bolognese 1.4.1 → 1.4.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 951d3917c38c323cd2a1d09b7d0f82fa63720406550216a9d3830f271514b5f3
-  data.tar.gz: 2c24cb509b6cf99e92d1ffd5eefa90c24ea0977d81000f8039a497ea25f9bec3
+  metadata.gz: 228903b64465ba1351f4856c7137f9a801c619d05412cd125795f69404a2debe
+  data.tar.gz: 88b6b30d92f41133f5d1a18ac656891fce74b00867f82f7a055ef90e01c6b6db
 SHA512:
-  metadata.gz: a1654b8029b926790983db5bc37684fffdc2575d7388cb1abe172476234a21cc834016693a0c211cb2d5ee69377821236686defcb064ea0fc77272a6a310d9bd
-  data.tar.gz: e360ecf63ac8c6bdbd1adbae524b5e9fefbeb0199c7953ea8c5aad320b4d04d7f4fabf9cc0490ca668ef50bb1ad12d0a9748e228e3c6b5d49f0142ea536fa41a
+  metadata.gz: 40039c3403e1f6fae2aea53913b98ad8e5c8a68a75116c11fd62a86728d1f5151c6a5ae6223dabeea95464baf778ca4ae56138ae583fa43d90a9320a47748e06
+  data.tar.gz: f2da6c0392eefe64d4c4a2e9af3cb49ff376a4df5566014e57ba69245fff37d6ac92419172f49c38e4af6b22ea27c72caf99287f95a0b76138f2a6bad40d860f

