Class: Relaton::Calconnect::Scraper

Inherits:
Object
  • Object
show all
Includes:
Core::ArrayWrapper, Core::HashKeysSymbolizer
Defined in:
lib/relaton/calconnect/scraper.rb

Constant Summary collapse

DOMAIN = 
"https://standards.calconnect.org/".freeze

Instance Method Summary collapse

Constructor Details

#initialize(errors = {}) ⇒ Scraper

Returns a new instance of Scraper.

Parameters:

  • errors (Hash) (defaults to: {}) —

    error tracking hash



15
16
17
 
# File 'lib/relaton/calconnect/scraper.rb', line 15

def initialize(errors = {})
  @errors = errors
end

Instance Method Details

#parse_page(hit) ⇒ Relaton::Calconnect::ItemData

Parse document page

Returns:



26
27
28
29
30
31
32
33
34
35
36
37
38
39
 
# File 'lib/relaton/calconnect/scraper.rb', line 26

def parse_page(hit) # rubocop:disable Metrics/AbcSize, Metrics/MethodLength
  hash = symbolize_hash_keys hit
  links = array(hash[:link])
  link = links.detect { |l| l[:type] == "rxl" }
  if link
    bib = fetch_bib_xml link[:content]
    update_links bib, links
  else
    hash.delete :fetched
    bib = hash_to_item hash
  end
  update_sources bib
  bib
end