Slurp / migrate

- Bugfix: Robot template 'parents' column IDs were not Mondo IDs
- Add: tests

Unit testing
- Add: test/ dir, and initialized with the necessary files to test slurp / migrate

.gitignore

@@ -78,3 +78,6 @@ src/scripts/.ipynb_checkpoints/*
 src/ontology/.template.db
 src/mappings/mondo-sources-all-lexical.sssom.tsv
 src/scripts/mondo_unmapped.tsv
+
+# Test
+test/output/

src/scripts/migrate.py

@@ -10,8 +10,10 @@
 - https://incatools.github.io/ontology-access-kit/intro/tutorial02.html
 """
 import os
+import sys
 from argparse import ArgumentParser
 from glob import glob
+from pathlib import Path
 from typing import Dict, List, Set
 
 import pandas as pd
@@ -20,7 +22,10 @@
 from oaklib.implementations import ProntoImplementation
 from oaklib.types import CURIE, URI
 
-from utils import CACHE_DIR, DOCS_DIR, PREFIX, PROJECT_DIR, Term, _get_all_owned_terms, _get_next_available_mondo_id, \
+SCRIPTS_DIR = Path(os.path.abspath(os.path.dirname(__file__)))
+PROJECT_ROOT = SCRIPTS_DIR.parent.parent
+sys.path.insert(0, str(PROJECT_ROOT))
+from src.scripts.utils import CACHE_DIR, DOCS_DIR, PREFIX, PROJECT_DIR, Term, _get_all_owned_terms, _get_next_available_mondo_id, \
     get_mondo_term_ids, _load_ontology, SLURP_DIR
 
 
@@ -103,9 +108,12 @@ def slurp(
         onto_config_path=onto_config_path, use_cache=use_cache)
     slurp_candidates: List[Term] = [x for x in owned_terms if all([x.curie not in y for y in [excluded, mapped]])]
     match_types: Dict = {}
+    mondo_id_map: Dict = {}
     for row in sssom_df.itertuples():
         # noinspection PyUnresolvedReferences
         match_types[row.object_id] = row.predicate_id
+        # noinspection PyUnresolvedReferences
+        match_types[row.object_id] = row.subject_id
 
     # Determine slurpable / migratable terms
     # To be migratable, the term (i) must not already be mapped, (ii) must not be excluded (e.g. not in
@@ -120,13 +128,15 @@ def slurp(
         else:
             next_mondo_id, mondo_term_ids = _get_next_available_mondo_id(next_mondo_id, max_id, mondo_term_ids)
             mondo_id = 'MONDO:' + str(next_mondo_id).zfill(7)  # leading 0-padding
+        qualified_parents = [p for p in t.direct_owned_parent_curies
+                             if p in match_types and match_types[p] in ['skos:exactMatch', 'skos:narrowMatch']]
+        qualified_mondo_paents = [mondo_id_map[p] for p in qualified_parents if p in mondo_id_map]
         mondo_label = t.label.lower() if t.label else ''
         terms_to_slurp.append({
             'mondo_id': mondo_id, 'mondo_label': mondo_label, 'xref': t.curie, 'xref_source': 'MONDO:equivalentTo',
             'original_label': t.label if t.label else '', 'definition': t.definition if t.definition else '',
             # if not in match_types, this should mean term is excluded or obsolete
-            'parents': '|'.join([p for p in t.direct_owned_parent_curies if p in match_types
-                                 and match_types[p] in ['skos:exactMatch', 'skos:narrowMatch']])})
+            'parents': '|'.join(qualified_mondo_paents)})
 
     # Sort, add robot row, save and return
     result = pd.DataFrame(terms_to_slurp)