Updated exception handling for invalid tabular and json files

CHANGELOG.md

@@ -1,3 +1,11 @@
+Release 0.3.1 July 3, 2023
+- Pinned the version of the pydantic and inflect libraries due to inflict.
+- Reorganized JSON output of remodeling summaries so that all of consistent form.
+- Fixed summarize_hed_tags_op so that tags were correctly categorized for output.
+- Minor refactoring to reduce code complexity.
+- BaseInput and Sidecar now raise HedFileError if input could not be read.
+
+
 Release 0.3.0 June 20, 2023
 - Introduction of partnered schema.
 - Improved error handling for schema validation.

LICENSE

@@ -1,4 +1,4 @@
-MIT License
+The MIT License (MIT)
 
 Copyright (c) 2020+ HED Standard Working Group
 

hed/errors/exceptions.py

@@ -11,7 +11,7 @@ class HedExceptions:
     INVALID_EXTENSION = 'invalidExtension'
 
     INVALID_DATAFRAME = 'INVALID_DATAFRAME'
-
+    INVALID_FILE_FORMAT = 'INVALID_FILE_FORMAT'
     # These are actual schema issues, not that the file cannot be found or parsed
     SCHEMA_HEADER_MISSING = 'HED_SCHEMA_HEADER_INVALID'
     HED_SCHEMA_HEADER_INVALID = 'HED_SCHEMA_HEADER_INVALID'

hed/models/base_input.py

@@ -45,6 +45,8 @@ def __init__(self, file, file_type=None, worksheet_name=None, has_column_names=T
             - A duplicate or empty column name appears
             - Cannot open the indicated file
             - The specified worksheet name does not exist
+            - If the sidecar file or tabular file had invalid format and could not be read.
+
          """
         if mapper is None:
             mapper = ColumnMapper()
@@ -77,7 +79,7 @@ def __init__(self, file, file_type=None, worksheet_name=None, has_column_names=T
                 self._dataframe = pandas.read_csv(file, delimiter='\t', header=pandas_header,
                                                   dtype=str, keep_default_na=True, na_values=None)
             except Exception as e:
-                raise HedFileError(HedExceptions.GENERIC_ERROR, str(e), self.name) from e
+                raise HedFileError(HedExceptions.INVALID_FILE_FORMAT, str(e), self.name) from e
             # Convert nan values to a known value
             self._dataframe = self._dataframe.fillna("n/a")
         elif input_type in self.EXCEL_EXTENSION:
@@ -96,7 +98,7 @@ def __init__(self, file, file_type=None, worksheet_name=None, has_column_names=T
         # todo: Can we get rid of this behavior now that we're using pandas?
         column_issues = ColumnMapper.check_for_blank_names(self.columns, allow_blank_names=allow_blank_names)
         if column_issues:
-            raise HedFileError(HedExceptions.BAD_COLUMN_NAMES, "Duplicate or blank columns found.  See issues.",
+            raise HedFileError(HedExceptions.BAD_COLUMN_NAMES, "Duplicate or blank columns found. See issues.",
                                self.name, issues=column_issues)
 
         self.reset_mapper(mapper)
@@ -287,7 +289,7 @@ def set_cell(self, row_number, column_number, new_string_obj, tag_form="short_ta
 
         Notes:
              Any attribute of a HedTag that returns a string is a valid value of tag_form.
-             
+
         :raises ValueError:
             - There is not a loaded dataframe
 

hed/models/sidecar.py

@@ -127,15 +127,13 @@ def load_sidecar_file(self, file):
         if not file:
             return {}
         elif isinstance(file, str):
+            if not self.name:
+                self.name = file
             try:
                 with open(file, "r") as fp:
-                    if not self.name:
-                        self.name = file
                     return self._load_json_file(fp)
             except OSError as e:
                 raise HedFileError(HedExceptions.FILE_NOT_FOUND, e.strerror, file) from e
-            except TypeError as e:
-                raise HedFileError(HedExceptions.FILE_NOT_FOUND, str(e), file) from e
         else:
             return self._load_json_file(file)