Allow provider specific examples

src/bioregistry/app/templates/resource.html

@@ -613,7 +613,8 @@ <h5 style="margin: 0">Providers</h5>
                 {% if providers %}
                 <p>
                     The local unique identifier <code>{{ example }}</code> is used to demonstrate the providers
-                    available for {{ name_pack.value }}. A guide for curating additional providers can be found
+                    available for {{ name_pack.value }}. Some providers may use a different example, which is displayed in the table below. 
+                    A guide for curating additional providers can be found
                     <a href="https://biopragmatics.github.io/bioregistry/curation/providers">here</a>.
                 </p>
                 {% else %}
@@ -651,7 +652,8 @@ <h5 class="card-header">Extra Providers</h5>
                         <tr>
                             <td><code>{{ provider.code }}</code></td>
                             <td>{{ provider.name }}</td>
-                            {% set url = provider.resolve(example) %}
+                            {% set example_id = provider.example if provider.example else example %}
+                            {% set url = provider.resolve(example_id) %}
                             <td><a href="{{ url }}">{{ url }}</a></td>
                         </tr>
                     {% endfor %}

src/bioregistry/app/ui.py

@@ -118,7 +118,7 @@ def resource(prefix: str) -> str | flask.Response:
 
     example = _resource.get_example()
     example_curie = _resource.get_example_curie(use_preferred=True)
-    example_extras = _resource.example_extras or []
+    example_extras = _resource.get_example_extras()
     example_curie_extras = [
         _resource.get_curie(example_extra, use_preferred=True) for example_extra in example_extras
     ]

src/bioregistry/benchmarks/uri_parsing.py

@@ -34,7 +34,7 @@ def iter_uris() -> Iterable[tuple[str, str, str, str]]:
         for metaprefix, url in manager.get_providers_list(prefix, example):
             if url.endswith(example):  # skip funny formats
                 yield prefix, example, metaprefix, url
-        for extra_example in resource.example_extras or []:
+        for extra_example in resource.get_example_extras():
             for metaprefix, url in manager.get_providers_list(prefix, extra_example):
                 if url.endswith(extra_example):
                     yield prefix, extra_example, metaprefix, url

src/bioregistry/curation/add_ontology_regexes.py

@@ -13,7 +13,7 @@
     example = resource.get_example()
     if example:
         rv.append(example)
-    rv.extend(resource.example_extras or [])
+    rv.extend(resource.get_example_extras())
     return rv
 
 

src/bioregistry/data/bioregistry.json

@@ -89050,6 +89050,7 @@
       {
         "code": "furna",
         "description": "FURNA (Functions of RNAs) is a database of ligand-RNA interactions and Gene Ontology annotations for RNAs in the Protein Data Bank (PDB).",
+        "example": "157d",
         "homepage": "https://seq2fun.dcmb.med.umich.edu/furna/",
         "name": "furna",
         "publications": [

src/bioregistry/resource_manager.py

@@ -899,7 +899,7 @@ def rasterized_resource(self, resource: Resource) -> Resource:
             download_rdf=resource.get_download_rdf(),
             # Registry properties
             example=resource.get_example(),
-            example_extras=resource.example_extras,
+            example_extras=resource.get_example_extras(),
             example_decoys=resource.example_decoys,
             uri_format=resource.get_uri_format(),
             rdf_uri_format=resource.get_rdf_uri_format(),

src/bioregistry/schema/schema.json

@@ -398,6 +398,19 @@
           "default": null,
           "description": "A list of publications about the provider. See the `indra` provider for `hgnc` for an example.",
           "title": "Publications"
+        },
+        "example": {
+          "anyOf": [
+            {
+              "type": "string"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "default": null,
+          "description": "An example local identifier, specific to the provider. Providing this value is only necessary if the example associated with the prefix for which this is a provider is not resolvable by the provider. The example identifier should exclude any redundant usage of the prefix. For example, a GO identifier should only look like ``1234567`` and not like ``GO:1234567``",
+          "title": "Example"
         }
       },
       "required": [

src/bioregistry/schema/struct.py

@@ -260,6 +260,14 @@
         default=None,
         description="A list of publications about the provider. See the `indra` provider for `hgnc` for an example.",
     )
+    example: str | None = Field(
+        default=None,
+        description="An example local identifier, specific to the provider. Providing this value is "
+        "only necessary if the example associated with the prefix for which this is a provider "
+        "is not resolvable by the provider. The example identifier should exclude any redundant "
+        "usage of the prefix. For example, a GO identifier should only "
+        "look like ``1234567`` and not like ``GO:1234567``",
+    )
 
     def resolve(self, identifier: str) -> str:
         """Resolve the identifier into a URI.
@@ -1355,9 +1363,18 @@
         example = self.get_example()
         if example:
             rv.append(example)
-        rv.extend(self.example_extras or [])
+        rv.extend(self.get_example_extras())
         return rv
 
+    def get_example_extras(self) -> list[str]:
+        """Aggregate manually curated examples with provider-specific examples."""
+        rv = set(self.example_extras or [])
+        if self.providers:
+            for provider in self.providers:
+                if provider.example:
+                    rv.add(provider.example)
+        return sorted(rv)
+
     def get_example_curie(self, use_preferred: bool = False) -> str | None:
         """Get an example CURIE, if an example identifier is available.
 

tests/test_data.py

@@ -421,15 +421,16 @@ def assert_is_valid_identifier(self, prefix: str, example: str) -> None:
     def test_extra_examples(self):
         """Test extra examples."""
         for prefix, entry in self.registry.items():
-            if not entry.example_extras:
+            example_extras = entry.get_example_extras()
+            if not example_extras:
                 continue
             primary_example = entry.get_example()
             with self.subTest(prefix=prefix):
                 self.assertIsNotNone(
                     primary_example, msg="entry has extra examples but not primary example"
                 )
 
-            for example in entry.example_extras:
+            for example in example_extras:
                 with self.subTest(prefix=prefix, identifier=example):
                     self.assertEqual(entry.standardize_identifier(example), example)
                     self.assertNotEqual(
@@ -438,8 +439,8 @@ def test_extra_examples(self):
                     self.assert_is_valid_identifier(prefix, example)
 
             self.assertEqual(
-                len(entry.example_extras),
-                len(set(entry.example_extras)),
+                len(example_extras),
+                len(set(example_extras)),
                 msg="duplicate extra examples",
             )