commonmeta-ruby 3.9.0 → 3.11.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: cc0c7f6508a3b1d909c7726fa761c35f8fa213112bebded5779f75b200dbbac7
-  data.tar.gz: 1030041c2ec0ea94c01e3f079cb6aa3ce8a5b5aebd3c270a71341bf4e87396fe
+  metadata.gz: 7bc4df0a364fc1ff53342960355c5fc534a0c479cf825cc0d856e282e42e7d49
+  data.tar.gz: 57dcd210498057807f46ffb851eb40b4db189636d915152a77de1fd1191012c9
 SHA512:
-  metadata.gz: a5446caf7d8cb8a32cf8fe4dbddbd11bade0cd68d577d80302c291540d61564b4268b51d846e406486b4ec26a1ae4682326fb79edf5d66de115f91ea845c5db0
-  data.tar.gz: 622d33b224c37ea1a131aad8efeef29ef9a80c4d536012fee61ca982b58dfca784a2edfaeed5d889f5a93b52d6d13da978d89e49000f50120cb8b9d71aa776a3
+  metadata.gz: 2a4265c7a8d17ab99b963459015232400bcd7541e4c827e2bf0863e9809fbe64d70d0ce6ced7dbcf2a9186beca2075be73617bd3d3cace2e4b9c67e39d8e756a
+  data.tar.gz: 765b9d3d29683badac4ed6b77ac6f86f6b939f5081d23016c1484a9f8b7e0d26039559a9f6d5c2e2b8fe5c06bc1d70d5e09e17f16a5a24c6c9bb84291a8c1548

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    commonmeta-ruby (3.9.0)
+    commonmeta-ruby (3.11.0)
       activesupport (>= 4.2.5, < 8.0)
       addressable (~> 2.8.1, < 2.8.2)
       base32-url (>= 0.7.0, < 1)
@@ -27,7 +27,7 @@ PATH
 GEM
   remote: https://rubygems.org/
   specs:
-    activesupport (7.1.2)
+    activesupport (7.1.3)
       base64
       bigdecimal
       concurrent-ruby (~> 1.0, >= 1.0.2)
@@ -47,7 +47,7 @@ GEM
     bibtex-ruby (6.1.0)
       latex-decode (~> 0.0)
       racc (~> 1.7)
-    bigdecimal (3.1.5)
+    bigdecimal (3.1.6)
     builder (3.2.4)
     citeproc (1.0.10)
       namae (~> 1.0)
@@ -56,7 +56,7 @@ GEM
       csl (~> 2.0)
     code-scanning-rubocop (0.6.1)
       rubocop (~> 1.0)
-    concurrent-ruby (1.2.2)
+    concurrent-ruby (1.2.3)
     connection_pool (2.4.1)
     crack (0.4.5)
       rexml
@@ -132,7 +132,7 @@ GEM
       crass (~> 1.0.2)
       nokogiri (>= 1.12.0)
     matrix (0.4.2)
-    minitest (5.21.1)
+    minitest (5.21.2)
     multi_json (1.15.0)
     mutex_m (0.2.0)
     namae (1.2.0)
@@ -143,7 +143,7 @@ GEM
       bigdecimal (>= 3.0)
     optimist (3.1.0)
     parallel (1.24.0)
-    parser (3.3.0.3)
+    parser (3.3.0.5)
       ast (~> 2.4.1)
       racc
     postrank-uri (1.1)
@@ -191,11 +191,11 @@ GEM
     rspec-xsd (0.1.0)
       nokogiri (~> 1.6)
       rspec (~> 3)
-    rubocop (1.59.0)
+    rubocop (1.60.2)
       json (~> 2.3)
       language_server-protocol (>= 3.17.0)
       parallel (~> 1.10)
-      parser (>= 3.2.2.4)
+      parser (>= 3.3.0.2)
       rainbow (>= 2.2.2, < 4.0)
       regexp_parser (>= 1.8, < 3.0)
       rexml (>= 3.2.5, < 4.0)

