Merge branch 'main' into feat/fk-musts

Author: martonvago

CHANGELOG.md

@@ -9,6 +9,12 @@ often, sometimes several in a day. It also means any individual release
 will not have many changes within it. Below is a list of releases along
 with what was changed within it.
 
+## 0.24.0 (2025-12-08)
+
+### Feat
+
+- :sparkles: check primary key exists (#218)
+
 ## 0.23.1 (2025-11-27)
 
 ### Fix

CITATION.cff

@@ -1,6 +1,8 @@
 title: "Check Data Package: Ensure the correctness and compliance of your Data Package"
-# TODO: Add abstract of Python package.
-abstract: ""
+abstract: |
+  Check your Data Package's metadata against the Data Package standard to ensure that
+  it's compliant with the standard. See https://check-datapackage.seedcase-project.org
+  for more information.
 authors:
   - family-names: Johnston
     given-names: Luke William
@@ -18,9 +20,12 @@ authors:
     given-names: Marton
     affiliation: "Steno Diabetes Center Aarhus"
     orcid: "https://orcid.org/0009-0007-4628-655X"
+  - family-names: Ostblom
+    given-names: Joel
+    orcid: "https://orcid.org/0000-0003-0051-3239"
 cff-version: 1.2.0
-# doi: ""
-# date-released: ""
+doi: "10.5281/zenodo.17733655"
+date-released: "2025-11-27"
 keywords:
   - "data package"
   - "data package checks"
@@ -31,9 +36,7 @@ keywords:
   - "metadata"
   - "metadata checks"
   - "metadata verification"
-  - "verification"
   - "data engineering"
-  - "metadata checks"
 license: MIT
 message: "If you use this Python package, please cite it using these metadata."
 repository-code: "https://github.com/seedcase-project/check-datapackage"

README.md

@@ -140,3 +140,26 @@ This project is licensed under the [MIT License](LICENSE.md).
 ## Changelog
 
 For a list of changes, see our [changelog](CHANGELOG.md) page.
+
+## Citing
+
+This project is part of the [Seedcase
+Project](https://seedcase-project.org), which is a collaborative effort
+to create a framework for data management and analysis in research. If
+you use this project in your work, please cite it as follows:
+
+Johnston L.W., Brødbæk S.K., Beicher K., Vago M., Ostblom J. (2025).
+Check Data Package: Ensure the correctness and compliance of your Data
+Package DOI: 10.5281/zenodo.17733655 URL:
+https://check-datapackage.seedcase-project.org
+
+Or as a BibTeX entry:
+
+    @misc{YourReferenceHere,
+    author = {Johnston, Luke William and Brødbæk, Signe Kirk and Beicher, Kristiane and Vago, Marton and Ostblom, Joel},
+    doi = {10.5281/zenodo.17733655},
+    month = {11},
+    title = {Check Data Package: Ensure the correctness and compliance of your Data Package},
+    url = {https://check-datapackage.seedcase-project.org},
+    year = {2025}
+    }

README.qmd

@@ -2,6 +2,9 @@
 format: gfm
 metadata-files:
   - _metadata.yml
+execute:
+  echo: false
+jupyter: python3
 ---
 
 <!-- NOTE: This README.md is auto-generated from README.qmd. Edit that file. -->
@@ -119,3 +122,21 @@ This project is licensed under the [MIT License](LICENSE.md).
 ## Changelog
 
 For a list of changes, see our [changelog](CHANGELOG.md) page.
+
+## Citing
+
+This project is part of the [Seedcase
+Project](https://seedcase-project.org), which is a collaborative effort
+to create a framework for data management and analysis in research. If
+you use this project in your work, please cite it as follows:
+
+```{python}
+#| output: asis
+!uvx --quiet cffconvert --format apalike
+```
+
+Or as a BibTeX entry:
+
+```{python}
+!uvx --quiet cffconvert --format bibtex
+```

pyproject.toml

@@ -1,19 +1,20 @@
 [project]
 name = "check-datapackage"
-version = "0.23.1"
-# TODO: Add a description of the package.
-description = ""
+version = "0.24.0"
+description = "Check your Data Package's metadata against the Data Package standard to ensure that it's compliant with the standard"
 authors = [
   {name = "Luke W. Johnston", email = "lwjohnst@gmail.com" },
   {name = "Kristiane Beicher", email = "kris.beicher@clin.au.dk" },
   {name = "Signe Kirk Brødbæk", email = "signekb@clin.au.dk" },
   {name = "Marton Vago", email = "marton.vago95@gmail.com" },
+  {name = "Joel Ostblom", email = "joel@joelostblom.com" },
 ]
 maintainers = [
   {name = "Luke W. Johnston", email = "lwjohnst@gmail.com" },
   {name = "Kristiane Beicher", email = "kris.beicher@clin.au.dk" },
   {name = "Signe Kirk Brødbæk", email = "signekb@clin.au.dk" },
   {name = "Marton Vago", email = "marton.vago95@gmail.com" },
+  {name = "Joel Ostblom", email = "joel@joelostblom.com" },
 ]
 readme = "README.md"
 license = "MIT"

