feat: update datacite model to version 4.6

I am not sure what the best way to deal with schema versions is. The
easiest thing and the current strategy is to provide the latest versions
that are backward-compatible with prior versions (within reason) i.e.:

* `CodeMetaV3` can accept v2 fields and turn them into v3 equivalents
* `DataCiteV46` implements the latest changes to v4, and does not break
   compatibility with any other v4 version

Citation File Format provides jsonschema so in this case, it would be
easy to provide seperate models

Author: sgfost

README.md

@@ -11,35 +11,31 @@ For converting between different standards, an extension of [CodeMeta](https://c
 
 <table><thead>
   <tr>
-    <th colspan="2">Schema</th>
-    <th colspan="3">Status<br></th>
+    <th>Schema</th>
+    <th>Pydantic model</th>
+    <th>Backward-compatible with<a href="#1"><sup>[1]</sup></a></th>
+    <th>Convert <i>to</i></th>
+    <th>Convert <i>from</i></th>
   </tr></thead>
 <tbody>
   <tr>
-    <td>Name<br></td>
-    <td>Version(s)</td>
-    <td>Pydantic Model</td>
-    <td>convert <b>to</b><br></td>
-    <td>convert <b>from</b><br></td>
-  </tr>
-  <tr>
-    <td><a href="https://codemeta.github.io/">CodeMeta</a><br></td>
-    <td><a href="https://w3id.org/codemeta/3.0"><code>v3</code></a></td>
-    <td>✅ *</td>
+    <td><a href="https://w3id.org/codemeta/3.0">CodeMeta v3</a></td>
+    <td>✅<a href="#2"><sup>[2]</sup></a></td>
+    <td>v2</td>
     <td>✅</td>
     <td>✅</td>
   </tr>
   <tr>
-    <td><a href="https://schema.datacite.org/">Datacite</a></td>
-    <td><a href="https://datacite-metadata-schema.readthedocs.io/en/4.5/"><code>v4.5</code></a><br></td>
+    <td><a href="https://datacite-metadata-schema.readthedocs.io/en/4.6">Datacite 4.6</a></td>
     <td>✅</td>
+    <td>4.0, 4.1, 4.2, 4.3, 4.4, 4.5</td>
     <td>✅</td>
     <td></td>
   </tr>
   <tr>
-    <td><a href="https://citation-file-format.github.io/">Citation File Format</a></td>
-    <td><a href="https://github.com/citation-file-format/citation-file-format/blob/bd0b31df69dccf11b31584585b5fb8c39d3e0e09/schema.json"><code>1.2.0</a></code></td>
+    <td><a href="https://citation-file-format.github.io/">Citation File Format 1.2.0</a></td>
     <td>✅</td>
+    <td></td>
     <td>✅</td>
     <td></td>
   </tr>
@@ -67,7 +63,11 @@ For converting between different standards, an extension of [CodeMeta](https://c
 </tbody>
 </table>
 
-\* The `CodeMeta` model is currently implemented as a pydantic **v1** model, due to a heavy reliance on [pydantic_schemaorg](https://github.com/lexiq-legal/pydantic_schemaorg) which has not been fully updated.
+##### [1]
+Lists the versions that can be safely used as input. Output will always use the specified version. For example, the `CodeMetaV3` model will accept v2 property names and automatically change them to v3 equivalents.
+
+##### [2]
+The `CodeMeta` model is currently implemented as a pydantic **v1** model, due to a heavy reliance on [pydantic_schemaorg](https://github.com/lexiq-legal/pydantic_schemaorg) which has not been fully updated.
 
