csvw-functions-extra 0.0.0

Python functions for CSVW files providing extra functionality beyond the CSVW standards

csvw_functions_extra

Python functions for CSVW files providing extra functionality beyond the CSVW standards

Installation

pip install git+https://github.com/stevenkfirth/csvw_functions_extra

The python package csvw_functions will also need to be installed.

API

get_normalized_metadata_table_group_dict

Description: Returns a normalized version of a CSVW metadata file.

csvw_functions_extra.get_normalized_metadata_table_group_dict(
        metadata_document_location
        )

• metadata_document_location (str): The filepath or url of the CSVW metadata file containing a Table Group object.

Returns (dict): A dictionary of the normalized CSVW Table Group object.

get_available_csv_file_names

Description: Returns the CSV file names of all tables in a CSVW metadata file.

csvw_extra_functions.get_available_csv_file_names(
        metadata_document_location
)

• metadata_document_location (str): The filepath or url of the CSVW metadata file containing a Table Group object.

Returns (list): A list of the https://purl.org/berg/csvw_functions_extra/vocab/csv_file_name value in each table.

download_table_group

Description: Reads a CSVW metadata file and downloads the CSV files from remote locations. This makes use of the https://purl.org/berg/csvw_functions_extra vocabulary.

csvw_functions_extra.download_table_group(
        metadata_document_location,
        data_folder,
        csv_file_names=None,  
        overwrite_existing_files=False,
        verbose=False
        )

For each table in the TableGroup object, the method is:

i) For individual CSV files:

1. The CSV file is downloaded using the url in https://purl.org/berg/csvw_functions_extra/vocab/csv_download_url.
2. The CSV is saved in the data_folder using the filename in https://purl.org/berg/csvw_functions_extra/vocab/csv_file_name

ii) For ZIP files:

1. The ZIP file is downloaded using the url in https://purl.org/berg/csvw_functions_extra/vocab/zip_download_url.
2. The ZIP is saved in the data_folder using the filename in https://purl.org/berg/csvw_functions_extra/vocab/zip_file_name
3. The CSV file is extracted from the ZIP file using the path in https://purl.org/berg/csvw_functions_extra/vocab/csv_zip_extract_path.
4. The CSV is saved in the data_folder using the filename in https://purl.org/berg/csvw_functions_extra/vocab/csv_file_name

iii) If an associated metadata file is present (this is separate to the CSVW metadata file):

1. This is downloaded using the url in https://purl.org/berg/csvw_functions_extra/vocab/metadata_download_url
2. If step 3 occurs, the associated metadata file is saved in the data_folder using the filename in https://purl.org/berg/csvw_functions_extra/vocab/csv_file_name with the additional suffix in https://purl.org/berg/csvw_functions_extra/vocab/metadata_file_suffix.

iv) For all cases:

1. A new version of the normalized CSVW metadata file is also saved in the data folder, with Table url values linking to the newly downloaded CSV files.

Arguments:

• metadata_document_location (str): The filepath or url of the CSVW metadata file containing a Table Group object.
• data_folder (str): The filepath of a local folder where the downloaded CSV data is to be saved to.
• csv_file_names (str or list): The csv_file_name values of the tables to be downloaded. If None then all tables are downloaded.
• overwrite_existing_files (bool): If True, then any existing CSV files in data_folder will be overwritten. If False, then no download occurs if there is an existing CSV file in data_folder.
• verbose (bool): If True, then this function prints intermediate variables and other useful information.

Returns (str): The local filename of the updated CSVW metadata file containing the new URLs for the newly downloaded tables.

get_metadata_table_group_dict

Description: Returns a CSVW metadata Table Group object.

csvw_functions_extra.get_metadata_table_group_dict(
        data_folder,
        metadata_filename
        )

Arguments:

• data_folder (str): The filepath of a local folder where the normalized CSVW metadata file is saved.
• metadata_filename (str): The filename of a CSVW metadata file which has been created by the download_table_group method and is located in the data folder.