data/lib/commonmeta/author_utils.rb

@@ -54,18 +54,22 @@ module Commonmeta
       id = parse_attributes(author.fetch("id", nil), first: true) ||
            parse_attributes(author.fetch("identifier", nil), first: true) ||
            parse_attributes(author.fetch("sameAs", nil), first: true)
-
+      id = normalize_orcid(id) || normalize_ror(id) if id.present?
+      
       # DataCite metadata
       if id.nil? && author["nameIdentifiers"].present?
         id = Array.wrap(author.dig("nameIdentifiers")).find do |ni|
-          ni["nameIdentifierScheme"] == "ORCID"
+          normalize_name_identifier(ni).present?
         end
-        id = id["nameIdentifier"] if id.present?
-        # Crossref metadata
+        id = normalize_name_identifier(id) if id.present?
+      # Crossref metadata
       elsif id.nil? && author["ORCID"].present?
         id = author.fetch("ORCID")
+        id = normalize_orcid(id)
+      # JSON Feed metadata
+      elsif id.nil? && author["url"].present?
+        id = author.fetch("url")       
       end
-      id = normalize_orcid(id) || normalize_ror(id)
 
       # parse author type, i.e. "Person", "Organization" or not specified
       type = author.fetch("type", nil)

data/lib/commonmeta/readers/commonmeta_reader.rb

@@ -12,7 +12,7 @@ module Commonmeta
         read_options = ActiveSupport::HashWithIndifferentAccess.new(options.except(:doi, :id, :url,
                                                                                    :sandbox, :validate, :ra))
         meta = string.present? ? JSON.parse(string) : {}
-        meta["schema_version"] = "https://commonmeta.org/commonmeta_v0.10.5.json"
+        meta["schema_version"] = "https://commonmeta.org/commonmeta_v0.10"
         meta.compact.merge(read_options)
       end
     end

data/lib/commonmeta/readers/datacite_reader.rb

@@ -55,7 +55,13 @@ module Commonmeta
         end
         contributors = get_authors(from_datacite(meta.fetch('creators', nil)))
         contributors += get_authors(from_datacite(meta.fetch('contributors', nil)))
-        publisher = { 'name' => meta.fetch('publisher', nil) }
+        if meta.fetch('publisher', nil).is_a?(Hash)
+          publisher = { 'name' => meta.fetch('publisher', nil).fetch('name', nil) }
+        elsif meta.fetch('publisher', nil).is_a?(String)
+          publisher = { 'name' => meta.fetch('publisher', nil) }
+        else
+          publisher = nil
+        end
 
         container = meta.fetch('container', nil)
         funding_references = meta.fetch('funding_references', nil)

data/lib/commonmeta/schema_utils.rb

@@ -5,7 +5,7 @@ require "pathname"
 
 module Commonmeta
   module SchemaUtils
-    COMMONMETA = File.read(File.expand_path("../../resources/commonmeta_v0.10.5.json",
+    COMMONMETA = File.read(File.expand_path("../../resources/commonmeta_v0.10.6.json",
                                             __dir__))
 
     def json_schema_errors

data/lib/commonmeta/utils.rb

@@ -223,6 +223,7 @@ module Commonmeta
       "Dissertation" => "Dissertation",
       "Event" => "Event",
       "Image" => "Image",
+      "Instrument" => "Instrument",
       "InteractiveResource" => "InteractiveResource",
       "Journal" => "Journal",
       "JournalArticle" => "JournalArticle",
@@ -237,6 +238,7 @@ module Commonmeta
       "Software" => "Software",
       "Sound" => "Sound",
       "Standard" => "Standard",
+      "StudyRegistration" => "StudyRegistration",
       "Text" => "Document",
       "Thesis" => "Dissertation",
       "Workflow" => "Workflow",
@@ -256,6 +258,7 @@ module Commonmeta
       "Event" => "Event",
       "Figure" => "Image",
       "Image" => "Image",
+      "Instrument" => "Instrument",
       "JournalArticle" => "JournalArticle",
       "LegalDocument" => "Text",
       "Manuscript" => "Text",
@@ -271,6 +274,7 @@ module Commonmeta
       "Software" => "Software",
       "Sound" => "Sound",
       "Standard" => "Standard",
+      "StudyRegistration" => "StudyRegistration",
       "WebPage" => "Text",
     }
 
