Merge pull request #668 from tomkinsc/ct-priorities-parsing-bugfix

fix parsing of priorities tsv file to allow spaces in sequence IDs

augur/filter.py

@@ -60,7 +60,7 @@ def read_priority_scores(fname):
         with open(fname, encoding='utf-8') as pfile:
             return defaultdict(float, {
                 elems[0]: float(elems[1])
-                for elems in (line.strip().split() for line in pfile.readlines())
+                for elems in (line.strip().split('\t') if '\t' in line else line.strip().split() for line in pfile.readlines())
             })
     except Exception as e:
         print(f"ERROR: missing or malformed priority scores file {fname}", file=sys.stderr)

tests/test_filter.py

@@ -57,6 +57,12 @@ def mock_run_shell_command(mocker):
     mocker.patch("augur.filter.run_shell_command")
 
 
+@pytest.fixture
+def mock_priorities_file_valid_with_spaces_and_tabs(mocker):
+    mocker.patch(
+        "builtins.open", mocker.mock_open(read_data="strain 1\t5\nstrain 2\t6\nstrain 3\t8\n")
+    )
+
 class TestFilter:
     def test_read_vcf_compressed(self):
         seq_keep, all_seq = augur.filter.read_vcf(
@@ -89,6 +95,14 @@ def test_read_priority_scores_malformed(self, mock_priorities_file_malformed):
             # builtins.open is stubbed, but we need a valid file to satisfy the existence check
             augur.filter.read_priority_scores("tests/builds/tb/data/lee_2015.vcf")
 
+    def test_read_priority_scores_valid_with_spaces_and_tabs(self, mock_priorities_file_valid_with_spaces_and_tabs):
+        # builtins.open is stubbed, but we need a valid file to satisfy the existence check
+        priorities = augur.filter.read_priority_scores(
+            "tests/builds/tb/data/lee_2015.vcf"
+        )
+
+        assert priorities == {"strain 1": 5, "strain 2": 6, "strain 3": 8}
+
     def test_read_priority_scores_does_not_exist(self):
         with pytest.raises(FileNotFoundError):
             augur.filter.read_priority_scores("/does/not/exist.txt")