Class: Html2rss::AutoSource::Segmenter::PrimaryLink

Inherits:
Object
  • Object
show all
Defined in:
lib/html2rss/auto_source/segmenter/primary_link.rb

Overview

Selects the strongest content-like primary link inside a container (port of Discovery::SemanticAnchorCandidates).

Instance Method Summary collapse

Constructor Details

#initialize(segmenter) ⇒ PrimaryLink

Returns a new instance of PrimaryLink.

Parameters:



11
12
13
14
15
16
# File 'lib/html2rss/auto_source/segmenter/primary_link.rb', line 11

def initialize(segmenter)
  @segmenter = segmenter
  @index = segmenter.index
  @link_resolver = segmenter.link_resolver
  @noise_policy = segmenter.noise_policy
end

Instance Method Details

#select(container) ⇒ SST::Node?

Parameters:

Returns:



21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
# File 'lib/html2rss/auto_source/segmenter/primary_link.rb', line 21

def select(container) # rubocop:disable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/PerceivedComplexity
  winners = {}

  container.find_all(&:link?).each do |anchor|
    next if @index.ignored_chrome?(anchor)

    facts = candidate_facts(anchor, container)
    next unless facts

    dest = facts[:destination]
    current = winners[dest]
    winners[dest] = facts if current.nil? || facts[:score] > current[:score]
  end

  winners.values.max_by { |f| f[:score] }&.fetch(:anchor)
end