@@ -370,6 +374,7 @@ module Commonmeta
       "CreativeWork" => "Other",
       "Dataset" => "Dataset",
       "Dissertation" => "Dissertation",
+      "Instrument" => "Instrument",
       "NewsArticle" => "Article",
       "Legislation" => "LegalDocument",
       "ScholarlyArticle" => "JournalArticle",
@@ -389,6 +394,7 @@ module Commonmeta
       "Event" => "CreativeWork",
       "Figure" => "CreativeWork",
       "Image" => "CreativeWork",
+      "Instrument" => "Instrument",
       "JournalArticle" => "ScholarlyArticle",
       "LegalDocument" => "Legislation",
       "Software" => "SoftwareSourceCode",
@@ -497,7 +503,9 @@ module Commonmeta
     def find_from_format_by_string(string)
       begin # try to parse as JSON
         hsh = MultiJson.load(string).to_h
-        if hsh.dig("schema_version").to_s.start_with?("https://commonmeta.org")
+        if hsh.dig("blog", "version") == "https://jsonfeed.org/version/1.1"
+          return "json_feed_item"
+        elsif hsh.dig("schema_version").to_s.start_with?("https://commonmeta.org")
           return "commonmeta"
         elsif hsh.dig("@context") && URI.parse(hsh.dig("@context")).host == "schema.org"
           return "schema_org"
@@ -507,7 +515,7 @@ module Commonmeta
           return "crossref"
         elsif hsh.dig("issued", "date-parts").present?
           return "csl"
-        elsif URI.parse(hsh.dig("@context")).to_s == "https://raw.githubusercontent.com/codemeta/codemeta/master/codemeta.jsonld"
+        elsif hsh.dig("@context") && URI.parse(hsh.dig("@context")).to_s == "https://raw.githubusercontent.com/codemeta/codemeta/master/codemeta.jsonld"
           return "codemeta"
         end
       rescue MultiJson::ParseError
@@ -649,6 +657,43 @@ module Commonmeta
       "https://ror.org/" + Addressable::URI.encode(ror)
     end
 
+    def normalize_name_identifier(hsh)
+      return nil unless hsh.present? && hsh.is_a?(Hash)
+
+      name_identifier = hsh["nameIdentifier"]
+      name_identifier_scheme = hsh["nameIdentifierScheme"]
+      scheme_uri = hsh["schemeURI"] || hsh["schemeUri"] 
+      return nil unless name_identifier.present?
+        
+      if name_identifier_scheme == "ORCID" || scheme_uri == "https://orcid.org"
+        return normalize_orcid(name_identifier)
+      elsif name_identifier_scheme == "ROR" || scheme_uri == "https://ror.org"
+        return normalize_ror(name_identifier)
+      elsif name_identifier_scheme == "ISNI" || scheme_uri == "https://isni.org"
+        return normalize_isni(name_identifier)
+      elsif validate_url(name_identifier) == "URL"
+        return name_identifier
+      elsif scheme_uri.present?
+        return scheme_uri + Addressable::URI.encode(name_identifier)
+      end
+
+      return nil
+    end
+
+    def validate_isni(isni)
+      isni = Array(%r{\A(?:(?:http|https)://)?(isni\.org/isni/)?(\d{4}[[:space:]-]\d{4}[[:space:]-]\d{4}[[:space:]-]\d{3}[0-9X]+)\z}.match(isni)).last
+      isni.gsub(/[[:space:]]/, "-") if isni.present?
+    end
+
+    def normalize_isni(isni)
+      # TODO fix validation
+      # isni = validate_isni(isni)
+      return nil unless isni.present?
+
+      # turn ISNI ID into URL
+      "https://isni.org/isni/" + Addressable::URI.encode(isni)
+    end
+
     # pick electronic issn if there are multiple
     # format issn as xxxx-xxxx
     def normalize_issn(input, options = {})

