Class: Relaton::Cie::DataFetcher
- Inherits:
-
Relaton::Core::DataFetcher
- Object
- Relaton::Core::DataFetcher
- Relaton::Cie::DataFetcher
- Defined in:
- lib/relaton/cie/data_fetcher.rb
Constant Summary collapse
- URL =
"https://www.techstreet.com/cie/searches/31156444?page=1&per_page=100"- RETRIABLE_ERRORS =
[ SocketError, Ferrum::TimeoutError, Ferrum::PendingConnectionsError, Ferrum::StatusError ].freeze
Instance Method Summary collapse
- #agent ⇒ Object
- #fetch(_source = nil) ⇒ Object
- #fetch_abstract(doc) ⇒ Array<Relaton::Bib::LocalizedMarkedUpString>
- #fetch_contributor(doc) ⇒ Array<Relaton::Bib::Contributor>
- #fetch_date(doc) ⇒ Array<Relaton::Bib::Date>
- #fetch_doc(url = URL) ⇒ Object
- #fetch_docid(hit, doc) ⇒ Array<Relaton::Bib::Docidentifier>
- #fetch_docnumber(hit) ⇒ Object
- #fetch_doctype ⇒ Object
- #fetch_edition(doc) ⇒ String
- #fetch_ext ⇒ Object
- #fetch_relation(doc) ⇒ Array<Relaton::Cie::Relation>
- #fetch_source(url) ⇒ Array<Relaton::Bib::Uri>
- #fetch_title(doc) ⇒ Array<Relaton::Bib::Title>
- #index ⇒ Object
- #log_error(msg) ⇒ Object
-
#parse_cie_code(code1, code2, doc = nil) ⇒ Object
rubocop:disable Metrics/CyclomaticComplexity.
- #parse_code(hit, doc = nil) ⇒ Object
- #parse_page(hit) ⇒ Object
- #primary_code(code, doc = nil) ⇒ Object
- #time_req ⇒ Object
- #to_bibxml(bib) ⇒ Object
- #to_xml(bib) ⇒ Object
- #to_yaml(bib) ⇒ Object
- #write_file(bib) ⇒ Object
Instance Method Details
#agent ⇒ Object
77 78 79 |
# File 'lib/relaton/cie/data_fetcher.rb', line 77 def agent @agent ||= BrowserAgent.new end |
#fetch(_source = nil) ⇒ Object
313 314 315 316 317 318 |
# File 'lib/relaton/cie/data_fetcher.rb', line 313 def fetch(_source = nil) fetch_doc report_errors ensure @agent&.quit end |
#fetch_abstract(doc) ⇒ Array<Relaton::Bib::LocalizedMarkedUpString>
220 221 222 223 224 225 226 227 228 229 230 |
# File 'lib/relaton/cie/data_fetcher.rb', line 220 def fetch_abstract(doc) content = doc.at('//div[contains(@class,"description")]')&.text&.strip if content.nil? || content.empty? @errors[:abstract] &&= true return [] end result = [Bib::Abstract.new(content: content, language: "en", script: "Latn")] @errors[:abstract] &&= result.empty? result end |
#fetch_contributor(doc) ⇒ Array<Relaton::Bib::Contributor>
234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 |
# File 'lib/relaton/cie/data_fetcher.rb', line 234 def fetch_contributor(doc) # rubocop:disable Metrics/AbcSize,Metrics/MethodLength,Metrics/CyclomaticComplexity = doc.xpath('//hgroup/p[not(@class="pub_date")]').text.gsub "\"", "" contribs = [] until .empty? /^(?<sname1>\S+(?:\sder?\s)?[^\s,]+) (?:,?\s(?<sname2>[\w-]{2,})(?=,\s+\w\.))? (?:,?\s(?<fname>W-T\.[\w-]{2,})(?!,\s+\w\.))? (?:(?:\s?,\s?|\s)(?<init>(?:\w(?:\s?\.|\s|,|$)[\s-]?)+))? (?:(?:[,;]\s*|\s+|\.|(?<=\s))(?:and\s)?)?/x =~ raise StandardError, "Author name not found in \"#{}\"" unless $LAST_MATCH_INFO .sub! $LAST_MATCH_INFO.to_s, "" sname = [sname1, sname2].compact.join " " surname = Bib::LocalizedString.new content: sname, language: "en", script: "Latn" forename = [] forename << Bib::FullNameType::Forename.new(content: fname, language: "en", script: "Latn") if fname (init&.strip || "").split(/(?:,|\.)(?:-|\s)?/).each do |int| forename << Bib::FullNameType::Forename.new(content: "", initial: int.strip, language: "en", script: "Latn") end fullname = Bib::FullName.new surname: surname, forename: forename person = Bib::Person.new name: fullname role = Bib::Contributor::Role.new type: "author" contribs << Bib::Contributor.new(person: person, role: [role]) @errors[:contributor_author] &&= contribs.empty? end org_name = Bib::TypedLocalizedString.new(content: "Commission Internationale de L'Eclairage") abbrev = Bib::LocalizedString.new content: "CIE" org_uri = Bib::Uri.new content: "cie.co.at" org = Bib::Organization.new(name: [org_name], abbreviation: abbrev, uri: [org_uri]) org_role = Bib::Contributor::Role.new type: "publisher" contribs << Bib::Contributor.new(organization: org, role: [org_role]) end |
#fetch_date(doc) ⇒ Array<Relaton::Bib::Date>