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    bolognese (1.4.1)
+    bolognese (1.4.2)
       activesupport (>= 4.2.5, < 6)
       benchmark_methods (~> 0.7)
       bibtex-ruby (~> 4.1)
@@ -30,7 +30,7 @@ PATH
 GEM
   remote: https://rubygems.org/
   specs:
-    activesupport (5.2.4)
+    activesupport (5.2.4.1)
       concurrent-ruby (~> 1.0, >= 1.0.2)
       i18n (>= 0.7, < 2)
       minitest (~> 5.1)
@@ -41,9 +41,9 @@ GEM
     bibtex-ruby (4.4.7)
       latex-decode (~> 0.0)
     builder (3.2.4)
-    citeproc (1.0.9)
+    citeproc (1.0.10)
       namae (~> 1.0)
-    citeproc-ruby (1.1.10)
+    citeproc-ruby (1.1.12)
       citeproc (~> 1.0, >= 1.0.9)
       csl (~> 1.5)
     codeclimate-test-reporter (1.0.9)
@@ -52,19 +52,19 @@ GEM
     concurrent-ruby (1.0.5)
     crack (0.4.3)
       safe_yaml (~> 1.0.0)
-    crass (1.0.5)
-    csl (1.5.0)
+    crass (1.0.6)
+    csl (1.5.1)
       namae (~> 1.0)
-    csl-styles (1.0.1.9)
+    csl-styles (1.0.1.10)
       csl (~> 1.0)
     diff-lcs (1.3)
     docile (1.1.5)
-    ebnf (1.1.3)
-      rdf (~> 3.0)
-      sxp (~> 1.0)
+    ebnf (1.2.0)
+      rdf (~> 3.1)
+      sxp (~> 1.1)
     edtf (3.0.5)
       activesupport (>= 3.0, < 7.0)
-    excon (0.67.0)
+    excon (0.71.1)
     faraday (0.17.0)
       multipart-post (>= 1.2, < 3)
     faraday-encoding (0.0.5)
@@ -80,17 +80,21 @@ GEM
       concurrent-ruby (~> 1.0)
     hashdiff (1.0.0)
     htmlentities (4.3.4)
-    i18n (1.7.0)
+    i18n (1.8.2)
       concurrent-ruby (~> 1.0)
     iso8601 (0.9.1)
     json (2.3.0)
-    json-ld (3.0.2)
-      multi_json (~> 1.12)
-      rdf (>= 2.2.8, < 4.0)
-    json-ld-preloaded (3.0.6)
-      json-ld (~> 3.0)
-      multi_json (~> 1.12)
-      rdf (~> 3.0)
+    json-canonicalization (0.2.0)
+    json-ld (3.1.0)
+      htmlentities (~> 4.3)
+      json-canonicalization (~> 0.1)
+      link_header (~> 0.0, >= 0.0.8)
+      multi_json (~> 1.14)
+      rack (~> 2.0)
+      rdf (~> 3.1)
+    json-ld-preloaded (3.1.0)
+      json-ld (~> 3.1)
+      rdf (~> 3.1)
     jsonlint (0.3.0)
       oj (~> 3)
       optimist (~> 3)
@@ -99,25 +103,25 @@ GEM
     loofah (2.4.0)
       crass (~> 1.0.2)
       nokogiri (>= 1.5.9)
-    maremma (4.4)
+    maremma (4.5)
       activesupport (>= 4.2.5, < 6)
       addressable (>= 2.3.6)
       builder (~> 3.2, >= 3.2.2)
-      excon (~> 0.67.0)
-      faraday (~> 0.17.0)
+      excon (~> 0.71.0)
+      faraday (= 0.17.0)
       faraday-encoding (~> 0.0.4)
       faraday_middleware (~> 0.13.1)
-      multi_json (~> 1.12)
       nokogiri (~> 1.10.4)
       oj (>= 2.8.3)
+      oj_mimic_json (~> 1.0, >= 1.0.1)
     mini_portile2 (2.4.0)
-    minitest (5.13.0)
+    minitest (5.14.0)
     multi_json (1.14.1)
     multipart-post (2.1.1)
     namae (1.0.1)
     nokogiri (1.10.7)
       mini_portile2 (~> 2.4.0)
-    oj (3.10.0)
+    oj (3.10.2)
     oj_mimic_json (1.0.1)
     optimist (3.0.0)
     postrank-uri (1.0.24)
@@ -125,44 +129,44 @@ GEM
       nokogiri (>= 1.8.0)
       public_suffix (>= 2.0.0, < 2.1)
     public_suffix (2.0.5)
-    rack (2.0.7)
+    rack (2.1.2)
     rack-test (0.8.3)
       rack (>= 1.0, < 3)
     rake (12.3.3)
-    rdf (3.0.13)
+    rdf (3.1.1)
       hamster (~> 3.0)
       link_header (~> 0.0, >= 0.0.8)
-    rdf-aggregate-repo (2.2.1)
-      rdf (>= 2.2, < 4.0)
-    rdf-rdfa (3.0.1)
-      haml (~> 5.0)
+    rdf-aggregate-repo (3.1.0)
+      rdf (~> 3.1)
+    rdf-rdfa (3.1.0)
+      haml (~> 5.1)
       htmlentities (~> 4.3)
-      rdf (~> 3.0)
-      rdf-aggregate-repo (>= 2.2, < 4.0)
-      rdf-xsd (~> 3.0)
+      rdf (~> 3.1)
+      rdf-aggregate-repo (~> 3.1)
+      rdf-xsd (~> 3.1)
     rdf-rdfxml (2.2.1)
       htmlentities (~> 4.3)
       rdf (>= 2.2, < 4.0)
       rdf-rdfa (>= 2.2, < 4.0)
       rdf-xsd (>= 2.2, < 4.0)
-    rdf-turtle (3.0.6)
-      ebnf (~> 1.1)
-      rdf (~> 3.0)
-    rdf-xsd (3.0.1)
-      rdf (~> 3.0)
+    rdf-turtle (3.1.0)
+      ebnf (~> 1.2)
+      rdf (~> 3.1)
+    rdf-xsd (3.1.0)
+      rdf (~> 3.1)
     rspec (3.9.0)
       rspec-core (~> 3.9.0)
       rspec-expectations (~> 3.9.0)
       rspec-mocks (~> 3.9.0)
-    rspec-core (3.9.0)
-      rspec-support (~> 3.9.0)
+    rspec-core (3.9.1)
+      rspec-support (~> 3.9.1)
     rspec-expectations (3.9.0)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.9.0)
-    rspec-mocks (3.9.0)
+    rspec-mocks (3.9.1)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.9.0)
-    rspec-support (3.9.0)
+    rspec-support (3.9.2)
     rspec-xsd (0.1.0)
       nokogiri (~> 1.6)
       rspec (~> 3)
@@ -172,17 +176,17 @@ GEM
       json (>= 1.8, < 3)
       simplecov-html (~> 0.10.0)
     simplecov-html (0.10.2)
-    sxp (1.0.2)
-      rdf (~> 3.0)
+    sxp (1.1.0)
+      rdf (~> 3.1)
     temple (0.8.2)
     thor (0.20.3)
     thread_safe (0.3.6)
     tilt (2.0.10)
-    tzinfo (1.2.5)
+    tzinfo (1.2.6)
       thread_safe (~> 0.1)
     unicode_utils (1.4.0)
     vcr (3.0.3)
-    webmock (3.7.6)
+    webmock (3.8.1)
       addressable (>= 2.3.6)
       crack (>= 0.3.2)
       hashdiff (>= 0.4.0, < 2.0.0)
@@ -204,4 +208,4 @@ DEPENDENCIES
   webmock (~> 3.0, >= 3.0.1)
 