 ## Installation
 

codemeticulous/cff/models.py

@@ -1575,7 +1575,7 @@ class Reference(BaseModel):
     """
 
 
-class CitationFileFormat(ByAliasExcludeNoneMixin, BaseModel):
+class CitationFileFormatV120(ByAliasExcludeNoneMixin, BaseModel):
     model_config = ConfigDict(
         extra="forbid",
         populate_by_name=True,
@@ -1694,3 +1694,6 @@ class CitationFileFormat(ByAliasExcludeNoneMixin, BaseModel):
     """
     The version of the software or dataset.
     """
+
+
+CitationFileFormat = CitationFileFormatV120

codemeticulous/codemeta/models.py

@@ -40,7 +40,7 @@ class VersionedLanguage(ComputerLanguage):
 SoftwareListOrSingle = Software | SoftwareList
 
 
-class CodeMeta(ByAliasExcludeNoneMixin, BaseModel):
+class CodeMetaV3(ByAliasExcludeNoneMixin, BaseModel):
     """CodeMeta v3 schema (supports v2 fields aliased to v3)
     see: https://codemeta.github.io/terms/
     and: https://github.com/codemeta/codemeta-generator/blob/master/js/validation/
@@ -272,3 +272,6 @@ def validate_sub_type(cls, value, base_class):
 
     class Config:
         allow_population_by_field_name = True
+
+
+CodeMeta = CodeMetaV3

codemeticulous/convert.py

@@ -1,5 +1,5 @@
 from codemeticulous.codemeta.models import CodeMeta
-from codemeticulous.datacite.models import DataciteV45
+from codemeticulous.datacite.models import DataCite
 from codemeticulous.cff.models import CitationFileFormat
 from codemeticulous.codemeta.convert import canonical_to_codemeta, codemeta_to_canonical
 from codemeticulous.datacite.convert import canonical_to_datacite, datacite_to_canonical
@@ -14,7 +14,7 @@
         "from_canonical": canonical_to_codemeta,
     },
     "datacite": {
-        "model": DataciteV45,
+        "model": DataCite,
         "format": "json",
         "to_canonical": datacite_to_canonical,
         "from_canonical": canonical_to_datacite,

codemeticulous/datacite/convert.py

@@ -26,7 +26,7 @@
     Contributor,
     ContributorType,
     Creator,
-    DataciteV45,
+    DataCite,
     DateModel,
     Description,
     NameIdentifier,
@@ -154,6 +154,7 @@
         "financial backer",
     },
     ContributorType.Supervisor: {"supervisor", "advisor", "overseer", "mentor"},
+    ContributorType.Translator: {"translator", "translation"},
     ContributorType.WorkPackageLeader: {
         "work package leader",
         "package leader",
@@ -222,9 +223,17 @@ def codemeta_actors_to_datacite(
             # match roles to contributor types
             matched_roles = set()
             for role in extractor.role_names:
-                normalized_role = role.lower().replace(" ", "").replace("-", "").replace("_", "")
-                for contributor_type, synonyms, in CONTRIBUTOR_TYPE_MAP.items():
-                    if normalized_role in {s.lower().replace(" ", "").replace("-", "").replace("_", "") for s in synonyms}:
+                normalized_role = (
+                    role.lower().replace(" ", "").replace("-", "").replace("_", "")
+                )
+                for (
+                    contributor_type,
+                    synonyms,
+                ) in CONTRIBUTOR_TYPE_MAP.items():
+                    if normalized_role in {
+                        s.lower().replace(" ", "").replace("-", "").replace("_", "")
+                        for s in synonyms
+                    }:
                         matched_roles.add(contributor_type)
             # if we have no matched roles, default to "Other"
             if not matched_roles:
@@ -235,13 +244,16 @@ def codemeta_actors_to_datacite(
                     Contributor(
                         name=extractor.name,
                         nameType=(
-                            "Organizational" if extractor.is_organization else "Personal"
+                            "Organizational"
+                            if extractor.is_organization
+                            else "Personal"
                         ),
                         givenName=extractor.given_names,
                         familyName=extractor.family_names,
                         nameIdentifiers=[NameIdentifier(**i) for i in name_identifiers]
                         or None,
-                        affiliation=[AffiliationItem(**a) for a in affiliations] or None,
+                        affiliation=[AffiliationItem(**a) for a in affiliations]
+                        or None,
                         contributorType=role_type.value,
                     )
                 )
@@ -307,7 +319,7 @@ def codemeta_language_fileformat_to_datacite_format(
 
 def canonical_to_datacite(
     data: CanonicalCodeMeta, ignore_existing_doi=False, **custom_fields
-) -> DataciteV45:
+) -> DataCite:
     primary_doi = (
         extract_doi_from_identifier(data.identifier)
         if not ignore_existing_doi
@@ -328,7 +340,7 @@ def canonical_to_datacite(
                 for note in release_notes
             ]
         )
-    return DataciteV45(
+    return DataCite(
         doi=primary_doi,
         prefix=doi_prefix,
         suffix=doi_suffix,
@@ -372,7 +384,7 @@ def canonical_to_datacite(
     )
 
