Update synonym parsing to handle OMO CURIEs (#166)

* Update synonym parsing

* Handle preferred curies

* Enable searching for synonym type even without specificity

Author: cthoyt

src/pyobo/reader.py

@@ -444,20 +444,23 @@ def _extract_synonym(
             break
 
     stype: Optional[SynonymTypeDef] = None
-    if specificity is not None:  # go fishing for a synonym type definition
-        for _stype in synonym_typedefs.values():
-            # Since there aren't a lot of carefully defined synonym definitions, it
-            # can appear as a string or curie. Therefore, we might see temporary prefixes
-            # get added, so we should check against full curies as well as local unique
-            # identifiers
-            if rest.startswith(_stype.curie):
-                rest = rest[len(_stype.curie) :].strip()
-                stype = _stype
-                break
-            elif rest.startswith(_stype.identifier):
-                rest = rest[len(_stype.identifier) :].strip()
-                stype = _stype
-                break
+    for _stype in synonym_typedefs.values():
+        # Since there aren't a lot of carefully defined synonym definitions, it
+        # can appear as a string or curie. Therefore, we might see temporary prefixes
+        # get added, so we should check against full curies as well as local unique
+        # identifiers
+        if rest.startswith(_stype.curie):
+            rest = rest[len(_stype.curie) :].strip()
+            stype = _stype
+            break
+        elif rest.startswith(_stype.preferred_curie):
+            rest = rest[len(_stype.preferred_curie) :].strip()
+            stype = _stype
+            break
+        elif rest.startswith(_stype.identifier):
+            rest = rest[len(_stype.identifier) :].strip()
+            stype = _stype
+            break
 
     if not rest.startswith("[") or not rest.endswith("]"):
         logger.warning("[%s:%s] problem with synonym: %s", prefix, identifier, s)

tests/test_get.py

@@ -19,6 +19,7 @@
     iterate_node_synonyms,
     iterate_node_xrefs,
 )
+from pyobo.struct.struct import acronym
 from tests.constants import TEST_CHEBI_OBO_PATH, chebi_patch
 
 
@@ -105,6 +106,7 @@ def test_extract_synonym(self):
         iupac_name = SynonymTypeDef.from_text("IUPAC NAME", lower=False)
         synoynym_typedefs = {
             "IUPAC_NAME": iupac_name,
+            acronym.curie: acronym,
         }
 
         for expected_synonym, text in [
@@ -137,6 +139,22 @@ def test_extract_synonym(self):
                 Synonym(name="LTEC I", specificity="EXACT"),
                 '"LTEC I" []',
             ),
+            (
+                Synonym(name="HAdV-A", specificity="BROAD", type=acronym),
+                '"HAdV-A" BROAD OMO:0003012 []',
+            ),
+            (
+                Synonym(name="HAdV-A", specificity="BROAD", type=acronym),
+                '"HAdV-A" BROAD omo:0003012 []',
+            ),
+            (
+                Synonym(name="HAdV-A", specificity="EXACT", type=acronym),
+                '"HAdV-A" OMO:0003012 []',
+            ),
+            (
+                Synonym(name="HAdV-A", specificity="EXACT", type=acronym),
+                '"HAdV-A" omo:0003012 []',
+            ),
         ]:
             with self.subTest(s=text):
                 actual_synonym = _extract_synonym(