Updatd DOI generation to generate both validated JSON and XML versions to better support syncing with DataCite and new features.

Author: bherr2

requirements-freeze.txt

@@ -2,143 +2,146 @@ airium==0.2.6
 annotated-types==0.7.0
 antlr4-python3-runtime==4.9.3
 appdirs==1.4.4
-argcomplete==3.6.2
+argcomplete==3.3.0
 arrow==1.3.0
-attrs==25.3.0
+attrs==23.2.0
 bcp47==0.1.0
-CacheControl==0.14.2
-cattrs==24.1.3
-certifi==2025.1.31
+CacheControl==0.14.0
+cattrs==24.1.0
+certifi==2024.6.2
 CFGraph==0.2.1
 chardet==5.2.0
-charset-normalizer==3.4.1
-check-jsonschema==0.33.0
-class_resolver==0.5.4
-click==8.1.8
+charset-normalizer==3.3.2
+check-jsonschema==0.28.4
+class_resolver==0.5.0
+click==8.1.7
 coloredlogs==15.0.1
-curies==0.10.11
-cwl-upgrader==1.2.12
-cwl-utils==0.37
-cwltool==3.1.20250110105449
+curies==0.7.9
+cwl-upgrader==1.2.11
+cwl-utils==0.33
+cwltool==3.1.20240508115724
+datacite==1.3.1
 defusedxml==0.7.1
-Deprecated==1.2.18
+Deprecated==1.2.14
 deprecation==2.1.0
-et_xmlfile==2.0.0
+et-xmlfile==1.1.0
 eutils==0.6.0
-fastobo==0.13.0
-filelock==3.18.0
+exceptiongroup==1.2.1
+fastobo==0.12.3
+filelock==3.14.0
 fqdn==1.5.1
 funowl==0.2.3
 graphviz==0.20.3
+greenlet==3.0.3
 hbreader==0.9.1
 humanfriendly==10.0
-idna==3.10
+idna==3.7
+idutils==1.5.0
 ijson==3.3.0
-importlib_resources==6.5.2
-iniconfig==2.1.0
-isodate==0.7.2
+importlib_metadata==8.7.0
+importlib_resources==6.4.5
+iniconfig==2.0.0
+isbnlib==3.10.14
+isodate==0.6.1
 isoduration==20.11.0
-Jinja2==3.1.6
+Jinja2==3.1.4
 json-flattener==0.1.9
 jsonasobj==1.3.1
 jsonasobj2==1.0.4
 jsonlines==4.0.0
 jsonpatch==1.33
-jsonpath-ng==1.7.0
-jsonpointer==3.0.0
-jsonschema==4.23.0
-jsonschema-specifications==2024.10.1
+jsonpath-ng==1.6.1
+jsonpointer==2.4
+jsonschema==4.22.0
+jsonschema-specifications==2023.12.1
 kgcl-rdflib==0.5.0
 kgcl_schema==0.6.9
 lark==1.2.2
-linkml==1.8.7
+linkml==1.7.10
 linkml-dataops==0.1.0
 linkml-renderer==0.3.1
-linkml-runtime==1.9.1
+linkml-runtime==1.7.6
 linkml_owl==0.4.2
-lxml==5.3.2
-markdown-it-py==3.0.0
-MarkupSafe==3.0.2
-mdurl==0.1.2
+lxml==5.2.2
+MarkupSafe==2.1.5
 mistune==3.0.2
 more-click==0.1.2
-msgpack==1.1.0
+msgpack==1.0.8
 mypy-extensions==1.0.0
-ndex2==3.10.0
-networkx==3.4.2
+ndex2==3.9.0
+networkx==3.3
 nodeenv==1.9.1
-numpy==2.2.4
-oaklib==0.6.22
+numpy==2.1.1
+oaklib==0.6.16
 ols-client==0.1.4
 ontoportal-client==0.0.4
-openpyxl==3.1.5
-packaging==24.2
-pandas==2.2.3
+openpyxl==3.1.3
+packaging==24.0
+pandas==2.2.2
 pansql==0.0.1
-parse==1.20.2
-platformdirs==4.3.7
+parse==1.20.1
+platformdirs==4.3.2
 pluggy==1.5.0
 ply==3.11
 prefixcommons==0.1.12
-prefixmaps==0.2.6
-pronto==2.7.0
+prefixmaps==0.2.4
+pronto==2.5.7
 prov==1.5.1
-psutil==7.0.0
-pydantic==2.11.3
-pydantic_core==2.33.1
+psutil==6.0.0
+pydantic==2.7.4
+pydantic_core==2.18.4
 pydot==2.0.0
-Pygments==2.19.1
 PyJSG==0.11.10
-pyparsing==3.2.3
+pyparsing==3.1.2
 PyShEx==0.8.1
 PyShExC==0.9.1
-pysolr==3.10.0
-pystow==0.7.0
-pytest==8.3.5
+pysolr==3.9.0
+pystow==0.5.5
+pytest==8.2.2
 pytest-logging==2015.11.4
 python-dateutil==2.9.0.post0
 PyTrie==0.4.0
-pytz==2025.2
-PyYAML==6.0.2
+pytz==2024.1
+PyYAML==6.0.1
 ratelimit==2.2.1
-rdflib==7.1.4
+rdflib==7.0.0
 rdflib-jsonld==0.6.1
 rdflib-shim==1.0.3
-referencing==0.36.2
-regress==2025.3.1
+referencing==0.35.1
+regress==0.4.5
 requests==2.32.3
 requests-cache==1.2.1
 requests-toolbelt==1.0.0
 rfc3339-validator==0.1.4
 rfc3987==1.3.8
-rich==14.0.0
-rich-argparse==1.7.0
-rpds-py==0.24.0
-ruamel.yaml==0.18.10
-schema-salad==8.9.20250408123006
-scipy==1.15.2
-semsql==0.4.0
-setuptools==78.1.0
+rpds-py==0.18.1
+ruamel.yaml==0.18.6
+ruamel.yaml.clib==0.2.8
+schema-salad==8.5.20240503091721
+scipy==1.14.1
+semsql==0.3.3
+shellescape==3.8.1
 ShExJSG==0.8.2
-six==1.17.0
+six==1.16.0
 sortedcontainers==2.4.0
 sparqlslurper==0.5.1
 SPARQLWrapper==2.0.0
-spython==0.3.14
-SQLAlchemy==2.0.40
+spython==0.3.13
+SQLAlchemy==2.0.30
 SQLAlchemy-Utils==0.38.3
-sssom==0.4.15
-sssom-schema==1.0.0
-tenacity==9.1.2
-tqdm==4.67.1
-types-python-dateutil==2.9.0.20241206
-typing-inspection==0.4.0
-typing_extensions==4.13.2
-tzdata==2025.2
+sssom==0.4.11
+sssom-schema==0.15.2
+tenacity==9.0.0
+tomli==2.0.1
+tqdm==4.66.5
+types-python-dateutil==2.9.0.20240316
+typing_extensions==4.12.1
+tzdata==2024.1
 uri-template==1.3.0
-url-normalize==2.2.0
-urllib3==2.4.0
+url-normalize==1.4.3
+urllib3==2.2.1
 validators==0.34.0
-watchdog==6.0.0
-webcolors==24.11.1
-wrapt==1.17.2
+watchdog==4.0.1
+webcolors==24.6.0
+wrapt==1.16.0
+zipp==3.23.0