data/lib/commonmeta/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Commonmeta
-  VERSION = '3.9.0'
+  VERSION = '3.11.0'
 end

data/resources/{commonmeta_v0.10.5.json → commonmeta_v0.10.6.json}

@@ -1,7 +1,7 @@
 {
   "$schema": "http://json-schema.org/draft-07/schema#",
-  "$id": "https://commonmeta.org/commonmeta_v0.10.5.json",
-  "title": "Commonmeta v0.10.5",
+  "$id": "https://commonmeta.org/commonmeta_v0.10.6.json",
+  "title": "Commonmeta v0.10.6",
   "description": "JSON representation of the Commonmeta schema.",
   "additionalProperties": false,
   "definitions": {
@@ -125,6 +125,7 @@
         "ReportSeries",
         "Report",
         "Software",
+        "StudyRegistration",
         "Other"
       ]
     },
@@ -508,7 +509,7 @@
       "description": "The schema version of the resource.",
       "type": "string",
       "enum": [
-        "https://commonmeta.org/commonmeta_v0.10.5.json",
+        "https://commonmeta.org/commonmeta_v0.10",
         "http://datacite.org/schema/kernel-3",
         "http://datacite.org/schema/kernel-4"
       ]
@@ -523,7 +524,14 @@
       "type": "array",
       "items": {
         "type": "string",
-        "enum": ["CLOCKSS", "LOCKSS", "Portico", "KB", "Internet Archive", "DWT"]
+        "enum": [
+          "CLOCKSS",
+          "LOCKSS",
+          "Portico",
+          "KB",
+          "Internet Archive",
+          "DWT"
+        ]
       },
       "uniqueItems": true
     }

data/resources/{datacite-v4.json → datacite-v45.json}

@@ -1,8 +1,8 @@
 {
   "$schema": "http://json-schema.org/draft-07/schema#",
-  "id": "datacite-v4.json",
-  "title": "DataCite v4",
-  "description": "JSON representation of the DataCite v4 schema.",
+  "id": "datacite-v45.json",
+  "title": "DataCite v4.5",
+  "description": "JSON representation of the DataCite v4.5 schema.",
   "additionalProperties": true,
   "definitions": {
     "nameType": {
@@ -110,6 +110,7 @@
         "Dissertation",
         "Event",
         "Image",
+        "Instrument",
         "InteractiveResource",
         "Journal",
         "JournalArticle",
@@ -123,6 +124,7 @@
         "Software",
         "Sound",
         "Standard",
+        "StudyRegistration",
         "Text",
         "Workflow",
         "Other"
@@ -187,7 +189,9 @@
         "IsRequiredBy",
         "Requires",
         "IsObsoletedBy",
-        "Obsoletes"
+        "Obsoletes",
+        "Collects",
+        "IsCollectedBy"
       ]
     },
     "descriptionType": {
@@ -276,7 +280,24 @@
       "uniqueItems": true
     },
     "publisher": {
-      "type": "string"
+      "type": ["object", "string"],
+      "anyOf": [
+        {
+          "type": "object",
+          "properties": {
+            "name": { "type": "string" },
+            "publisherIdentifier": { "type": "string", "format": "uri" },
+            "publisherIdentifierScheme": { "type": "string" },
+            "schemeUri": { "type": "string", "format": "uri" },
+            "lang": { "type": "string" }
+          },
+          "required": ["name"]
+        },
+        {
+          "type": "string",
+          "minLength": 1
+        }
+      ]
     },
     "publication_year": {
       "type": "integer"

data/resources/kernel-4/include/datacite-relationType-v4.xsd

@@ -46,6 +46,8 @@
       <xs:enumeration value="IsRequiredBy" />
       <xs:enumeration value="Obsoletes" />
       <xs:enumeration value="IsObsoletedBy" />
+      <xs:enumeration value="Collects" />
+      <xs:enumeration value="IsCollectedBy" />
     </xs:restriction>
   </xs:simpleType>
 </xs:schema>