165 166 167 168 169 170 171 172 173 |
# File 'lib/relaton/cie/data_fetcher.rb', line 165 def fetch_date(doc) result = doc.xpath("//h3[.='Published:']/following-sibling::span").map do |d| pd = d.text.strip on = pd.match?(/^\d{4}(?:[^-]|$)/) ? pd : Date.strptime(pd, "%m/%d/%Y").strftime("%Y-%m-%d") Bib::Date.new(type: "published", at: on) end @errors[:date] &&= result.empty? result end |
#fetch_doc(url = URL) ⇒ Object
320 321 322 323 324 325 326 327 328 329 330 331 |
# File 'lib/relaton/cie/data_fetcher.rb', line 320 def fetch_doc(url = URL) result = time_req { agent.get url } result.xpath("//li[@data-product]").each { |hit| parse_page hit } np = result.at '//a[@class="next_page"]' if np next_href = np[:href] next_url = next_href.start_with?("http") ? next_href : "https://www.techstreet.com#{next_href}" fetch_doc next_url else index.save end end |
#fetch_docid(hit, doc) ⇒ Array<Relaton::Bib::Docidentifier>
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 |
# File 'lib/relaton/cie/data_fetcher.rb', line 92 def fetch_docid(hit, doc) code, code2 = parse_code hit, doc docid = [] if code && !code.strip.empty? docid << Bib::Docidentifier.new(type: "CIE", content: code, primary: true) @errors[:docid_1] &&= false else @errors[:docid_1] &&= true end if code2 && !code2.strip.empty? type2 = code2.match(/\w+/).to_s docid << Relaton::Bib::Docidentifier.new(type: type2, content: code2.strip) @errors[:docid_2] &&= false else @errors[:docid_2] &&= true end isbn = doc.at('//h3[contains(.,"ISBN")]/following-sibling::span')&.text if isbn && !isbn.strip.empty? docid << Bib::Docidentifier.new(type: "ISBN", content: isbn) @errors[:docid_isbn] &&= false else @errors[:docid_isbn] &&= true end docid end |
#fetch_docnumber(hit) ⇒ Object
145 146 147 |
# File 'lib/relaton/cie/data_fetcher.rb', line 145 def fetch_docnumber(hit) parse_code(hit).first.sub(/^CIE\s(?:ISO\s)?/, "") end |
#fetch_doctype ⇒ Object
271 272 273 |
# File 'lib/relaton/cie/data_fetcher.rb', line 271 def fetch_doctype Bib::Doctype.new(content: "document") end |
#fetch_edition(doc) ⇒ String
177 178 179 180 181 182 183 184 185 186 187 |
# File 'lib/relaton/cie/data_fetcher.rb', line 177 def fetch_edition(doc) ed = doc.at("//h3[.='Edition:']/following-sibling::span") @errors[:edition] &&= true return unless ed content = ed.text.slice(/^\d+(?=(st|nd|rd|th))/) if content @errors[:edition] = false Bib::Edition.new(content: content) end end |
#fetch_ext ⇒ Object
267 268 269 |
# File 'lib/relaton/cie/data_fetcher.rb', line 267 def fetch_ext Ext.new(doctype: fetch_doctype, flavor: "cie") end |
#fetch_relation(doc) ⇒ Array<Relaton::Cie::Relation>
191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 |
# File 'lib/relaton/cie/data_fetcher.rb', line 191 def fetch_relation(doc) # rubocop:disable Metrics/AbcSize,Metrics/MethodLength rels = doc.xpath('//section[@class="history"]/ol/li[not(contains(@class,"selected-product"))]').map do |rel| ref = rel.at("a") url = "https://www.techstreet.com#{ref[:href]}" title = Bib::Title.from_string ref.at('p/span[@class="title"]').text did = ref.at("h3").text docid = [Bib::Docidentifier.new(type: "CIE", content: did, primary: true)] on = ref.at("p/time") date = [Bib::Date.new(type: "published", at: on[:datetime])] source = [Bib::Uri.new(type: "src", content: url)] bibitem = ItemData.new docidentifier: docid, title: title, source: source, date: date type = ref.at('//li/i[contains(@class,"historical")]') ? "updates" : "updatedBy" Bib::Relation.new(type: type, bibitem: bibitem) end @errors[:relation] &&= rels.empty? rels end |
#fetch_source(url) ⇒ Array<Relaton::Bib::Uri>
211 212 213 214 215 216 |
# File 'lib/relaton/cie/data_fetcher.rb', line 211 def fetch_source(url) @errors[:source] &&= url.nil? || url.empty? return [] if url.nil? || url.empty? [Bib::Uri.new(type: "src", content: url)] end |
#fetch_title(doc) ⇒ Array<Relaton::Bib::Title>