 BUNDLED WITH
-   2.0.2
+   2.1.4

data/lib/bolognese/readers/datacite_reader.rb

@@ -133,8 +133,8 @@ module Bolognese
         end.compact
         dates = Array.wrap(meta.dig("dates", "date")).map do |r|
           if r.is_a?(Hash) && date = sanitize(r["__content__"]).presence
-            if Date.edtf(date).present?
-              { "date" => date, 
+            if Date.edtf(date).present? || Bolognese::Utils::UNKNOWN_INFORMATION.key?(date)
+              { "date" => date,
                 "dateType" => parse_attributes(r, content: "dateType"),
                 "dateInformation" => parse_attributes(r, content: "dateInformation")
               }.compact

data/lib/bolognese/utils.rb

@@ -322,6 +322,19 @@ module Bolognese
       "WebSite" => "misc"
     }
 
+    UNKNOWN_INFORMATION = {
+      ":unac" => "temporarily inaccessible",
+      ":unal" => "unallowed, suppressed intentionally",
+      ":unap" => "not applicable, makes no sense",
+      ":unas" => "value unassigned (e.g., Untitled)",
+      ":unav" => "value unavailable, possibly unknown",
+      ":unkn" => "known to be unknown (e.g., Anonymous, Inconnue)",
+      ":none" => "never had a value, never will",
+      ":null" => "explicitly and meaningfully empty",
+      ":tba" => "to be assigned or announced later",
+      ":etal" => "too numerous to list (et alia)"
+    }
+
     def find_from_format(id: nil, string: nil, ext: nil, filename: nil)
       if id.present?
         find_from_format_by_id(id)

data/lib/bolognese/version.rb

@@ -1,3 +1,3 @@
 module Bolognese
-  VERSION = "1.4.1"
+  VERSION = "1.4.2"
 end

data/lib/bolognese.rb

@@ -7,6 +7,7 @@ require 'postrank-uri'
 require 'bibtex'
 require 'colorize'
 require 'loofah'
+require 'json/ld/preloaded'
 require 'json/ld'
 require 'rdf/turtle'
 require 'rdf/rdfxml'