data/resources/kernel-4/include/datacite-resourceType-v4.xsd

@@ -22,6 +22,7 @@
       <xs:enumeration value="Dissertation" />
       <xs:enumeration value="Event" />
       <xs:enumeration value="Image" />
+      <xs:enumeration value="Instrument" />
       <xs:enumeration value="InteractiveResource" />
       <xs:enumeration value="Journal" />
       <xs:enumeration value="JournalArticle" />
@@ -35,6 +36,7 @@
       <xs:enumeration value="Software" />
       <xs:enumeration value="Sound" />
       <xs:enumeration value="Standard" />
+      <xs:enumeration value="StudyRegistration" />
       <xs:enumeration value="Text" />
       <xs:enumeration value="Workflow" />
       <xs:enumeration value="Other" />

data/resources/kernel-4/metadata.xsd

@@ -2,7 +2,7 @@
 <!-- Revision history
   2010-08-26 Complete revision according to new common specification by the metadata work group after review. AJH, DTIC
   2010-11-17 Revised to current state of kernel review, FZ, TIB
-  2011-01-17 Complete revsion after community review. FZ, TIB
+  2011-01-17 Complete revision after community review. FZ, TIB
   2011-03-17 Release of v2.1: added a namespace; mandatory properties got minLength; changes in the definitions of relationTypes IsDocumentedBy/Documents and isCompiledBy/Compiles; changes type of property "Date" from xs:date to xs:string. FZ, TIB
   2011-06-27 v2.2: namespace: kernel-2.2, additions to controlled lists "resourceType", "contributorType", "relatedIdentifierType", and "descriptionType". Removal of intermediate include-files.
   2013-07-24 v3.0: namespace: kernel-3.0; delete LastMetadataUpdate & MetadateVersionNumber; additions to controlled lists "contributorType", "dateType", "descriptionType", "relationType", "relatedIdentifierType" & "resourceType"; deletion of "StartDate" & "EndDate" from list "dateType" and "Film" from "resourceType";  allow arbitrary order of elements; allow optional wrapper elements to be empty; include xml:lang attribute for title, subject & description; include attribute schemeURI for nameIdentifier of creator, contributor & subject; added new attributes "relatedMetadataScheme", "schemeURI" & "schemeType" to relatedIdentifier; included new property "geoLocation"
@@ -12,7 +12,8 @@
   2018-09-08 v4.1.1 Make schema 4.1 backwards compatible to 4.0 by allowing geolocation elements in any order
   2019-02-14 v4.2: Addition of dateType value "Withdrawn", relationType values "Obsoletes", "isObsoletedBy", addition of new subproperties for Rights: rightsIdentifier, rightsIdentifierScheme, schemeURI, addition of the XML language attribute to the properties Creator, Contributor and Publisher for organizational names, don't check format of DOI
   2019-07-13 v4.3: Addition of new subproperties for Affiliation: "affiliationIdentifier", "affiliationIdentifierScheme", "schemeURI", addition of new sub-property for funderIdentifier: "schemeURI", addition of new funderIdentifierScheme: "ROR", added documentation for nameIdentifier 
-  2021-03-08 v4.4: Addition of new property relatedItem, relationType value "isPublishedIn", subject subproperty "classificationCode", controlled list "numberType", additional 13 properties for controlled list "resourceType" -->
+  2021-03-08 v4.4: Addition of new property relatedItem, relationType value "isPublishedIn", subject subproperty "classificationCode", controlled list "numberType", additional 13 properties for controlled list "resourceType" 
+  2024-01-22 v4.5: Addition of new subproperties for publisher: "publisherIdentifier", "publisherIdentifierScheme", and "schemeURI"; addition of new resourceTypeGeneral values "Instrument" and "StudyRegistration"; addition of new relationType values "Collects" and "IsCollectedBy".-->
 <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://datacite.org/schema/kernel-4" targetNamespace="http://datacite.org/schema/kernel-4" elementFormDefault="qualified" xml:lang="EN">