 
-def datacite_to_canonical(data: DataciteV45) -> CanonicalCodeMeta:
+def datacite_to_canonical(data: DataCite) -> CanonicalCodeMeta:
     raise NotImplementedError(
         "DataCite metadata is not yet supported as an input format"
     )

codemeticulous/datacite/models.py

@@ -1,5 +1,5 @@
 """
-generated by datamodel-codegen (schema/datacite/schema45.json)
+generated by datamodel-codegen (schema/datacite/schema46.json)
 with options:
     --output-model-type pydantic_v2.BaseModel \
     --field-constraints \
@@ -11,9 +11,9 @@
     --disable-timestamp
 
 MANUAL CHANGES:
-- add ByAliasExcludeNoneMixin to DataciteV45
+- add ByAliasExcludeNoneMixin to DataCiteV46
 - add identifier and identifierType to Container, this is included in examples
-  for schema 4.5 but not in the schema itself..
+  but not in the source jsonschema
 """
 
 from __future__ import annotations
@@ -163,6 +163,7 @@ class ContributorType(Enum):
     RightsHolder = "RightsHolder"
     Sponsor = "Sponsor"
     Supervisor = "Supervisor"
+    Translator = "Translator"
     WorkPackageLeader = "WorkPackageLeader"
     Other = "Other"
 
@@ -179,6 +180,7 @@ class DateType(Enum):
     Available = "Available"
     Copyrighted = "Copyrighted"
     Collected = "Collected"
+    Coverage = "Coverage"
     Created = "Created"
     Issued = "Issued"
     Submitted = "Submitted"
@@ -190,6 +192,7 @@ class DateType(Enum):
 
 class ResourceTypeGeneral(Enum):
     Audiovisual = "Audiovisual"
+    Award = "Award"
     Book = "Book"
     BookChapter = "BookChapter"
     Collection = "Collection"
@@ -210,6 +213,7 @@ class ResourceTypeGeneral(Enum):
     PeerReview = "PeerReview"
     PhysicalObject = "PhysicalObject"
     Preprint = "Preprint"
+    Project = "Project"
     Report = "Report"
     Service = "Service"
     Software = "Software"
@@ -225,6 +229,7 @@ class RelatedIdentifierType(Enum):
     ARK = "ARK"
     arXiv = "arXiv"
     bibcode = "bibcode"
+    CTSR = "CTSR"
     DOI = "DOI"
     EAN13 = "EAN13"
     EISSN = "EISSN"
@@ -237,6 +242,7 @@ class RelatedIdentifierType(Enum):
     LSID = "LSID"
     PMID = "PMID"
     PURL = "PURL"
+    RRID = "RRID"
     UPC = "UPC"
     URL = "URL"
     URN = "URN"
@@ -280,6 +286,8 @@ class RelationType(Enum):
     Requires = "Requires"
     IsObsoletedBy = "IsObsoletedBy"
     Obsoletes = "Obsoletes"
+    isTranslationOf = "isTranslationOf"
+    hasTranslation = "hasTranslation"
 
 
 class RelatedObject(BaseModel):
@@ -469,7 +477,7 @@ class RelatedItem(RelatedObject):
     relationType: RelationType
 
 
-class DataciteV45(ByAliasExcludeNoneMixin, BaseModel):
+class DataciteV46(ByAliasExcludeNoneMixin, BaseModel):
     model_config = ConfigDict(
         extra="forbid",
         populate_by_name=True,
@@ -502,3 +510,6 @@ class DataciteV45(ByAliasExcludeNoneMixin, BaseModel):
         "http://datacite.org/schema/kernel-4"
     )
     container: Optional[Container] = None
+
+
+DataCite = DataciteV46