src/check_datapackage/check.py

@@ -143,7 +143,14 @@ class for more details, especially about the default values.
 def _check_keys(properties: dict[str, Any], issues: list[Issue]) -> list[Issue]:
     """Check that primary and foreign keys exist."""
     # Primary keys
-    key_issues = []
+    resources_with_pk = _get_fields_at_jsonpath(
+        "$.resources[?(length(@.schema.primaryKey) > 0 || @.schema.primaryKey == '')]",
+        properties,
+    )
+    resources_with_pk = _keep_resources_with_no_issue_at_property(
+        resources_with_pk, issues, "schema.primaryKey"
+    )
+    key_issues = _flat_map(resources_with_pk, _check_primary_key)
 
     # Foreign keys
     resources_with_fk = _get_fields_at_jsonpath(
@@ -179,6 +186,27 @@ def _keep_resources_with_no_issue_at_property(
     )
 
 
+def _check_primary_key(resource: PropertyField) -> list[Issue]:
+    """Check that primary key fields exist in the resource."""
+    pk_fields = resolve("/schema/primaryKey", resource.value)
+    pk_fields_list = _key_fields_as_str_list(pk_fields)
+    unknown_fields = _get_unknown_key_fields(pk_fields_list, resource.value)
+
+    if not unknown_fields:
+        return []
+
+    return [
+        Issue(
+            jsonpath=f"{resource.jsonpath}.schema.primaryKey",
+            type="primary-key",
+            message=(
+                f"No fields found in resource for primary key fields: {unknown_fields}."
+            ),
+            instance=pk_fields,
+        )
+    ]
+
+
 def _check_foreign_keys(
     resource: PropertyField, properties: dict[str, Any]
 ) -> list[Issue]:

tests/test_check.py

@@ -91,6 +91,60 @@ def test_fails_properties_with_pattern_mismatch():
     assert issues[0].jsonpath == "$.contributors[0].path"
 
 
+@mark.parametrize("primary_key", ["id", ["id", "name"]])
+def test_pass_good_primary_key(primary_key):
+    properties = example_package_properties()
+    properties["resources"][0]["schema"]["primaryKey"] = primary_key
+    properties["resources"][0]["schema"]["fields"].extend(
+        [
+            {"name": "id", "type": "integer"},
+            {"name": "name", "type": "string"},
+        ]
+    )
+
+    issues = check(properties)
+
+    assert issues == []
+
+
+@mark.parametrize("primary_key", ["", "last_name", ["first_name", "last_name"]])
+def test_fail_primary_key_with_unknown_fields(primary_key):
+    properties = example_package_properties()
+    properties["resources"][0]["schema"]["primaryKey"] = primary_key
+
+    issues = check(properties)
+
+    assert len(issues) == 1
+    assert issues[0].jsonpath == "$.resources[0].schema.primaryKey"
+    assert issues[0].type == "primary-key"
+    assert issues[0].instance == primary_key
+
+
+@mark.parametrize("primary_key", [None, 123, [], [123, "a_field"]])
+def test_do_not_check_bad_primary_key_against_fields(primary_key):
+    properties = example_package_properties()
+    properties["resources"][0]["schema"]["primaryKey"] = primary_key
+
+    issues = check(properties)
+
+    assert len(issues) == 1
+    assert issues[0].type != "primary-key"
+
+
+def test_do_not_check_primary_key_against_bad_field():
+    properties = example_package_properties()
+    properties["resources"][0]["schema"]["primaryKey"] = "eye-colour"
+    properties["resources"][0]["schema"]["fields"].append(
+        # Bad name
+        {"name": 123, "type": "integer"},
+    )
+
+    issues = check(properties)
+
+    assert len(issues) == 1
+    assert issues[0].type != "primary-key"
+
+
 def test_pass_good_foreign_keys_same_resource():
     properties = example_package_properties()
     properties["resources"][0]["schema"]["fields"].extend(
@@ -832,16 +886,6 @@ def test_fail_foreign_keys_with_bad_array_item():
     )
 
 
-@mark.parametrize("primary_key", ["id", ["name", "address"]])
-def test_pass_good_primary_key(primary_key):
-    properties = example_package_properties()
-    properties["resources"][0]["schema"]["primaryKey"] = primary_key
-
-    issues = check(properties)
-
-    assert issues == []
-
-
 def test_fail_primary_key_of_bad_type():
     properties = example_package_properties()
     properties["resources"][0]["schema"]["primaryKey"] = 123