Update to new process of building metadata and zenodo release

.github/scripts/generate_metadata_files.py

@@ -0,0 +1,106 @@
+"""Generate CITATION.cff and bioschemas.yml from metadata.yml."""
+
+import json
+
+import yaml
+
+METADATA_FILE = "metadata.yml"
+CITATION_FILE = "CITATION.cff"
+BIOSCHEMAS_FILE = "bioschemas.yml"
+
+TITLE_SUFFIX = " - CodeRefinery lesson"
+
+SPDX_LICENSE_URLS = {
+    "CC-BY-4.0": "https://creativecommons.org/licenses/by/4.0/",
+}
+
+
+def load_yaml(path):
+    with open(path, encoding="utf-8") as f:
+        return yaml.safe_load(f)
+
+
+def license_url(spdx_id):
+    return SPDX_LICENSE_URLS.get(spdx_id, spdx_id)
+
+
+def bioschemas_authors(authors):
+    result = []
+    for author in authors:
+        if "family-names" in author:
+            result.append(
+                {
+                    "@type": "Person",
+                    "name": f"{author['given-names']} {author['family-names']}",
+                }
+            )
+        else:
+            result.append({"@type": "Organization", "name": author["name"]})
+    return result
+
+
+def write_citation_cff(meta):
+    citation = {
+        "cff-version": "1.2.0",
+        "message": (
+            "If you use this lesson material, please cite it using "
+            "these metadata."
+        ),
+        "authors": meta["authors"],
+        "title": meta["title"] + TITLE_SUFFIX,
+        "type": "dataset",
+        "abstract": meta.get("abstract", ""),
+        "version": meta["version"],
+        "doi": meta["doi"],
+        "date-released": meta["version"],
+        "url": meta["url"],
+        "license": meta["license"],
+        "repository-code": meta["repository-code"],
+    }
+
+    if meta.get("maintainers"):
+        citation["contact"] = meta["maintainers"]
+
+    with open(CITATION_FILE, "w", encoding="utf-8") as f:
+        yaml.safe_dump(citation, f, sort_keys=False, allow_unicode=True)
+
+
+def write_bioschemas(meta):
+    bioschemas = {
+        "@context": "https://schema.org/",
+        "@type": "LearningResource",
+        "@id": meta["url"],
+        "description": meta.get("abstract", ""),
+        "keywords": ", ".join(meta.get("keywords", [])),
+        "name": meta["title"] + TITLE_SUFFIX,
+        "author": bioschemas_authors(meta["authors"]),
+        # Disabled for now.
+        # "maintainer": bioschemas_authors(meta.get("maintainers", [])),
+        "about": meta.get("abstract", ""),
+        "audience": meta.get("audience", ""),
+        "competencyRequired": meta.get("competencyRequired", ""),
+        "educationalLevel": meta.get("educationalLevel", ""),
+        "identifier": f"https://doi.org/{meta['doi']}",
+        "inLanguage": meta.get("inLanguage", ""),
+        "learningResourceType": meta.get("learningResourceType", ""),
+        "license": license_url(meta["license"]),
+        "teaches": meta.get("teaches", ""),
+        "url": meta["url"],
+        "accessibilitySummary": meta.get("accessibilitySummary", ""),
+        "isPartOf": meta.get("isPartOf", ""),
+        "version": meta["version"],
+    }
+
+    with open(BIOSCHEMAS_FILE, "w", encoding="utf-8") as f:
+        json.dump(bioschemas, f, indent=4)
+        f.write("\n")
+
+
+def main():
+    meta = load_yaml(METADATA_FILE)
+    write_citation_cff(meta)
+    write_bioschemas(meta)
+
+
+if __name__ == "__main__":
+    main()

.github/scripts/publish_to_zenodo.py

@@ -13,6 +13,8 @@
 
 BASE_URL = "https://zenodo.org/api"
 
+TITLE_SUFFIX = " - CodeRefinery lesson"
+
 
 json_headers = {
     "Authorization": f"Bearer {ACCESS_TOKEN}",
@@ -23,47 +25,50 @@
     "Authorization": f"Bearer {ACCESS_TOKEN}"
 }
 
-def load_citation():
-    with open("CITATION.cff", encoding="utf-8") as f:
+def load_metadata():
+    with open("metadata.yml", encoding="utf-8") as f:
         return yaml.safe_load(f)
 
 
-def cff_to_creators(authors):
+def authors_to_creators(authors):
     creators = []
 
     for author in authors:
         if "family-names" in author:
-            creators.append(
-                {
-                    "name": (
-                        f"{author['family-names']}, "
-                        f"{author['given-names']}"
-                    )
-                }
-            )
+            creator = {
+                "name": (
+                    f"{author['family-names']}, "
+                    f"{author['given-names']}"
+                )
+            }
         else:
-            creators.append({"name": author["name"]})
+            creator = {"name": author["name"]}
+
+        if author.get("orcid"):
+            creator["orcid"] = author["orcid"].removeprefix("https://orcid.org/")
+
+        creators.append(creator)
 
     return creators
 
 
 
-# Load metadata from CITATION.cff
+# Load metadata from metadata.yml
 
-cff = load_citation()
+meta = load_metadata()
 
-description = cff.get("abstract", "")
+description = meta.get("abstract", "")
 
-if cff.get("repository-code"):
+if meta.get("repository-code"):
     description += (
         f'<br><p>Source code: '
-        f'<a href="{cff["repository-code"]}">{cff["repository-code"]}</a></p>'
+        f'<a href="{meta["repository-code"]}">{meta["repository-code"]}</a></p>'
     )
 
