Add URI example

Author: cthoyt

docs/source/preprocessing.rst

@@ -53,6 +53,30 @@ OBO Foundry's PURL rules
     >>> converter.parse_curie("APOLLO:SV_1234567")
     ReferenceTuple('APOLLO_SV', '1234567')
 
+The CURIE and URI rewrites are unified. Therefore, you can also use a URI as a rewrite,
+such as handling Creative Commons license URLs, which unfortunately aren't themselves
+part of a semantic space for licenses. Luckily, SPDX is, and we can remap to that.
+
+.. code-block:: python
+
+    import curies
+    from curies import PreprocessingRules, PreprocessingConverter, PreprocessingRewrites
+
+    rules = PreprocessingRules(
+        rewrites=PreprocessingRewrites(
+            full={"http://creativecommons.org/licenses/by/3.0/": "spdx:CC-BY-3.0",},
+        )
+    )
+
+    converter = curies.get_obo_converter()
+    converter.add_prefix("spdx", "https://spdx.org/licenses/")
+    converter = PreprocessingConverter.from_converter(
+        converter, rules=rules,
+    )
+
+    >>> converter.parse_uri("http://creativecommons.org/licenses/by/3.0/")
+    ReferenceTuple('spdx', 'CC-BY-3.0')
+
 Some rewrite rules only apply to a specific resource, because of its own quirks in
 curation or encoding. For example, CHMO encodes OrangeBook entries with ``orange`` as a
 prefix, which is not typically specific enough to warrant curating ``orange`` as a

src/curies/preprocessing.py

@@ -367,8 +367,8 @@ def parse_uri(
         :param uri: The URI to parse and standardize
         :param strict: If the URI can't be parsed, should an error be thrown? Defaults
             to false.
-        :param return_none: A dummy value, do not use. If given as False,
-            will raise a not implemented error
+        :param return_none: A dummy value, do not use. If given as False, will raise a
+            not implemented error
         :param context: Is there a context, e.g., an ontology prefix that should be
             applied to the remapping and blocklist rules?
         :param block_action: What action should be taken when the blocklist is invoked?

tests/test_preprocessing.py

@@ -39,7 +39,10 @@ def setUpClass(cls) -> None:
         """Set up the test case."""
         cls.rules = PreprocessingRules(
             rewrites=PreprocessingRewrites(
-                full={"is_a": "rdf:type"},
+                full={
+                    "is_a": "rdf:type",
+                    "http://creativecommons.org/licenses/by/3.0/": "spdx:CC-BY-3.0",
+                },
                 prefix={
                     "OMIM:PS": "omim.ps:",
                     "omim:PS": "omim.ps:",
@@ -76,6 +79,7 @@ def setUpClass(cls) -> None:
                 "omim": "https://omim.org/MIM:",
                 "omim.ps": "https://omim.org/phenotypicSeries/PS",
                 "pubmed": "http://rdf.ncbi.nlm.nih.gov/pubchem/reference/",
+                "spdx": "https://spdx.org/licenses/",
             }
         )
 
@@ -120,6 +124,10 @@ def test_global_full_rewrite(self) -> None:
         """Test global full string rewrite."""
         self.assertEqual(ReferenceTuple("rdf", "type"), self.converter.parse("is_a"))
         self.assertEqual(ReferenceTuple("rdf", "type"), self.converter.parse_curie("is_a"))
+        self.assertEqual(
+            ReferenceTuple("spdx", "CC-BY-3.0"),
+            self.converter.parse_uri("http://creativecommons.org/licenses/by/3.0/"),
+        )
 
     def test_global_prefix_rewrite(self) -> None:
         """Test global prefix rewrite."""