relaton-nist 0.5.0 → 0.5.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: af67a70401309cab1e7d7b4511db20533a82927fc71aa4c439ee895bce140d2a
-  data.tar.gz: aa32fd8f52b1b3513153fdf6ee2d0898d16dfe4e86b2890303002aa0720274d3
+  metadata.gz: 84efea59e2f35758f44b21e42574ebc3bfbda66f95d7d1c905ae645d42d55711
+  data.tar.gz: c58d8aa8f2bc6da70c04793d196f90b4b5a0ab7868b97f7ef17fc3df3ff5f17b
 SHA512:
-  metadata.gz: d47764a549b16704b6672bed22add68438e0ef42be86f97a0d71dc2d3dbb8b371b96b1e120402ca2f9c3631ab57134dcf9341540b669bcdc7dcea11174a20336
-  data.tar.gz: 22e80353033d4b57d754a3b48b5515370cc1512a134bb534897c0e75cdcf72a3f1fe5444cc9e724aa6ac57bc6996ab825426547a513ebe4eb264cbecd03f354d
+  metadata.gz: 22d62f0a3a04d03efa7ca6bf2f8d2dda988da7346fac312ac58d43998a712f9bd35bf2481f2a6aebfd4507dcc4b6e8d9f75bc10de0fd51e5e2e9f22edeeecad2
+  data.tar.gz: 8193871dba1c5c64ba39839ed38eb34632c653ef55daf4e6850a5d4fe73446af64d27fd7eb2170ed73d0b79efb96ddbb8fee37871b986422b4ea0c1586157b7d

data/.gitignore

@@ -13,3 +13,4 @@
 .rspec_status
 pubs-export.json
 Gemfile.lock
+.byebug_history

data/lib/relaton_nist/scrapper.rb

@@ -87,7 +87,12 @@ module RelatonNist
       # @return [Array<Nokogiri::HTML::Document, String>]
       def get_page(url)
         uri = URI url
-        resp = Net::HTTP.get_response(uri) # .encode("UTF-8")
+        resp = Net::HTTP.get_response(uri)
+        %r{(?<=newLocation = 'https://' \+ window.location.hostname \+ ')(?<path>[^']+)} =~ resp.body
+        if path
+          uri = URI HitCollection::DOMAIN + path
+          resp = Net::HTTP.get_response(uri)
+        end
         Nokogiri::HTML(resp.body)
       rescue SocketError, Timeout::Error, Errno::EINVAL, Errno::ECONNRESET, EOFError,
              Net::HTTPBadResponse, Net::HTTPHeaderSyntaxError, Net::ProtocolError,
@@ -104,7 +109,7 @@ module RelatonNist
                    else
                      doc.at(
                        "//div[contains(@class, 'publications-detail')]/h3",
-                     )&.text&.strip
+                     )&.text&.strip.sub(/(?<=\w)\([^\)]+\)$/) { |m| " " + m.upcase}.squeeze " "
                    end
         item_ref ||= "?"
         [RelatonBib::DocumentIdentifier.new(id: item_ref, type: "NIST")]
@@ -138,7 +143,7 @@ module RelatonNist
           when "withdrawn"
             stage = "final"
             subst = "withdrawn"
-          when "draft"
+          when /^draft/
             stage = "draft-public"
             subst = "active"
           else
@@ -352,9 +357,9 @@ module RelatonNist
           links << { type: "uri", content: doc["uri"] } if doc["uri"]
           doi = "https://doi.org/" + doc["doi"] if doc["doi"]
         else
-          pub = doc.at "//p/strong[.='Publication:']"
-          pdf = pub.at "./following-sibling::a[.=' Local Download']"
-          doi = pub.at("./following-sibling::a[contains(.,'(DOI)')]")&.attr :href
+          pub = doc.at "//p/strong[contains(., 'Publication:')]"
+          pdf = pub&.at "./following-sibling::a[.=' Local Download']"
+          doi = pub&.at("./following-sibling::a[contains(.,'(DOI)')]")&.attr :href
           links << { type: "pdf", content: pdf[:href] } if pdf
         end
         links << { type: "doi", content: doi } if doi
@@ -417,10 +422,12 @@ module RelatonNist
                  else idx + 1
                  end
 
-          content = s.text.match(/^[^\(]+/).to_s.strip.gsub "  ", " "
+          content = s.text.match(/^[^\(]+/).to_s.strip.squeeze " "
 
-          ref = case content.match(/\w+/).to_s
-                when "Draft" then content.match(/(?<=Draft\s).+/).to_s + " (#{iter}PD)"
+          ref = case s.text
+                when /^Draft/ then content.match(/(?<=Draft\s).+/).to_s + " (#{iter}PD)"
+                when /\(Draft\)/ then content + " (#{iter}PD)"
+                else content
                 end
 
           fref = RelatonBib::FormattedRef.new(

data/lib/relaton_nist/version.rb

@@ -1,3 +1,3 @@
 module RelatonNist
-  VERSION = "0.5.0".freeze
+  VERSION = "0.5.1".freeze
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: relaton-nist
 version: !ruby/object:Gem::Version
-  version: 0.5.0
+  version: 0.5.1
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2020-01-11 00:00:00.000000000 Z
+date: 2020-01-16 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: byebug