Add upper- and lowercase prefix synonyms

src/bioregistry/data/bioregistry.json

@@ -36468,7 +36468,6 @@
       "name": "Drosophila Phenotype Ontology",
       "prefix": "FBcv"
     },
-    "part_of": "flybase",
     "pattern": "^\\d{7}$",
     "prefixcommons": {
       "bioportal": "1017",
@@ -41924,6 +41923,9 @@
         "Phenomics",
         "Comparative Genomics",
         "Omics"
+      ],
+      "synonyms": [
+        "GEO"
       ]
     },
     "go": {

src/bioregistry/data/contexts.json

@@ -1,7 +1,9 @@
 {
   "obo": {
     "blacklist": [
-      "icd9"
+      "icd9",
+      "orphanet",
+      "gro"
     ],
     "custom_prefix_map": {
       "PMID": "https://www.ncbi.nlm.nih.gov/pubmed/"
@@ -29,7 +31,10 @@
       "default"
     ],
     "prefix_remapping": {
+      "GEO": "ncbi.geo",
+      "cpga": "GRO",
       "ensembl": "ENSEMBL",
+      "geogeo": "GEO",
       "icd10": "ICD10WHO",
       "orphanet.ordo": "Orphanet",
       "pubmed": "PMID",

src/bioregistry/record_accumulator.py

@@ -133,6 +133,7 @@ def get_converter(
         converter = curies.remap_curie_prefixes(converter, remapping)
     if rewiring:
         converter = curies.rewire(converter, rewiring)
+    converter = _enrich_converter_synonyms(converter)
     return converter
 
 
@@ -340,3 +341,24 @@ def _add_prefix_prefixes(
         )
 
     return [record for _, record in sorted(records.items())]
+
+
+def _enrich_converter_synonyms(converter: Converter) -> Converter:
+    return Converter([_enrich_record_synonyms(r) for r in converter.records])
+
+
+def _enrich_record_synonyms(record: curies.Record) -> curies.Record:
+    sss = set()
+    for s in [record.prefix, *record.prefix_synonyms]:
+        sss.update(_generate_variants(s))
+    record.prefix_synonyms = sorted(sss - {record.prefix})
+    return record
+
+
+def _generate_variants(s: str):
+    yield s
+    yield s.lower()
+    yield s.upper()
+    yield s.replace("_", "")
+    yield s.replace("_", "").upper()
+    yield s.replace("_", "").lower()

tests/test_contexts.py

@@ -5,6 +5,8 @@
 import json
 import unittest
 
+import pytest
+
 import bioregistry
 from bioregistry import Resource, manager
 from bioregistry.constants import CONTEXTS_PATH
@@ -44,6 +46,8 @@ def test_obo_context(self):
         self.assertEqual(f"{p}/KISAO_", prefix_map["KISAO"])
         self.assertIn("FBcv", prefix_map)
         self.assertEqual(f"{p}/FBcv_", prefix_map["FBcv"])
+        self.assertNotIn("geo", prefix_map)
+        self.assertIn("ncbi.geo", prefix_map)
         self.assertIn("GEO", prefix_map)
         self.assertEqual(f"{p}/GEO_", prefix_map["GEO"])
         self.assertEqual("https://www.ncbi.nlm.nih.gov/pubmed/", prefix_map["PMID"])
@@ -58,15 +62,22 @@ def test_obo_context(self):
             msg="When overriding, this means that bioregistry prefix isn't properly added to the synonyms list",
         )
 
