Revised the imports for MATLAB and added additional allowed font extensions

hed/tools/analysis/annotation_util.py

@@ -4,8 +4,10 @@
 import re
 
 import pandas as pd
+from pandas import DataFrame, Series
 from hed.models.sidecar import Sidecar
 from hed.models.tabular_input import TabularInput
+
 from hed.errors.exceptions import HedFileError
 from hed.models import df_util
 from hed.tools.bids.bids_dataset import BidsDataset
@@ -63,21 +65,6 @@ def df_to_hed(dataframe, description_tag=True):
     return hed_dict
 
 
-def series_to_factor(series):
-    """Convert a series to an integer factor list.
-
-    Parameters:
-        series (Series) - Series to be converted to a list.
-
-    Returns:
-        list - contains 0's and 1's, empty, 'n/a' and np.NAN are converted to 0.
-    """
-    replaced = series.replace('n/a', False)
-    filled = replaced.fillna(False)
-    bool_list = filled.astype(bool).tolist()
-    return [int(value) for value in bool_list]
-
-
 def extract_tags(hed_string, search_tag):
     """ Extract all instances of specified tag from a tag_string.
 
@@ -129,6 +116,7 @@ def generate_sidecar_entry(column_name, column_values=None):
         sidecar_entry["HED"] = hed
     return sidecar_entry
 
+
 def get_bids_dataset(data_root):
     """ Return a BIDS dataset object given a path to a dataset root.
 
@@ -141,6 +129,7 @@ def get_bids_dataset(data_root):
     """
     return BidsDataset(data_root)
 
+
 def hed_to_df(sidecar_dict, col_names=None):
     """ Return a 4-column dataframe of HED portions of sidecar.
 
@@ -202,6 +191,21 @@ def merge_hed_dict(sidecar_dict, hed_dict):
             sidecar_dict[key]['Levels'] = value_dict['Levels']
 
 
+def series_to_factor(series):
+    """Convert a series to an integer factor list.
+
+    Parameters:
+        series (Series) - Series to be converted to a list.
+
+    Returns:
+        list - contains 0's and 1's, empty, 'n/a' and np.NAN are converted to 0.
+    """
+    replaced = series.replace('n/a', False)
+    filled = replaced.fillna(False)
+    bool_list = filled.astype(bool).tolist()
+    return [int(value) for value in bool_list]
+
+
 def str_to_tabular(tsv_str, sidecar=None):
     """ Return a TabularInput a tsv string.
 
@@ -223,9 +227,11 @@ def strs_to_sidecar(sidecar_strings):
          sidecar_strings (string or list):  String or strings representing sidecars.
 
      Returns:
-         Sidecar:  the merged sidecar from the list.
+         Sidecar or None:  the merged sidecar from the list.
      """
 
+    if not sidecar_strings:
+        return None
     if not isinstance(sidecar_strings, list):
         sidecar_strings = [sidecar_strings]
     if sidecar_strings:
@@ -236,6 +242,7 @@ def strs_to_sidecar(sidecar_strings):
     else:
         return None
 
+
 def to_factor(data, column=None):
     """Convert data to an integer factor list.
 
@@ -246,11 +253,11 @@ def to_factor(data, column=None):
     Returns:
         list - contains 0's and 1's, empty, 'n/a' and np.NAN are converted to 0.
     """
-    if isinstance(data, pd.Series):
+    if isinstance(data, Series):
         series = data
-    elif isinstance(data, pd.DataFrame) and column:
+    elif isinstance(data, DataFrame) and column:
         series = data[column]
-    elif isinstance(data, pd.DataFrame):
+    elif isinstance(data, DataFrame):
         series = data.iloc[:, 0]
     else:
         raise HedFileError("CannotConvertToFactor",

hed/tools/visualization/tag_word_cloud.py

@@ -46,7 +46,7 @@ def create_wordcloud(word_dict, mask_path=None, background_color=None, width=400
     kwargs.setdefault('min_font_size', 8)
     if 'font_path' not in kwargs:
         kwargs['font_path'] = None
-    elif kwargs['font_path'] and not kwargs['font_path'].lower().endswith((".ttf", ".otf")):
+    elif kwargs['font_path'] and not kwargs['font_path'].lower().endswith((".ttf", ".otf", ".ttc")):
         raise HedFileError("InvalidFontPath", f"Font {kwargs['font_path']} not valid on this system", "")
 
     wc = WordCloud(background_color=background_color, mask=mask_image,

tests/tools/analysis/test_annotation_util.py

@@ -9,6 +9,7 @@
 from hed.models.sidecar import Sidecar
 from hed.models.hed_string import HedString
 from hed.models.tabular_input import TabularInput
+
 from hed.tools.analysis import annotation_util
 from hed.tools.analysis.tabular_summary import TabularSummary
 from hed.tools.util import io_util
@@ -196,17 +197,6 @@ def test_generate_sidecar_entry(self):
         self.assertIsInstance(entry2['HED'], str,
                               "generate_sidecar_entry HED entry should be str when no column values")
 
