Publish links

.github/workflows/tests.yml

@@ -22,7 +22,7 @@ jobs:
     steps:
       - uses: actions/checkout@v2
       - name: Set up Python ${{ matrix.python-version }}
-        uses: actions/setup-python@v2
+        uses: actions/setup-python@v4
         with:
           python-version: ${{ matrix.python-version }}
           cache : 'pip'

examples/publishing-guides/dataset_publishing.ipynb

@@ -291,7 +291,15 @@
     "\n",
     "# publication year (optional)\n",
     "# The default is the current calendar year\n",
-    "publication_year = 2023"
+    "publication_year = 2023\n",
+    "\n",
+    "# links (optional)\n",
+    "# These are urls that link to additional information about the dataset\n",
+    "# in the form [{\"type\":str, \"doi\":str, \"url\":str, \"description\":str, \"bibtex\":str}]\n",
+    "links = {\"type\": \"other_paper\",\n",
+    "        \"doi\": \"10.1072/xyz\",\n",
+    "        \"url\": \"https://doi.org/10.1072/xyz\"\n",
+    "          }"
    ]
   },
   {
@@ -323,19 +331,19 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 4,
    "metadata": {
     "id": "iSxfIFmxdBjS"
    },
    "outputs": [],
    "source": [
     "# publish to Foundry! returns a result object we can inspect\n",
-    "res = f.publish_dataset(example_iris_metadata, title, authors, https_data_path=data_path, short_name=short_name)"
+    "res = f.publish_dataset(example_iris_metadata, title, authors, https_data_path=data_path, short_name=short_name, links=links)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 5,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -350,7 +358,7 @@
        "True"
       ]
      },
-     "execution_count": 15,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -371,7 +379,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 6,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -383,13 +391,13 @@
     {
      "data": {
       "text/plain": [
-       "{'source_id': 'colab_example_iris_1682542160_v1.1',\n",
+       "{'source_id': 'colab_example_iris_1691186523_v1.1',\n",
        " 'success': True,\n",
        " 'error': None,\n",
        " 'status_code': 202}"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -680,9 +688,9 @@
    "provenance": []
   },
   "kernelspec": {
-   "display_name": "https_pub_example",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
-   "name": "https_pub_example"
+   "name": "python3"
   },
   "language_info": {
    "codemirror_mode": {
@@ -694,7 +702,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.16"
+   "version": "3.9.4"
   }
  },
  "nbformat": 4,

foundry/foundry.py

@@ -395,6 +395,8 @@ def publish_dataset(
             dataset_doi (str): The DOI for this dataset (not an associated paper).
             related_dois (list): DOIs related to this dataset,
                     not including the dataset's own DOI (for example, an associated paper's DOI).
+            links (list[dict]): List of dicts describing links associated with the dataset; in the format:
+                {"type":str, "doi":str, "url":str, "description":str, "bibtex":str}
 
         Returns
         -------
@@ -425,6 +427,17 @@ def publish_dataset(
         self.connect_client.set_project_block(
             self.config.metadata_key, foundry_metadata)
 
+        # add links
+        links = kwargs.get("links")
+        if links is not None:
+            # make sure it's a list
+            links = links if isinstance(links, list) else [links]
+            for link in links:
+                # validate links
+                self.validate_link(link)
+            # add valid links - need to add all at once
+            self.connect_client.add_links(links)
+
         # upload via HTTPS if specified
         if https_data_path:
             # gather auth'd clients necessary for publication to endpoint
@@ -808,6 +821,15 @@ def validate_metadata(self, metadata):
                 field_name = ".".join([item for item in error['loc'] if isinstance(item, str)])
                 error_description = error['msg']
                 error_message = f"""There is an issue validating the metadata for the field '{field_name}':
-                The error message returned is: '{error_description}'."""
+                                The error message returned is: '{error_description}'."""
                 logger.error(error_message)
             raise e
+
+    def validate_link(self, link):
+        valid_keys = ["type", "doi", "url", "description", "bibtex"]
+        link_keys = [*link]
+        if not all(key in valid_keys for key in link_keys):
+            raise ValueError(f"A key in one of the submitted link ({link_keys}) is not of the valid options: "
+                             f"{valid_keys}")
+        else:
+            return True

tests/test_foundry.py

@@ -356,6 +356,62 @@ def test_publish_with_https():
     _write_test_data(local_path)
 
     res = f.publish_dataset(pub_test_metadata, title, authors, https_data_path=local_path, short_name=short_name)
+    assert res['success']
+    assert res['source_id'] == f"_test_{short_name}_v1.1"
+
+
+@pytest.mark.skipif(bool(is_gha), reason="Not run as part of GHA CI")
+def test_publish_bad_links_with_https():
+    """System test: Assess the end-to-end publication of a dataset via HTTPS
+    """
+
+    f = Foundry(index="mdf-test", authorizers=auths)
+    timestamp = datetime.now().timestamp()
+    title = "https_publish_test_{:.0f}".format(timestamp)
+    short_name = "https_pub_{:.0f}".format(timestamp)
+    authors = ["A Scourtas"]
+    local_path = "./data/https_test"
+    links = {"horse": "link", "doi": "3", "url": "www.test.com", "description": "string", "bibtex": "bib"}
+
+    # create test JSON to upload (if it doesn't already exist)
+    _write_test_data(local_path)
+
+    with pytest.raises(Exception) as exc_info:
+        f.publish_dataset(pub_test_metadata,
+                          title,
+                          authors,
+                          https_data_path=local_path,
+                          short_name=short_name,
+                          links=links)
+    # err = exc_info.value
+    # assert hasattr(err, '__cause__')
+    # assert isinstance(err.__cause__, ValueError)
+    assert isinstance(exc_info.type(), ValueError)
+    _delete_test_data(f)
+
+
+@pytest.mark.skipif(bool(is_gha), reason="Not run as part of GHA CI")
+def test_publish_links_with_https():
+    """System test: Assess the end-to-end publication of a dataset via HTTPS
+    """
+
+    f = Foundry(index="mdf-test", authorizers=auths)
+    timestamp = datetime.now().timestamp()
+    title = "https_publish_test_{:.0f}".format(timestamp)
+    short_name = "https_pub_{:.0f}".format(timestamp)
+    authors = ["A Scourtas"]
+    local_path = "./data/https_test"
+    links = {"type": "link", "doi": "3", "url": "www.test.com", "description": "string", "bibtex": "bib"}
+
+    # create test JSON to upload (if it doesn't already exist)
+    _write_test_data(local_path)
+
+    res = f.publish_dataset(pub_test_metadata,
+                            title,
+                            authors,
+                            https_data_path=local_path,
+                            short_name=short_name,
+                            links=links)
 
     assert res['success']
     assert res['source_id'] == f"_test_{short_name}_v1.1"