data/spec/fixtures/datacite-example-complicated-tba.xml

@@ -0,0 +1,56 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<resource xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://datacite.org/schema/kernel-4" xsi:schemaLocation="http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4/metadata.xsd">
+  <identifier identifierType="DOI">10.5072/testpub</identifier>
+  <creators>
+    <creator>
+      <creatorName nameType="Personal">Smith, John</creatorName>
+    </creator>
+    <creator>
+      <creatorName>つまらないものですが</creatorName>
+      <nameIdentifier nameIdentifierScheme="ISNI" schemeURI="http://isni.org/isni/">0000000134596520</nameIdentifier>
+    </creator>
+  </creators>
+  <titles>
+    <title>:unav</title>
+  </titles>
+  <publisher>Springer</publisher>
+  <publicationYear>2010</publicationYear>
+  <subjects>
+    <subject subjectScheme="DDC">830 German &amp; related literatures</subject>
+    <subject>Polish Literature</subject>
+  </subjects>
+  <dates>
+    <date dateType="Other" dateInformation="Correction">:tba</date>
+  </dates>
+  <contributors>
+    <contributor contributorType="DataCollector">
+      <contributorName>Doe, John</contributorName>
+      <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="http://orcid.org/">0000-0001-5393-1421</nameIdentifier>
+    </contributor>
+  </contributors>
+  <language>GER</language>
+  <resourceType resourceTypeGeneral="Text">Monograph</resourceType>
+  <alternateIdentifiers>
+    <alternateIdentifier alternateIdentifierType="ISBN">937-0-4523-12357-6</alternateIdentifier>
+  </alternateIdentifiers>
+  <relatedIdentifiers>
+    <relatedIdentifier resourceTypeGeneral="Text" relatedIdentifierType="DOI" relationType="IsPartOf">10.5272/oldertestpub</relatedIdentifier>
+  </relatedIdentifiers>
+  <sizes>
+    <size>256 pages</size>
+  </sizes>
+  <formats>
+    <format>:null</format>
+  </formats>
+  <version>2</version>
+  <rightsList>
+    <rights xml:lang="eng" rightsURI="http://creativecommons.org/licenses/by-nd/2.0/">Creative Commons Attribution-NoDerivs 2.0 Generic</rights>
+  </rightsList>
+  <descriptions>
+    <description descriptionType="Abstract">
+      Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam et justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea
+      takimata sanctus est Lorem ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam et justo duo dolores
+      et ea rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem ipsum dolor sit amet.
+    </description>
+  </descriptions>
+</resource>

data/spec/fixtures/datacite-example-escaped-text.xml

@@ -0,0 +1,56 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<resource xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://datacite.org/schema/kernel-4" xsi:schemaLocation="http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4/metadata.xsd">
+  <identifier identifierType="DOI">10.5072/testpub</identifier>
+  <creators>
+    <creator>
+      <creatorName nameType="Personal">Smith, John</creatorName>
+    </creator>
+    <creator>
+      <creatorName>つまらないものですが</creatorName>
+      <nameIdentifier nameIdentifierScheme="ISNI" schemeURI="http://isni.org/isni/">0000000134596520</nameIdentifier>
+    </creator>
+  </creators>
+  <titles>
+    <title>Some initial text&lt;the rest of the text won't display.</title>
+  </titles>
+  <publisher>Springer</publisher>
+  <publicationYear>2010</publicationYear>
+  <subjects>
+    <subject subjectScheme="DDC">830 German &amp; related literatures</subject>
+    <subject>Polish Literature</subject>
+  </subjects>
+  <dates>
+    <date dateType="Other" dateInformation="Correction">2012-12-13</date>
+  </dates>
+  <contributors>
+    <contributor contributorType="DataCollector">
+      <contributorName>Doe, John</contributorName>
+      <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="http://orcid.org/">0000-0001-5393-1421</nameIdentifier>
+    </contributor>
+  </contributors>
+  <language>GER</language>
+  <resourceType resourceTypeGeneral="Text">Monograph</resourceType>
+  <alternateIdentifiers>
+    <alternateIdentifier alternateIdentifierType="ISBN">937-0-4523-12357-6</alternateIdentifier>
+  </alternateIdentifiers>
+  <relatedIdentifiers>
+    <relatedIdentifier resourceTypeGeneral="Text" relatedIdentifierType="DOI" relationType="IsPartOf">10.5272/oldertestpub</relatedIdentifier>
+  </relatedIdentifiers>
+  <sizes>
+    <size>256 pages</size>
+  </sizes>
+  <formats>
+    <format>pdf</format>
+  </formats>
+  <version>2</version>
+  <rightsList>
+    <rights xml:lang="eng" rightsURI="http://creativecommons.org/licenses/by-nd/2.0/">Creative Commons Attribution-NoDerivs 2.0 Generic</rights>
+  </rightsList>
+  <descriptions>
+    <description descriptionType="Abstract">
+      Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam et justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea
+      takimata sanctus est Lorem ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam et justo duo dolores
+      et ea rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem ipsum dolor sit amet.
+    </description>
+  </descriptions>
+</resource>