Merge pull request #669 from nextstrain/fix-numeric-names

ensure columns strain and name in metadata are parsed as string
nextstrain · Feb 12, 2021 · fa5fa43 · fa5fa43
2 parents b2a313c + f266805
commit fa5fa43
Show file tree

Hide file tree

Showing 2 changed files with 13 additions and 0 deletions.
diff --git a/augur/util_support/metadata_file.py b/augur/util_support/metadata_file.py
@@ -93,4 +93,5 @@ def parse_file(self):
             sep=None,  # csv.Sniffer will automatically detect sep
             engine="python",
             skipinitialspace=True,
+            dtype={"strain":"string", "name":"string"}
         ).fillna("")
diff --git a/tests/util_support/test_metadata_file.py b/tests/util_support/test_metadata_file.py
@@ -117,3 +117,15 @@ def test_metadata_delimiter_autodetect(self, tmpdir, prepare_file):
             "strainB": {"strain": "strainB", "location": "nevada", "quality": "good"},
         }
         assert list(columns) == ["strain", "location", "quality"]
+
+    def test_metadata_strain_type(self, tmpdir, prepare_file):
+        prepare_file(
+            """
+            strain\tlocation
+            1\tWashington
+            2\tOregon
+            """
+        )
+
+        records, columns = MetadataFile(f"{tmpdir}/metadata.txt").read()
+        assert "1" in records