Use curies' Converter in strict=False mode. (#409)

- [x] Fixes #408 
- [x] Fixes #269 

This way, when a `prefix_map` has duplicate `uri_prefix` or `prefix`
from the user, `curies` will not throw an error. `sssom-py` already
gives priority to a user-defined prefix map over the default one (which
now is EPM from bioregistry).

---------

Co-authored-by: Nico Matentzoglu <[email protected]>

src/sssom/parsers.py

@@ -667,7 +667,7 @@ def from_obographs(
     :return: An SSSOM data frame (MappingSetDataFrame)
     """
     _ensure_prefix_map(prefix_map)
-    converter = Converter.from_prefix_map(prefix_map)
+    converter = Converter.from_prefix_map(prefix_map, strict=False)
     ms = _init_mapping_set(meta)
     mlist: List[Mapping] = []
     # bad_attrs = {}

tests/data/hp-subset-metadata.yml

@@ -0,0 +1,2 @@
+curie_map:
+  HP: http://example/obo/HP_

tests/data/hp-subset.json

@@ -0,0 +1,59 @@
+{
+  "graphs" : [ {
+    "nodes" : [ {
+      "id" : "http://example/obo/HP_0011770",
+      "meta" : {
+        "definition" : {
+          "val" : "A type of hyperparathyroidism that occurs following kidney transplantation, which is a treatment for secondary hyperparathyroidism. Although kidney transplantation leads to a normalization of serum calcium and parathyroid hormone in most patients. The state of persistent hypercalcemia and hyperparathyroidism is referred to as tertiary hyperparathyroidism.",
+          "xrefs" : [ "DDD:spark" ]
+        },
+        "xrefs" : [ {
+          "val" : "SNOMEDCT_US:78200003"
+        }, {
+          "val" : "UMLS:C0271858"
+        } ],
+        "basicPropertyValues" : [ {
+          "pred" : "http://www.geneontology.org/formats/oboInOwl#created_by",
+          "val" : "peter"
+        }, {
+          "pred" : "http://www.geneontology.org/formats/oboInOwl#creation_date",
+          "val" : "2012-04-22T04:38:20Z"
+        }, {
+          "pred" : "http://www.geneontology.org/formats/oboInOwl#hasOBONamespace",
+          "val" : "human_phenotype"
+        } ]
+      },
+      "type" : "CLASS",
+      "lbl" : "Tertiary hyperparathyroidism"
+    }, {
+      "id" : "http://example/obo/HP_0009450",
+      "meta" : {
+        "definition" : {
+          "val" : "Increased width of the proximal phalanx of the 3rd finger.",
+          "xrefs" : [ "HPO:curators" ]
+        },
+        "xrefs" : [ {
+          "val" : "UMLS:C4024355"
+        } ],
+        "synonyms" : [ {
+          "pred" : "hasExactSynonym",
+          "val" : "Broad innermost bone of middle finger",
+          "xrefs" : [ "ORCID:0000-0001-5208-3432" ],
+          "synonymType" : "http://purl.obolibrary.org/obo/hp#layperson"
+        } ],
+        "basicPropertyValues" : [ {
+          "pred" : "http://www.geneontology.org/formats/oboInOwl#hasOBONamespace",
+          "val" : "human_phenotype"
+        }, {
+          "pred" : "http://www.geneontology.org/formats/oboInOwl#created_by",
+          "val" : "doelkens"
+        }, {
+          "pred" : "http://www.geneontology.org/formats/oboInOwl#creation_date",
+          "val" : "2009-01-14T04:33:41Z"
+        } ]
+      },
+      "type" : "CLASS",
+      "lbl" : "Broad proximal phalanx of the 3rd finger"
+    } ]
+    } ]
+}

tests/test_parsers.py

@@ -13,6 +13,7 @@
 from rdflib import Graph
 
 from sssom.context import _raise_on_invalid_prefix_map, get_default_metadata
+from sssom.io import parse_file
 from sssom.parsers import (
     from_alignment_minidom,
     from_obographs,
@@ -223,3 +224,25 @@ def test_read_sssom_table(self):
                         self.assertEqual(imported_df.iloc[idx][k], v)
                     else:
                         self.assertEqual(imported_df.iloc[idx][k], v)
+
+    def test_parse_obographs_merged(self):
+        """Test parsing OBO Graph JSON using custom prefix_map."""
+        hp_json = f"{test_data_dir}/hp-subset.json"
+        hp_meta = f"{test_data_dir}/hp-subset-metadata.yml"
+        outfile = f"{test_out_dir}/hp-subset-parse.tsv"
+
+        with open(hp_meta, "r") as f:
+            data = yaml.safe_load(f)
+            custom_curie_map = data["curie_map"]
+
+        with open(outfile, "w") as f:
+            parse_file(
+                input_path=hp_json,
+                prefix_map_mode="merged",
+                clean_prefixes=True,
+                input_format="obographs-json",
+                metadata_path=hp_meta,
+                output=f,
+            )
+        msdf = parse_sssom_table(outfile)
+        self.assertTrue(custom_curie_map.items() <= msdf.prefix_map.items())