Returns (dict): A dictionary of the CSVW Table Group object.

get_metadata_table_dict

Description: Returns a CSVW metadata Table object.

csvw_functions_extra.get_metadata_table_dict(
        sql_table_name,
        metadata_table_group_dict=None,
        data_folder=None,
        metadata_filename=None
        )

Arguments:

• sql_table_name (str): The https://purl.org/berg/csvw_functions_extra/vocab/sql_table_name value of the table.
• metadata_table_group_dict (dict): A dictionary of a metadata Table Group object, such as the return value of get_metadata_table_group_dict.
• data_folder (str): The filepath of a local folder where the normalized CSVW metadata file is saved.
• metadata_filename (str): The filename of a CSVW metadata file which has been created by the download_table_group method and is located in the data folder.

Returns (dict): A dictionary of the CSVW Table object.

Notes: If supplied the metadata_table_group_dict will be used to access the table object. If not supplied then the table object is accessed using the file located using data_folder and metadata_filename.

get_metadata_column_dict

Description: Returns a CSVW metadata Column object.

csvw_functions_extra.get_metadata_column_dict(
        column_name,
        sql_table_name,
        metadata_table_group_dict=None,
        data_folder=None,
        metadata_filename=None
        )

Arguments:

• column_name (str): The name value of a column in a CSVW TableSchema object.
• sql_table_name (str): The https://purl.org/berg/csvw_functions_extra/vocab/sql_table_name value of the table.
• metadata_table_group_dict (dict): A dictionary of a metadata Table Group object, such as the return value of get_metadata_table_group_dict.
• data_folder (str): The filepath of a local folder where the normalized CSVW metadata file is saved.
• metadata_filename (str): The filename of a CSVW metadata file which has been created by the download_table_group method and is located in the data folder.

Notes: If supplied the metadata_table_group_dict will be used to access the table object. If not supplied then the table object is accessed using the file located using data_folder and metadata_filename.

Returns (dict): A dictionary of the CSVW Column object.

get_metadata_sql_table_names

Description: Returns a list of the SQL table names in the CSVW metadata file.

csvw_functions_extra.get_metadata_sql_table_names(
        metadata_table_group_dict=None,
        data_folder=None,
        metadata_filename=None
        )

Arguments:

• metadata_table_group_dict (dict): A dictionary of a metadata Table Group object, such as the return value of get_metadata_table_group_dict.
• data_folder (str): The filepath of a local folder where the normalized CSVW metadata file is saved.
• metadata_filename (str): The filename of a CSVW metadata file which has been created by the download_table_group method and is located in the data folder.

Notes: If supplied the metadata_table_group_dict will be used to access the table object. If not supplied then the table object is accessed using the file located using data_folder and metadata_filename.

Returns (list): A list of the https://purl.org/berg/csvw_functions_extra/vocab/sql_table_name values of all tables in the CSVW metadata file.

get_metadata_columns_codes

Description: Returns lookup dictionaries for the lookup codes for one or more columns.

csvw_functions_extra.get_metadata_columns_codes(
        sql_table_name,
        column_names = None,
        metadata_table_group_dict = None,
        data_folder = None,
        metadata_filename=None
        )

Arguments:

• sql_table_name (str): The https://purl.org/berg/csvw_functions_extra/vocab/sql_table_name value of the table.
• column_names (str, list or None): The name value of one or more column in a CSVW TableSchema object. If None, then all columns are returned.
• metadata_table_group_dict (dict): A dictionary of a metadata Table Group object, such as the return value of get_metadata_table_group_dict.
• data_folder (str): The filepath of a local folder where the normalized CSVW metadata file is saved.
• metadata_filename (str): The filename of a CSVW metadata file which has been created by the download_table_group method and is located in the data folder.

Returns (dict of dicts): A dictionary with:

• keys: the names of the column(s)
• values: a dictionary with keys as lookup codes and values as code descriptions.

import_table_group_to_sqlite

Description: Reads a CSVW metadata file and imports the CSV data into a SQLite database. This makes use of the https://purl.org/berg/csvw_functions_extra vocabulary.