+    @pytest.mark.slow
     def test_obo_converter(self):
         """Test getting a converter from a context."""
         converter = manager.get_converter_from_context("obo")
         self.assertEqual("ICD10WHO", converter.standardize_prefix("icd10"))
         self.assertEqual("Orphanet", converter.standardize_prefix("ordo"))
-        self.assertEqual("GO", converter.standardize_prefix("GO"))
-        self.assertEqual("GO", converter.standardize_prefix("gomf"))
+        self.assertEqual("GO", converter.standardize_prefix("GO", strict=True))
+        self.assertEqual("GO", converter.standardize_prefix("gomf", strict=True))
         self.assertEqual("https://www.ncbi.nlm.nih.gov/pubmed/", converter.bimap["PMID"])
+        self.assertEqual("GO", converter.standardize_prefix("go", strict=True))
         self.assertEqual("GO", converter.standardize_prefix("go"))
+        self.assertEqual("PMID", converter.standardize_prefix("pmid", strict=True))
+        self.assertEqual("PMID", converter.standardize_prefix("pubmed", strict=True))
+        self.assertEqual("PMID", converter.standardize_prefix("PubMed", strict=True))
+        self.assertEqual("PMID", converter.standardize_prefix("PUBMED"))
+        self.assertEqual("PMID", converter.standardize_prefix("PMID"))
         self.assertEqual("oboInOwl", converter.standardize_prefix("oboinowl"))
 
     def test_data(self):
@@ -95,36 +106,9 @@ def test_data(self):
                     self.valid_metaprefixes.union({"obofoundry.preferred", "preferred", "default"}),
                 )
             remapping = context.prefix_remapping or {}
-            _valid_remapping_prefixes = set(
-                bioregistry.get_prefix_map(
-                    uri_prefix_priority=context.uri_prefix_priority,
-                )
-            )
+            _valid_remapping_prefixes = set(manager.converter.prefix_map)
             for prefix in remapping:
-                # Currently this allows overwriting of existing prefixes
                 self.assertIn(prefix, _valid_remapping_prefixes)
 
-            _valid_custom_prefixes = set(
-                bioregistry.get_prefix_map(
-                    remapping=remapping,
-                    uri_prefix_priority=context.uri_prefix_priority,
-                )
-            )
-            invalid_custom_prefixes = {
-                prefix
-                for prefix in (context.custom_prefix_map or {})
-                if prefix not in _valid_custom_prefixes
-            }
-            self.assertEqual(
-                0,
-                len(invalid_custom_prefixes),
-                msg=f"""
-
-    All prefixes in the custom prefix mapping should either be canonical prefixes or generated by the prefix remapping
-
-    Invalid prefixes: {", ".join(sorted(invalid_custom_prefixes))}
-""",
-            )
-
             for blacklist_prefix in context.blacklist or []:
                 self.assertIn(blacklist_prefix, self.valid_prefixes)

tests/test_manager.py

@@ -4,6 +4,8 @@
 
 import unittest
 
+import pytest
+
 import bioregistry
 from bioregistry import Manager, parse_curie
 from bioregistry.export.rdf_export import get_full_rdf
@@ -261,3 +263,16 @@ def test_external_registry_mappings(self):
         self.assertIn("loggerhead", res.source_only)
         # This is a non-ontology so it won't get in OBO Foundry
         self.assertIn("DCTERMS", res.target_only)
+
+    @pytest.mark.slow
+    def test_converter(self):
+        """Test standardizing prefixes with the converter."""
+        converter = self.manager.get_converter()
+        self.assertEqual("meddra", converter.standardize_prefix("MEDDRA"))
+        self.assertEqual("meddra", converter.standardize_prefix("MedDRA"))
+
+        self.assertEqual("pubmed", converter.standardize_prefix("PMID"))
+        self.assertEqual("pubmed", converter.standardize_prefix("PUBMED"))
+        self.assertEqual("pubmed", converter.standardize_prefix("pmid"))
+        self.assertEqual("pubmed", converter.standardize_prefix("pubmed"))
+        self.assertEqual("pubmed", converter.standardize_prefix("PubMed"))

tox.ini

@@ -27,7 +27,7 @@ envlist =
 
 [testenv]
 commands =
-    coverage run -p -m pytest --durations=20 {posargs:tests}
+    coverage run -p -m pytest --durations=20 --disable-warnings {posargs:tests}
     coverage combine
     coverage xml
 deps =