151 152 153 154 155 156 157 158 159 160 161 |
# File 'lib/relaton/cie/data_fetcher.rb', line 151 def fetch_title(doc) t = doc.at("//hgroup/h2/text()", "//hgroup/h1/text()") unless t && !t.text.strip.empty? @errors[:title] &&= true return [] end result = Bib::Title.from_string t.text.strip @errors[:title] &&= result.empty? result end |
#index ⇒ Object
81 82 83 |
# File 'lib/relaton/cie/data_fetcher.rb', line 81 def index @index ||= Index.find_or_create :cie, file: "index-v1.yaml" end |
#log_error(msg) ⇒ Object
85 86 87 |
# File 'lib/relaton/cie/data_fetcher.rb', line 85 def log_error(msg) Util.error msg end |
#parse_cie_code(code1, code2, doc = nil) ⇒ Object
rubocop:disable Metrics/CyclomaticComplexity
137 138 139 140 141 142 143 |
# File 'lib/relaton/cie/data_fetcher.rb', line 137 def parse_cie_code(code1, code2, doc = nil) # rubocop:disable Metrics/CyclomaticComplexity code = code1.size > 25 && code2 ? "CIE #{code2.sub(/,(\sPages)?/, '')}" : code1 add = doc&.at("//hgroup/h2")&.text&.match(/(Add)endum\s(\d+)$/) return code unless add "#{code} #{add[1]} #{add[2]}" end |
#parse_code(hit, doc = nil) ⇒ Object
118 119 120 121 122 123 |
# File 'lib/relaton/cie/data_fetcher.rb', line 118 def parse_code(hit, doc = nil) code = hit.at("h3/a").text.strip.squeeze(" ").sub(/\u25b9/, "").gsub(" / ", "/") c2idx = %r{(?:\(|/)(?<c2>(?:ISO|IEC)\s[^()]+)} =~ code code = code[0...c2idx].strip if c2idx [primary_code(code, doc), c2] end |
#parse_page(hit) ⇒ Object
295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 |
# File 'lib/relaton/cie/data_fetcher.rb', line 295 def parse_page(hit) # rubocop:disable Metrics/AbcSize,Metrics/MethodLength url = hit.at('h3/a')[:href] doc = time_req { agent.get url } item = ItemData.new( type: "standard", source: fetch_source(url), docnumber: fetch_docnumber(hit), docidentifier: fetch_docid(hit, doc), title: fetch_title(doc), abstract: fetch_abstract(doc), date: fetch_date(doc), edition: fetch_edition(doc), contributor: fetch_contributor(doc), relation: fetch_relation(doc), language: "en", script: "Latn", ext: fetch_ext ) write_file item rescue StandardError => e Util.error do "Document: #{url}\n#{e.}\n#{e.backtrace}" end end |
#primary_code(code, doc = nil) ⇒ Object
125 126 127 128 129 130 131 132 133 134 135 |
# File 'lib/relaton/cie/data_fetcher.rb', line 125 def primary_code(code, doc = nil) /^(?<code1>[^(]+)(?:\((?<code2>[a-zA-Z]+\d+,(?:\sPages)?[^)]+))?/ =~ code if code1&.match?(/^CIE/) parse_cie_code code1, code2, doc elsif (pcode = doc&.at('//h3[.="Product Code(s):"]/following-sibling::span')) "CIE #{pcode.text.strip.match(/[^,]+/)}" else num = code.match(/(?<=\()\w{2}\d+,.+(?=\))/).to_s.gsub(/,(?=\s)/, "").gsub(/,(?=\S)/, " ") "CIE #{num}" end end |
#time_req ⇒ Object
340 341 342 343 344 345 346 347 348 349 350 351 352 |
# File 'lib/relaton/cie/data_fetcher.rb', line 340 def time_req tries = 0 begin tries += 1 sleep [4 - (Time.now - @last_request_time).to_i, 0].max if @last_request_time yield rescue *RETRIABLE_ERRORS => e retry if tries < 4 raise e ensure @last_request_time = Time.now end end |
#to_bibxml(bib) ⇒ Object
292 |
# File 'lib/relaton/cie/data_fetcher.rb', line 292 def to_bibxml(bib) = bib.to_rfcxml |
#to_xml(bib) ⇒ Object
290 |
# File 'lib/relaton/cie/data_fetcher.rb', line 290 def to_xml(bib) = bib.to_xml(bibdata: true) |
#to_yaml(bib) ⇒ Object
291 |
# File 'lib/relaton/cie/data_fetcher.rb', line 291 def to_yaml(bib) = bib.to_yaml |
#write_file(bib) ⇒ Object
276 277 278 279 280 281 282 283 284 285 286 287 288 |
# File 'lib/relaton/cie/data_fetcher.rb', line 276 def write_file(bib) # rubocop:disable Metrics/AbcSize,Metrics/MethodLength id = bib.docidentifier[0].content file = output_file id if @files.include? file Util.warn do "File #{file} exists. Docid: #{bib.docidentifier[0].content}\n" \ "Link: #{bib.source.detect { |l| l.type == 'src' }.content}" end else @files << file end index.add_or_update bib.docidentifier[0].content, file File.write file, serialize(bib), encoding: "UTF-8" end |