-if cff.get("url"):
+if meta.get("url"):
     description += (
         f'<br><p>Lesson website: '
-        f'<a href="{cff["url"]}">{cff["url"]}</a></p>'
+        f'<a href="{meta["url"]}">{meta["url"]}</a></p>'
     )
 
 description += (
@@ -72,15 +77,21 @@ def cff_to_creators(authors):
 )
 
 metadata = {
-    "title": cff["title"],
-    "upload_type": "dataset",
+    "title": meta["title"] + TITLE_SUFFIX,
+    "upload_type": "lesson",
     "description": description,
-    "creators": cff_to_creators(cff["authors"]),
-    "keywords": cff.get("keywords", []),
-    "version": str(cff.get("version")) if cff.get("version") is not None else None,
-    "license": cff.get("license"),
+    "creators": authors_to_creators(meta["authors"]),
+    "keywords": meta.get("keywords", []),
+    "version": meta.get("version"),
+    "license": meta.get("license"),
 }
 
+if meta.get("maintainers"):
+    contributors = authors_to_creators(meta["maintainers"])
+    for contributor in contributors:
+        contributor["type"] = "ContactPerson"
+    metadata["contributors"] = contributors
+
 
 # Discard any unpublished draft left over from a previous failed run.
 # Zenodo only allows one unpublished new-version draft per concept at
@@ -161,45 +172,14 @@ def cff_to_creators(authors):
 
     print(f"Removed inherited file {f['filename']}")
 
-
-# Upload release archive
-
+# Upload lesson PDF from the gh-pages branch
+# (built by sphinx.yml as OUTPUT_BASENAME.pdf, OUTPUT_BASENAME being
+# "{owner}-{repo}")
 
 tag = os.environ["GITHUB_REF_NAME"]
 repo_name = os.environ['GITHUB_REPOSITORY']
 owner, repo = repo_name.split("/", 1)
 
-
-archive_url = (
-    f"https://github.com/"
-    f"{repo_name}"
-    f"/archive/refs/tags/{tag}.zip"
-)
-
-archive_name = f"{owner}-{repo}-{tag}.zip"
-
-download = requests.get(archive_url)
-download.raise_for_status()
-
-with open(archive_name, "wb") as fp:
-    fp.write(download.content)
-
-with open(archive_name, "rb") as fp:
-    r = requests.put(
-        f"{bucket_url}/{archive_name}",
-        data=fp,
-        headers=upload_headers
-    )
-
-r.raise_for_status()
-
-print(f"Uploaded {archive_name}")
-
-
-# Upload lesson PDF from the gh-pages branch
-# (built by sphinx.yml as OUTPUT_BASENAME.pdf, OUTPUT_BASENAME being
-# "{owner}-{repo}")
-
 pdf_source_name = f"{owner}-{repo}.pdf"
 pdf_name = f"{owner}-{repo}-{tag}.pdf"
 
@@ -225,29 +205,33 @@ def cff_to_creators(authors):
 
 print(f"Uploaded {pdf_name}")
 
+# Upload release archive
 
-# Show the PDF first (Zenodo previews the first file in the list).
 
-r = requests.get(
-    f"{BASE_URL}/deposit/depositions/{deposition_id}/files",
-    headers=json_headers,
+archive_url = (
+    f"https://github.com/"
+    f"{repo_name}"
+    f"/archive/refs/tags/{tag}.zip"
 )
 
-r.raise_for_status()
+archive_name = f"{owner}-{repo}-{tag}.zip"
 
-files = r.json()
+download = requests.get(archive_url)
+download.raise_for_status()
 
-sort_order = sorted(files, key=lambda f: f["filename"] != pdf_name)
+with open(archive_name, "wb") as fp:
+    fp.write(download.content)
 
-r = requests.put(
-    f"{BASE_URL}/deposit/depositions/{deposition_id}/files/sort",
-    data=json.dumps([{"id": f["id"]} for f in sort_order]),
-    headers=json_headers,
-)
+with open(archive_name, "rb") as fp:
+    r = requests.put(
+        f"{bucket_url}/{archive_name}",
+        data=fp,
+        headers=upload_headers
+    )
 
 r.raise_for_status()
 
-print(f"File order: {[f['filename'] for f in sort_order]}")
+print(f"Uploaded {archive_name}")
 
 
 # Update metadata

.github/workflows/generate-metadata.yml

@@ -0,0 +1,32 @@
+name: Generate metadata files
+
+on:
+  push:
+    branches: [main]
+    paths:
+      - metadata.yml
+
+jobs:
+  generate:
+    name: Generate CITATION.cff and bioschemas.yml
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Install dependencies
+        run: pip install pyyaml
+
+      - name: Generate CITATION.cff and bioschemas.yml
+        run: python .github/scripts/generate_metadata_files.py
+
+      - name: Commit generated files
+        run: |
+          git config user.name "github-actions[bot]"
+          git config user.email "github-actions[bot]@users.noreply.github.com"
+          git add CITATION.cff bioschemas.yml
+          git diff --staged --quiet || git commit -m "Regenerate CITATION.cff and bioschemas.yml from metadata.yml"
+          git push

.gitignore

@@ -3,3 +3,4 @@ _build/
 venv/
 .*.sw?
 *DS_Store
+*__pycache__

content/conf.py

@@ -9,9 +9,9 @@
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
 #
-# import os
-# import sys
-# sys.path.insert(0, os.path.abspath('.'))
+import os
+import sys
+sys.path.insert(0, os.path.abspath('.'))
 
 
 # -- Project information -----------------------------------------------------
@@ -40,6 +40,7 @@
     "sphinx_rtd_theme_ext_color_contrast",
     "sphinx_coderefinery_branding",
     "sphinx_bioschemas",
+    "lesson_metadata",
 ]
 
 # Settings for myst_nb: