changes to . VCF value mapping

Author: ecrum19

src/run_conversion.sh

@@ -159,6 +159,17 @@ count_triples_json() {
   echo "}"
 }
 
+# Replace plain VCF null marker literals (`"."`) with typed null literals.
+# Ontology alignment:
+#   "."  ->  "."^^vcfr:Null
+annotate_null_literals_nt() {
+  local nt_path="$1"
+  local tmp_path
+  tmp_path="$(mktemp "${nt_path}.nullfix.XXXXXX")"
+  sed -E 's/"\."([[:space:]]+\.)/"."^^<https:\/\/w3id.org\/vcf-rdfizer\/vocab#Null>\1/g' "$nt_path" > "$tmp_path"
+  mv "$tmp_path" "$nt_path"
+}
+
 
 # Convert elapsed clock text from `time` output to numeric seconds.
 elapsed_to_seconds() {
@@ -266,6 +277,17 @@ else
   OUTPUT_PATH="$OUT_DIR/$OUT_NAME"
 fi
 
+# Apply ontology-compliant null datatype annotation to produced RDF files.
+if [[ -f "$OUTPUT_PATH" ]]; then
+  annotate_null_literals_nt "$OUTPUT_PATH"
+elif [[ -d "$OUTPUT_PATH" ]]; then
+  shopt -s nullglob
+  for RDF_NT in "$OUTPUT_PATH"/*.nt; do
+    annotate_null_literals_nt "$RDF_NT"
+  done
+  shopt -u nullglob
+fi
+
 OUT_SIZE=$(stat_size "$OUTPUT_PATH")
 TRIPLES_JSON=$(count_triples_json "$OUTPUT_PATH")
 

test/test_run_conversion_unit.py

@@ -532,6 +532,61 @@ def test_run_conversion_aggregate_skips_duplicate_part_payloads(self):
             self.assertEqual(lines.count("<dup> <p> <o> ."), 1)
             self.assertEqual(lines.count("<uniq> <p> <o> ."), 1)
 
+    def test_run_conversion_rewrites_dot_literal_to_typed_null(self):
+        """Plain literal dot is normalized to ontology null typed literal in RDF output."""
+        with tempfile.TemporaryDirectory() as td:
+            tmp_path = Path(td)
+            fake_bin = tmp_path / "bin"
+            fake_bin.mkdir()
+            make_executable(
+                fake_bin / "java",
+                """#!/usr/bin/env bash
+set -euo pipefail
+if [[ "${1:-}" == "-version" ]]; then
+  echo 'openjdk version "11.0.0"' >&2
+  exit 0
+fi
+out=""
+while [[ $# -gt 0 ]]; do
+  if [[ "$1" == "-o" ]]; then
+    out="$2"
+    shift 2
+    continue
+  fi
+  shift
+done
+mkdir -p "$out"
+printf '<s> <p> "." .\\n' > "$out/part-000"
+""",
+            )
+
+            out_dir = tmp_path / "out"
+            metrics_dir = tmp_path / "metrics"
+            rules = tmp_path / "rules.ttl"
+            rules.write_text("@prefix ex: <http://example.org/> .\n")
+            vcf = tmp_path / "input.vcf"
+            vcf.write_text("##fileformat=VCFv4.2\n#CHROM\tPOS\n1\t5\n")
+
+            env = env_with_path(fake_bin)
+            env.update(
+                {
+                    "JAR": "fake.jar",
+                    "IN": str(rules),
+                    "IN_VCF": str(vcf),
+                    "OUT_DIR": str(out_dir),
+                    "OUT_NAME": "rdf",
+                    "LOGDIR": str(metrics_dir),
+                    "RUN_ID": "run-null-dot",
+                    "TIMESTAMP": "2026-01-01T00:00:00",
+                }
+            )
+
+            result = subprocess.run(["bash", str(SCRIPT)], env=env, capture_output=True, text=True)
+            self.assertEqual(result.returncode, 0, msg=result.stderr)
+            merged_nt = out_dir / "rdf" / "rdf.nt"
+            text = merged_nt.read_text()
+            self.assertIn('"."^^<https://w3id.org/vcf-rdfizer/vocab#Null> .', text)
+
 
 if __name__ == "__main__":
     unittest.main()

test/test_vcf_rdfizer_unit.py

@@ -1100,6 +1100,8 @@ def fake_run(cmd, cwd=None, env=None):
             self.assertEqual(len(commands), 2)
             self.assertIn("/opt/vcf-rdfizer/vcf_as_tsv.sh", commands[0])
             self.assertIn("/opt/vcf-rdfizer/run_conversion.sh", commands[1])
+            self.assertTrue(any(str(arg).endswith(":/data/in:ro") for arg in commands[1]))
+            self.assertTrue(any(str(arg).startswith("IN_VCF=/data/in/") for arg in commands[1]))
 
     def test_main_full_mode_batch_layout_compresses_each_rml_part(self):
         """Batch layout compresses each part and prints one consolidated size summary."""

vcf_rdfizer.py

@@ -2118,6 +2118,8 @@ def fail_current(stage: str, message: str):
         run_cmd = [
             *docker_run_base(),
             "-v",
+            f"{str(input_mount_dir)}:/data/in:ro",
+            "-v",
             f"{str(generated_rules_dir)}:/data/rules:ro",
             "-v",
             f"{str(tsv_dir)}:/data/tsv:ro",