Merge branch 'release/0.3.5'

setup.py

@@ -43,7 +43,7 @@
     # For a discussion on single-sourcing the version across setup.py and the
     # project code, see
     # https://packaging.python.org/en/latest/single_source_version.html
-    version='0.3.4',  # Required
+    version='0.3.5',  # Required
 
     # This is a one-line description or tagline of what your project does. This
     # corresponds to the "Summary" metadata field:

tests/test_parsers.py

@@ -160,6 +160,23 @@ def test_parses_freq_from_freq(self):
         p = special_parser(line)
         assert p.alt_allele_freq == 0.25, "Parses frequency as is"
 
+    def test_pathological_chromosome_fails(self):
+        # Regression test for a badly specified user input file: restrict chromosomes to a whitelist, because sometimes
+        #   people manage to slip non-categorical fields to tabix (and then tabix eats all the RAM on the server)
+        line = '1\tchr1:722408:G:C\t722408\tc\tg\t0.9298\tchr1:722408'
+        parser_options = {
+            "alt_col": 5, "pos_col": 3, "ref_col": 4, "beta_col": None, "chrom_col": 2, "pvalue_col": 6,
+            "stderr_beta_col": None, "is_neg_log_pvalue": None
+        }
+
+        parser = parsers.GenericGwasLineParser(**parser_options)
+        with pytest.raises(
+                exceptions.LineParseException,
+                match="Chromosome 1:722408:G:C is not a valid option. Must be one of: '1 2 3 4 5 6 7 8 9 10 11 12 13 "
+                      "14 15 16 17 18 19 20 21 22 23 24 25 M MT X Y'"
+        ):
+            parser(line)
+
 
 class TestStandardGwasParser:
     def test_parses_locuszoom_standard_format(self, standard_gwas_parser):

zorp/__init__.py

@@ -1,6 +1,6 @@
 from distutils.version import LooseVersion
 
-__version__ = '0.3.4'
+__version__ = '0.3.5'
 __version_info__ = tuple(LooseVersion(__version__).version)
 
 __all__ = [

zorp/parser_utils.py

@@ -119,3 +119,10 @@ def human_to_zero(value):
         return value
     else:
         return value - 1
+
+
+def natural_sort(items: ty.Iterable):
+    """Natural sort a list of strings. Used for human-friendly error messages, eg, from a `set` of allowed strings"""
+    convert = lambda text: int(text) if text.isdigit() else text.lower()  # noqa: E731
+    alphanum_key = lambda key: [convert(c) for c in re.split('([0-9]+)', key)]  # noqa: E731
+    return sorted(items, key=alphanum_key)

zorp/parsers.py

@@ -13,6 +13,15 @@
 from .const import MISSING_VALUES
 from . import exceptions, parser_utils as utils
 
+# Whitelist of allowed chromosomes. It's ok to add more values, as long as we have some kind of whitelist.
+#  The generic parser uses these as a safeguard, because when people slip a non-categorical value into the chrom field,
+#  tabix uses all the RAM on the system and then crashes horribly.
+ALLOWED_CHROMS = frozenset({
+    '1', '2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12', '13', '14', '15', '16', '17', '18', '19', '20',
+    '21', '22', '23', '24', '25',
+    'X', 'Y', 'M', 'MT'
+})
+
 
 class BasicVariant:
     """
@@ -160,6 +169,11 @@ def inner(line):
 
             chrom = chrom.upper()
 
+            if chrom not in ALLOWED_CHROMS:
+                options = ' '.join(utils.natural_sort(ALLOWED_CHROMS))
+                raise exceptions.LineParseException(
+                    f"Chromosome {chrom} is not a valid option. Must be one of: '{options}'")
+
             # Explicit columns will override a value from the marker, by design
             if _ref_col is not None:
                 ref = fields[_ref_col]