   <xs:import namespace="http://www.w3.org/XML/1998/namespace" schemaLocation="include/xml.xsd" />
   <xs:include schemaLocation="include/datacite-titleType-v4.xsd" />
@@ -105,6 +106,9 @@
           <xs:complexType>
             <xs:simpleContent>
               <xs:extension base="nonemptycontentStringType">
+                <xs:attribute name="publisherIdentifier" type="xs:string" use="optional" />
+                <xs:attribute name="publisherIdentifierScheme" type="xs:string" use="optional" />
+                <xs:attribute name="schemeURI" type="xs:anyURI" use="optional" />
                 <xs:attribute ref="xml:lang" />
               </xs:extension>
             </xs:simpleContent>
@@ -161,7 +165,7 @@
             <xs:sequence>
               <xs:element name="contributor" minOccurs="0" maxOccurs="unbounded">
                 <xs:annotation>
-                  <xs:documentation>The institution or person responsible for collecting, creating, or otherwise contributing to the developement of the dataset.</xs:documentation>
+                  <xs:documentation>The institution or person responsible for collecting, creating, or otherwise contributing to the development of the dataset.</xs:documentation>
                   <xs:documentation>The personal name format should be: Family, Given.</xs:documentation>
                 </xs:annotation>
                 <xs:complexType>
@@ -524,7 +528,7 @@ Use the complete title of a license and include version information if applicabl
                     </xs:element>
                     <xs:element name="number" minOccurs="0">
                       <xs:annotation>
-                        <xs:documentation>Issue number or name of the related item.</xs:documentation>
+                        <xs:documentation>Number of the related item e.g. report number of article number.</xs:documentation>
                       </xs:annotation>
                       <xs:complexType>
                         <xs:simpleContent>
@@ -641,19 +645,19 @@ Use the complete title of a license and include version information if applicabl
       <xs:pattern value="\d{2}(\d{2}|\?\?|\d(\d|\?))(-(\d{2}|\?\?))?~?\??" />
       <!--  
       The following pattern is for  yearMonthDay - yyyymmdd,  where 'dd' may be '??'  so '200412??' means "some day during the month of 12/2004". 
-      The whole  string may be followed by '?' or '~'  to mean "questionable" or "approximate".     Hypens are  not allowed for this pattern. 
+      The whole  string may be followed by '?' or '~'  to mean "questionable" or "approximate".     Hyphens are  not allowed for this pattern. 
       -->
       <xs:pattern value="\d{6}(\d{2}|\?\?)~?\??" />
       <!--  
 
       The following pattern is for date and time with T separator:'yyyymmddThhmmss'. 
-      Hypens in date and colons in time not allowed for this pattern.   
+      Hyphens in date and colons in time not allowed for this pattern.   
       -->
       <xs:pattern value="\d{8}T\d{6}" />
       <!--  
 
       The following pattern is for a date range. in years: 'yyyy/yyyy'; or year/month: yyyy-mm/yyyy-mm, or year/month/day: yyyy-mm-dd/yyyy-mm-dd. Beginning or end of range value may be 'unknown'. End of range value may be 'open'.
-      Hypens mandatory when month is present. 
+      Hyphens mandatory when month is present. 
       -->
       <xs:pattern value="((-)?(\d{4}(-\d{2})?(-\d{2})?)|unknown)/((-)?(\d{4}(-\d{2})?(-\d{2})?)|unknown|open)" />
     </xs:restriction>

data/spec/fixtures/commonmeta.json

@@ -406,6 +406,6 @@
   ],
   "subjects": [],
   "provider": "Crossref",
-  "schema_version": "https://commonmeta.org/commonmeta_v0.10.5.json",
+  "schema_version": "https://commonmeta.org/commonmeta_v0.10",
   "state": "findable"
 }