Method:

1. If not already present, a SQLite database named database_name is created in the data_folder.
2. For each table in the TableGroup object, the local CSV file is located in the data_folder using https://purl.org/berg/csvw_functions_extra/vocab/csv_file_name.
3. The CSV file is imported into the SQLite database into a table named using https://purl.org/berg/csvw_functions_extra/vocab/sql_table_name.
4. Primary key field(s) are set up using the information in the CSVW TableSchema primaryKey value.
5. Indexes are set up on columns if https://purl.org/berg/csvw_functions_extra/vocab/sqlsetindex is True.

Call signature:

csvw_functions_extra.import_table_group_to_sqlite(
        metadata_filename,
        data_folder,
        database_name,
        csv_file_names=None, 
        remove_existing_tables=False,
        verbose=False

Arguments:

• metadata_filename (str): The filename of a CSVW metadata file which has been created by the download_table_group method and is located in the data folder.
• data_folder (str): The filepath of a local folder where the downloaded CSV data is located and the SQLite database is stored.
• database_name (str): The name of the SQLite database, relative to the data_folder.
• csv_file_names (str or list): The csv_file_name values of the tables to be imported. If None then all CSV files are imported.
• overwrite_existing_tables (bool): If True, then before importing the CSV data any associated existing table in the database is removed and recreated.
• verbose (bool): If True, then this function prints intermediate variables and other useful information.

Returns: None

add_index

Description: Adds an SQlite index to a column in a SQlite database.

csvw_functions_extra.add_index(
        fields,
        table_name,
        data_folder,
        database_name,
        unique=False,
        verbose=False
        )

Arguments:

• fields *(str or list): The field(s) (i.e. columns) to add the index to.
• table_name (str): The name of the table in the SQLite database.
• data_folder (str): The filepath of a local folder where the SQLite database is stored.
• database_name (str): The name of the SQLite database, relative to the data_folder.
• unique (bool): If True, then a unique index is created.

get_all_table_names_in_database

Description: Returns a list of all table names in the database.

csvw_functions_extra.get_all_table_names_in_database(
        data_folder,
        database_name        
        )

Arguments:

• data_folder (str): The filepath of a local folder where the SQLite database is stored.
• database_name (str): The name of the SQLite database, relative to the data_folder.

Returns (list): A list of all table names in the SQLite database.

get_field_names

Description: Returns a list of the field names in a database table.

csvw_functions_extra.get_field_names(
        table_name,
        data_folder,
        database_name,
        verbose=False
        )

Arguments:

• table_name (str): The name of the table in the SQLite database.
• data_folder (str): The filepath of a local folder where the SQLite database is stored.
• database_name (str): The name of the SQLite database, relative to the data_folder.

Returns (list): A list of all field names of the table in the SQLite database.

get_row_count

Description: Returns the number of rows from a table.

csvw_functions_extra.get_row_count(
        table_name,
        data_folder
        database_name,
        filter_by=None,
        group_by=None,
        verbose=False
        )

Arguments:

• table_name (str): The name of the table in the SQLite database.
• data_folder (str): The filepath of a local folder where the SQLite database is stored.
• database_name (str): The name of the SQLite database, relative to the data_folder.
• filter_by (dict): A dictionary with information to filter the rows - see get_where_string.
• group_by (list): A list of field names to group by.

Returns (list): A list of result dictionaries.

get_rows

Description: Returns one or more rows from a table in the database.

csvw_functions_extra.get_rows(
        table_name,
        data_folder,
        database_name,
        filter_by = None,  
        fields = None,  
        limit = None,
        replace_codes = False,
        metadata_filename = None,
        verbose = False
        )

Arguments:

• table_name (str): The name of the table in the SQLite database.
• data_folder (str): The filepath of a local folder where the SQLite database is stored.
• database_name (str): The name of the SQLite database, relative to the data_folder.
• filter_by (dict): A dictionary with information to filter the rows - see get_where_string.
• fields (list): A list of field names to return.
• limit (integer): The number of rows to return. If None, then all rows are returned.
• replace_codes *(bool):
• metadata_filename (str): Required only if replace_codes is True.

