fixups

countess/plugins/hgvs_parser.py

@@ -22,11 +22,7 @@ class HgvsParserPlugin(PandasTransformDictToDictPlugin):
  multi = BooleanParam("Multiple rows", False)
 
  def process_dict(self, data: dict):
- try:
- value = data[str(self.column)]
- except KeyError:
- return {}
-
+ value = data.get(str(self.column))
  if type(value) is not str:
  return {}
 
@@ -39,7 +35,7 @@ def process_dict(self, data: dict):
  guides += self.guides_str.value.split(";")
 
  if m := re.match(r"(?:([\w.]+):)?([ncg]\.)(.*)", value):
- output["reference"] = m.group(1) or ''
+ output["reference"] = m.group(1) or ""
  output["prefix"] = m.group(2)
  value = m.group(3)
 
@@ -55,18 +51,18 @@ def process_dict(self, data: dict):
 
  max_variations = int(self.max_var)
  variations = [v for v in variations if v not in guides]
- if len(variations) > max_variations:
- return {}
 
  output_vars: list[Optional[str]] = [None] * max_variations
  output_locs: list[Optional[str]] = [None] * max_variations
- for n, v in enumerate(variations):
- if self.split:
- if m := re.match(r"([\d_]+)(.*)", v):
- output_locs[n] = m.group(1)
- output_vars[n] = m.group(2)
- continue
- output_vars[n] = v
+
+ if len(variations) <= max_variations:
+ for n, v in enumerate(variations):
+ if self.split:
+ if m := re.match(r"([\d_]+)(.*)", v):
+ output_locs[n] = m.group(1)
+ output_vars[n] = m.group(2)
+ continue
+ output_vars[n] = v
 
  if self.multi:
  output["var"] = output_vars

countess/utils/parallel.py

@@ -1,9 +1,10 @@
 import gc
 import logging
+import threading
+import time
 from multiprocessing import Process, Queue, Value
 from os import cpu_count, getpid
-from queue import Empty, Full
-import time
+from queue import Empty
 from typing import Callable, Iterable
 
 try:
@@ -73,9 +74,9 @@ def multiprocess_map(function, values, *args, **kwargs):
  output_queue: Queue = Queue(maxsize=3)
 
  def __process(): # pragma: no cover
- # this is run in a pool of `nproc` processes to handle resource-intensive
- # processes which don't play nicely with the GIL.
- # XXX Coverage doesn't seem to understand this so we exclude it from coverage.
+  # this is run in a pool of `nproc` processes to handle resource-intensive
+  # processes which don't play nicely with the GIL.
+  # XXX Coverage doesn't seem to understand this so we exclude it from coverage.
 
  for data_in in input_queue:
  for data_out in function(data_in, *args, **(kwargs or {})):

tests/plugins/test_hgvs_parser.py

@@ -84,18 +84,16 @@ def test_hgvs_parser_bad():
 
  df = plugin.process_dataframe(df2)
 
- print(df)
- assert all(np.isnan(df["var_1"]))
- #assert np.isnan(df["var_1"].iloc[0])
- #assert np.isnan(df["var_1"].iloc[1])
- #assert np.isnan(df["var_1"].iloc[2])
+ assert np.isnan(df["var_1"].iloc[0])
+ assert np.isnan(df["var_1"].iloc[1])
+ assert np.isnan(df["var_1"].iloc[2])
 
 
 def test_hgvs_parser_very_bad():
  plugin = HgvsParserPlugin()
  plugin.set_parameter("column", "hgvs")
 
- dfi = pd.DataFrame([{'a': 1}])
+ dfi = pd.DataFrame([{"a": 1}])
  dfo = plugin.process_dataframe(dfi)
 
  assert all(dfo == dfi)

tests/test_parameters.py

@@ -296,5 +296,5 @@ def test_arrayparam_minmax():
  assert len(ap) == 2
 
  # FIX minimum and maximum constraints!
- #ap.del_subparam(ap[1])
- #assert len(ap) == 2
+ # ap.del_subparam(ap[1])
+ # assert len(ap) == 2

tests/test_plugins.py

@@ -4,13 +4,8 @@
 import pandas as pd
 import pytest
 
-<<<<<<< HEAD
-import countess
-from countess.core.plugins import BasePlugin, FileInputPlugin, PandasProductPlugin, get_plugin_classes
-=======
 from countess.core.parameters import ColumnChoiceParam, StringParam
 from countess.core.plugins import (
- BasePlugin,
  FileInputPlugin,
  PandasConcatProcessPlugin,
  PandasProductPlugin,
@@ -19,7 +14,6 @@
  PandasTransformSingleToSinglePlugin,
  get_plugin_classes,
 )
->>>>>>> bf58ca03bb7b84f59ec878258834878f375eb369
 
 empty_entry_points_dict = {"countess_plugins": []}
 
@@ -122,7 +116,7 @@ class FIP(FileInputPlugin):
  def num_files(self):
  return 3
 
- def load_file(self, file_number, row_limit):
+ def load_file(self, file_number, row_limit=None):
  if row_limit is None:
  row_limit = 1000000
  return [f"hello{file_number}"] * row_limit