Update

biopragmatics · Nov 2, 2023 · 0004f34 · 0004f34
1 parent b6ed42b
commit 0004f34
Show file tree

Hide file tree

Showing 4 changed files with 20 additions and 41 deletions.
diff --git a/src/bioregistry/data/bioregistry.json b/src/bioregistry/data/bioregistry.json
@@ -33721,7 +33721,6 @@
       "name": "Drosophila Phenotype Ontology",
       "prefix": "FBcv"
     },
-    "part_of": "flybase",
     "pattern": "^\\d{7}$",
     "prefixcommons": {
       "bioportal": "1017",
@@ -38805,6 +38804,9 @@
         "Phenomics",
         "Comparative Genomics",
         "Omics"
+      ],
+      "synonyms": [
+        "GEO"
       ]
     },
     "go": {

diff --git a/src/bioregistry/data/contexts.json b/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/"
@@ -29,8 +31,11 @@
       "default"
     ],
     "prefix_remapping": {
+      "GEO": "ncbi.geo",
       "ensembl": "ENSEMBL",
+      "geogeo": "GEO",
       "icd10": "ICD10WHO",
+      "cpga": "GRO",
       "orphanet.ordo": "Orphanet",
       "pubmed": "PMID",
       "snomedct": "SCTID",

diff --git a/src/bioregistry/record_accumulator.py b/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
 
 
@@ -159,13 +160,6 @@ def _get_records(  # noqa: C901
         resource.prefix: resource.get_priority_prefix(priority=prefix_priority)
         for resource in resource_dict.values()
     }
-    dd = defaultdict(list)
-    for k, v in primary_prefixes.items():
-        dd[v.lower()].append(k)
-    dd = {k: v for k, v in dd.items() if len(v) > 1}
-    if dd:
-        raise ValueError(f"Duplicate prefixes: {dd}")
-
     pattern_map = {
         prefix: pattern
         for prefix in primary_prefixes
@@ -338,20 +332,22 @@ def _add_prefix_prefixes(
         primary_uri_prefix = primary_uri_prefixes[prefix]
         if not primary_prefix or not primary_uri_prefix:
             continue
-        record = curies.Record(
+        records[prefix] = curies.Record(
             prefix=primary_prefix,
             prefix_synonyms=sorted(secondary_prefixes[prefix] - {primary_prefix}),
             uri_prefix=primary_uri_prefix,
             uri_prefix_synonyms=sorted(secondary_uri_prefixes[prefix] - {primary_uri_prefix}),
             pattern=pattern_map.get(prefix),
         )
-        record = _enrich_synonyms(record)
-        records[prefix] = record
 
     return [record for _, record in sorted(records.items())]
 
 
-def _enrich_synonyms(record: curies.Record) -> curies.Record:
+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))

diff --git a/tests/test_contexts.py b/tests/test_contexts.py
@@ -8,6 +8,7 @@
 import pytest
 
 import bioregistry
+import curies
 from bioregistry import Resource, manager
 from bioregistry.constants import CONTEXTS_PATH
 
@@ -46,6 +47,8 @@ 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"])
@@ -102,36 +105,9 @@ 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)