-    def test_series_to_factor(self):
-        series1 = Series([1.0, 2.0, 3.0, 4.0])
-        factor1 = annotation_util.series_to_factor(series1)
-        self.assertEqual(len(series1), len(factor1))
-        self.assertEqual(sum(factor1), len(factor1))
-        series2 = Series(['a', '', None, np.NAN, 'n/a'])
-        factor2 = annotation_util.series_to_factor(series2)
-        self.assertEqual(len(series2), len(factor2))
-        self.assertEqual(sum(factor2), 1)
-
-
     def test_generate_sidecar_entry_non_letters(self):
         entry1 = annotation_util.generate_sidecar_entry('my !#$-123_10', 
                                                         column_values=['apple 1', '@banana', 'grape%cherry&'])
@@ -294,7 +284,7 @@ def test_merge_hed_dict_full(self):
         skip_columns = ["onset", "duration", "sample", "trial", "response_time"]
         value_columns = ["rep_lag", "stim_file", "value"]
         event_files = io_util.get_file_list(self.bids_root_path, extensions=[".tsv"], name_suffix="_events",
-                                    exclude_dirs=exclude_dirs)
+                                            exclude_dirs=exclude_dirs)
         value_sum = TabularSummary(value_cols=value_columns, skip_cols=skip_columns)
         value_sum.update(event_files)
         sidecar_template = value_sum.extract_sidecar_template()
@@ -305,6 +295,37 @@ def test_merge_hed_dict_full(self):
         annotation_util.merge_hed_dict(example_sidecar, spreadsheet_sidecar)
         self.assertEqual(6, len(example_sidecar), 'merge_hed_dict merges with the correct length')
 
+        def test_to_factor(self):
+            series1 = Series([1.0, 2.0, 3.0, 4.0])
+            factor1 = annotation_util.to_factor(series1)
+            self.assertEqual(len(series1), len(factor1))
+            self.assertEqual(sum(factor1), len(factor1))
+            series2 = Series(['a', '', None, np.NAN, 'n/a'])
+            factor2 = annotation_util.to_factor(series2)
+            self.assertEqual(len(series2), len(factor2))
+            self.assertEqual(sum(factor2), 1)
+            data = {
+                'Name': ['Alice', '', 'n/a', 1.0],  # Contains a space
+                'Age': [25, np.NaN, 35, 0]
+            }
+            df = DataFrame(data)
+            factor3 = annotation_util.to_factor(df, column='Name')
+            self.assertEqual(sum(factor3), 2)
+            factor4 = annotation_util.to_factor(df)
+            self.assertEqual(sum(factor4), 2)
+            with self.assertRaises(HedFileError) as context5:
+                annotation_util.to_factor(data)
+
+    def test_series_to_factor(self):
+        series1 = Series([1.0, 2.0, 3.0, 4.0])
+        factor1 = annotation_util.series_to_factor(series1)
+        self.assertEqual(len(series1), len(factor1))
+        self.assertEqual(sum(factor1), len(factor1))
+        series2 = Series(['a', '', None, np.NAN, 'n/a'])
+        factor2 = annotation_util.series_to_factor(series2)
+        self.assertEqual(len(series2), len(factor2))
+        self.assertEqual(sum(factor2), 1)
+
     def test_strs_to_sidecar(self):
         with open(self.json_path, 'r') as fp:
             sidecar_dict = json.load(fp)
@@ -336,7 +357,6 @@ def test_to_strlist(self):
         self.assertEqual(str_list2[0], 'Red,Sensory-event')
         self.assertEqual(str_list2[2], '')
 
-
     def test_flatten_cat_col(self):
         col1 = self.sidecar2c["a"]
         col2 = self.sidecar2c["b"]
@@ -358,13 +378,14 @@ def test_flatten_cat_col(self):
                          "_flatten_cat_col should use the Description tag if available")
 
     def test_flatten_cat_col_only_description(self):
-        keys, values, descriptions, tags = annotation_util._flatten_cat_col("event_type",
-                                                            {"HED": {"code1": "Description/Code 1 here."}})
+        keys, values, descriptions, tags = \
+            annotation_util._flatten_cat_col("event_type", {"HED": {"code1": "Description/Code 1 here."}})
         self.assertIsInstance(tags, list)
         self.assertEqual(tags[0], 'n/a')
 
     def test_flatten_val_col_only_description(self):
-        keys, values, descriptions, tags = annotation_util._flatten_val_col("response", {"HED": "Description/Code 1 here."})
+        keys, values, descriptions, tags = annotation_util._flatten_val_col("response",
+                                                                            {"HED": "Description/Code 1 here."})
         self.assertEqual(descriptions[0], 'Code 1 here.')
         self.assertFalse(tags[0])