seed.lib.mappings package¶
Submodules¶
seed.lib.mappings.mapper module¶
-
seed.lib.mappings.mapper.create_column_regexes(raw_columns)¶ Take the columns in the format below and sanitize the keys and add in the regex.
Parameters: raw_data – list of strings (columns names from imported file) Returns: list of dict
-
seed.lib.mappings.mapper.get_pm_mapping(raw_columns, mapping_data=None, resolve_duplicates=True)¶ Create and return Portfolio Manager (PM) mapping for a given version of PM and the given list of column names.
The method will take the raw_columns (from the CSV/XLSX file) and attempt to normalize the column names so that they can be mapped to the data in the pm-mapping.json[‘from_field’].
seed.lib.mappings.mapping_columns module¶
-
class
seed.lib.mappings.mapping_columns.MappingColumns(raw_columns, dest_columns, previous_mapping=None, map_args=None, default_mappings=None, threshold=0)¶ Bases:
objectThis class handles the probabilistic mapping of unknown columns to defined fields. This is mainly used in the build_column_mapping API endpoint.
-
add_mappings(raw_column, mappings, previous_mapping=False)¶ Add mappings to the data structure for later processing.
Parameters: - raw_column – list of strings
- mappings – list of tuples of potential mappings and confidences
- previous_mapping – boolean, if true these these mappings will take precedence
Returns: Bool, whether or not the mapping was added
-
apply_threshold(threshold)¶ Remove mapping suggestions that do not meet the defined threshold
This method is forced as part of the workflow for now, but could easily be made as a separate call.
Parameters: threshold – int, min value to be greater than or equal to. Returns: None
-
duplicates¶ Check for duplicate initial mapping results.
Returns: List of raw col
-
final_mappings¶ Return the final mappings in a format that can be used downstream from this method {
“raw_column_1”: (‘table’, ‘db_column_1’, confidence), “raw_column_2”: (‘table’, ‘db_column_1’, confidence),}
-
first_suggested_mapping(raw_column)¶ Grab the first suggested mapping for a raw column
Parameters: raw_column – String Returns: tuple of the mapping (‘table’, ‘field’, confidence), or ()
-
resolve_duplicate(dup_map_field, raw_columns)¶ Parameters: - dup_map_field – String, name of the field that is a duplicate
- columns – list, raw columns that mapped to the same result
Returns: None
-
set_initial_mapping_cmp(raw_column)¶ Set the initial_mapping_cmp helper item in the self.data hash. This is used to detect if there are any duplicates.
Parameters: raw_column – String, name of the raw column to set the initial_mapping_cmp Returns: None
-
-
seed.lib.mappings.mapping_columns.sort_duplicates(a, b)¶ Custom sort for the duplicate hash to decide which raw column will get the mapping suggestion