requirements.txt

@@ -4,3 +4,4 @@ check-jsonschema
 linkml-owl
 linkml # use version required by linkml-owl above
 semsql
+datacite

src/finalizing/misc-files.js

@@ -2,29 +2,26 @@ import { resolve } from 'path';
 import { listDeployed } from '../list.js';
 import { readMetadata, readNormalizedMetadata } from '../normalization/utils.js';
 import { getDigitalObjectInformation } from '../utils/digital-object.js';
-import { writeDoiXml } from './utils/generate-doi-xml.js';
+import { writeDoiFiles } from './utils/generate-doi-files.js';
 import { writeIndexHtml } from './utils/generate-index-html.js';
 import { writeReadmeMd } from './utils/generate-readme-md.js';
 
 export function miscDoFiles(context) {
-  const metadata = readMetadata(context);
-  writeDoiXml(context, metadata);
-
+  deployDoiXml(context);
   const normalizedMetadata = readNormalizedMetadata(context);
   writeReadmeMd(context, normalizedMetadata);
   writeIndexHtml(context, normalizedMetadata);
 }
 
 export function miscFiles(context) {
-  for (const dataset of listDeployed(context).filter(doName => !doName.endsWith('/latest'))) {
+  for (const dataset of listDeployed(context).filter((doName) => !doName.endsWith('/latest'))) {
     const doPath = resolve(context.doHome, dataset);
     const selectedDigitalObject = getDigitalObjectInformation(doPath, context.purlIri);
     miscDoFiles({ ...context, selectedDigitalObject });
   }
 }
 
-
 export function deployDoiXml(context) {
   const metadata = readMetadata(context);
-  writeDoiXml(context, metadata);
+  writeDoiFiles(context, metadata);
 }

src/finalizing/templates/doi-xml.njk

@@ -0,0 +1,18 @@
+import json
+import sys
+
+from datacite import schema45
+
+if __name__ == "__main__":
+    input = sys.argv[1]
+    output = sys.argv[2]
+
+    json_data = json.load(open(input, "r", encoding="utf8"))
+    data = json_data["data"]["attributes"]
+
+    schema45.validator.validate(data)
+
+    doc = schema45.tostring(data)
+
+    with open(output, "w") as out:
+        out.write(doc)