From fa94fbc6a551fdb52a750626752e372a8d58d1ce Mon Sep 17 00:00:00 2001
From: Holger Frey <frey@imtek.de>
Date: Wed, 17 Feb 2021 10:08:15 +0100
Subject: [PATCH] added silenced parsing of data files

---
 sensospot_data/parser.py | 17 ++++++++++++++---
 tests/test_parser.py     | 35 ++++++++++++++++++++++++++++++++++-
 2 files changed, 48 insertions(+), 4 deletions(-)

diff --git a/sensospot_data/parser.py b/sensospot_data/parser.py
index 75e36d2..fc1de19 100755
--- a/sensospot_data/parser.py
+++ b/sensospot_data/parser.py
@@ -69,8 +69,8 @@ def _cleanup_data_columns(data_frame):
 
 
 def parse_file(data_file):
-    """ parses one data file and adds metadata to result 
-    
+    """parses one data file and adds metadata to result
+
     will race a ValueError, if metadata could not be extracted
     """
     measurement_info = _extract_measurement_info(Path(data_file))
@@ -81,11 +81,22 @@ def parse_file(data_file):
     return _cleanup_data_columns(data_frame)
 
 
+def _silenced_parse_file(data_file):
+    """parses one data file and adds metadata
+
+    returns data frame or None on ValueError
+    """
+    try:
+        return parse_file(data_file)
+    except ValueError:
+        return None
+
+
 def parse_multiple_files(file_list):
     """ parses a list of file paths to one combined dataframe """
     if not file_list:
         raise ValueError("Empty file list provided")
-    collection = (parse_file(path) for path in file_list)
+    collection = (_silenced_parse_file(path) for path in file_list)
     filtered = (frame for frame in collection if frame is not None)
     data_frame = next(filtered)
     for next_frame in filtered:
diff --git a/tests/test_parser.py b/tests/test_parser.py
index a568ec6..3c97fe9 100644
--- a/tests/test_parser.py
+++ b/tests/test_parser.py
@@ -179,6 +179,39 @@ def test_parse_file(example_file):
     assert result["Exposure.Id"][0] == 1
 
 
+def test_parse_file_raises_error(example_dir):
+    from sensospot_data.parser import parse_file
+
+    csv_file = (
+        example_dir / EXAMPLE_DIR_WITH_PARAMS / "should_raise_value_error.csv"
+    )
+
+    with pytest.raises(ValueError):
+        parse_file(csv_file)
+
+
+def test_silenced_parse_file_returns_data_frame(example_file):
+    from sensospot_data.parser import _silenced_parse_file
+
+    result = _silenced_parse_file(example_file)
+
+    assert result["Well.Row"][0] == "A"
+    assert result["Well.Column"][0] == 1
+    assert result["Exposure.Id"][0] == 1
+
+
+def test_silenced_parse_file_returns_none_on_error(example_dir):
+    from sensospot_data.parser import _silenced_parse_file
+
+    csv_file = (
+        example_dir / EXAMPLE_DIR_WITH_PARAMS / "should_raise_value_error.csv"
+    )
+
+    result = _silenced_parse_file(csv_file)
+
+    assert result is None
+
+
 @pytest.mark.parametrize(
     "file_list",
     [
@@ -225,7 +258,7 @@ def test_list_csv_files(example_dir):
 
     result = list(list_csv_files(example_dir / EXAMPLE_DIR_WITH_PARAMS))
 
-    assert len(result) == 36 * 3
+    assert len(result) == (36 * 3) + 1  # 36 wells, 3 exposure + one error file
     assert all(str(item).endswith(".csv") for item in result)
     assert all(not item.stem.startswith(".") for item in result)