there are subfolders to be mapped.
  if subcollections.empty?
    # Collect files that match the content type extension and group them
    # under a common key for each slug (this is so we can merge multiple
    # files with the same name together into a single content type, a
    # specific pattern found in some legacy content folders).
    #
    # Ideally, this code should be deleted once we have a clean workflow
    # and can experiment with decoupling different strategies for
    # expansion/enrichment of content objects.
    objects = start_node.out(:file).all.select do |file|
      file.props[:name].end_with?(*exts)
    end.group_by do |file|
      file.props[:slug]
    end

    # This is a massive hack to deal with situations where we don’t
    # recurse down the list of directories. The best way to clean it up
    # will be to document the different supported mapping formats and
    # URL generation strategies and break these up into separate
    # traversal objects for each particular style of content organisation.
    if index.nil?
      index = graph.create_node do |collection_node|
        collection_node.label = :collection
        collection_node.props[:type] = type
        collection_node.props[:name] = type
        collection_node.props[:slug] = type.to_s
        collection_node.props[:title] = metadata[:title]

        # Override default status so that mapped index collections always show
        # up in the resulting document manifest, when they don’t have
        # associated metadata. This is the opposite of how individual pieces
        # of content behave (default to draft status if one isn’t supplied).
        collection_node.props[:status] = if data[:status]
          data[:status]
        else
          "published"
        end

        # TODO: URL generation might need to happen elsewhere
        collection_node.props[:url] = if data[:url]
          data[:url]
        else
          "/#{type}/"
        end
      end
    end

    build_content_nodes(graph, objects, type, index)
  end

  # Go through each subcollection and expand content nodes step by step.
  subcollections.each do |path, index|
    # Group files matching the same slug under a common key
    objects = graph.n(path: path).out(:file).all.select do |file|
      file.props[:name].end_with?(*exts)
    end.group_by do |file|
      file.props[:slug]
    end

    build_content_nodes(graph, objects, type, index)
  end
end