Returns (list): A list of result dictionaries.

get_sql_table_names_in_database

Description: Returns a list of table names in the database which are also present in the CSVW metadata file.

csvw_functions_extra.get_sql_table_names_in_database(
        data_folder,
        database_name,
        metadata_filename
        )

Arguments:

• data_folder (str): The filepath of a local folder where the SQLite database is stored.
• database_name (str): The name of the SQLite database, relative to the data_folder.
• metadata_filename (str): The filename of a CSVW metadata file which has been created by the download_table_group method and is located in the data folder.

Returns: A list of table names in the database which are also present as https://purl.org/berg/csvw_functions_extra/vocab/sql_table_name values in the CSVW metadata file.

run_sql

Description: Runs an SQL query on the database and returns the result.

csvw_functions_extra.run_sql(
        sql_query,
        data_folder,
        database_name,
        verbose=False
        )

Arguments:

• sql_query (str): A SQL query.
• data_folder (str): The filepath of a local folder where the SQLite database is stored.
• database_name (str): The name of the SQLite database, relative to the data_folder.

Returns (list): A list of dictionaries where each dictionary contains one set of results - keys are the field (column) names and values are the data values.

convert_to_iterator

Description: Converts a value to a list.

csvw_functions_extra.convert_to_iterator(
        x
        )

Arguments: x *(int, float, string, list, tuple, None): The value to be converted to an iterator.

Returns (list): A list of value(s), where:

• A number is converted to a list of the number, e.g. 2 -> [2]
• A string to a list of the string, e.g. 'abc' -> ['abc']
• A list (or other iterable) remains the same, e.g. [1,2,3] -> [1,2,3]
• None is converted to an empty list, e.g. None -> []

get_field_string

Description: Converts a list of field names into a string for use in a SQL query.

get_field_string(
        fields = None
        )

Arguments:

• fields (str, list or None): The field name(s)

Returns (str): A string of the field names, where:

• None is converted to *
• A string is converted to a quoted string, e.g. 'field' -> ' "field1" '
• A list is converted to a series of quoted strings separated by commas, e.g. ['field1','field2'] -> ' "field1","field2" '

get_group_by_string

Description: Converts a list of field names into two strings for use in GROUP BY clauses in a SQL query.

get_group_by_string(
        group_by = None
        )

Arguments:

• group_by (str, list or None): The field name(s) to group by

Returns (tuple): A two-item tuple of (group_by_fields,group_by_string), where:

• None is converted to ('', '')
• 'field1' is converted to (' "field1", ', 'GROUP BY "field1" ')
• ['field1','field2'] is converted to (' "field1","field2", ', 'GROUP BY "field1","field2" ')

get_where_string

Description: Converts a dictionary of field names and values into a string for use in WHERE clauses in a SQL query.

get_where_string(
        filter_by = None 
        )

Arguments:

• group_by (dict or None): The field name(s) and values to filter by

Returns (str): A string to use in a WHERE clause, where:

• None is converted to ''
• {'field1': 1} is converted to ' WHERE ("field1" = 1)'
• {'field1': 1, 'field2': 'a', 'field3': None} is converted to ' WHERE ("field1" = 1) AND ("field2" = "a") AND ("field3" = Null)'
• {'field1': [1,2]} is converted to ' WHERE ("field1" IN (1,2))'
• {'field1': ['a','b']} is converted to ' WHERE ("field1" IN ("a","b"))'
• {'field1': {'BETWEEN':[1,2]}} is converted to ' WHERE ("field1" BETWEEN (1 AND 2))'
• {'field1': {'BETWEEN':['a','b']}} is converted to ' WHERE ("field1" BETWEEN ("a" AND "b"))' TO DO: Include True and False

CSVW vocabulary

Vocabulary on CSVW column metadata objects

• https://purl.org/berg/csvw_functions/vocab/codes: (JSON object) An object (dictionary) relating any codes using in the column data to a string with the meaning of the codes.

• https://purl.org/berg/csvw_functions/vocab/column_notes: (string) A description of the contents of the column.

• https://purl.org/berg/csvw_functions_extra/vocab/sqlsetindex: Boolean. If true then an SQLite index is set up on this column when the data is imported into the database.

Vocabulary on CSVW table metadata objects

Required:

• https://purl.org/berg/csvw_functions_extra/vocab/csv_file_name: The name for the newly downloaded CSV file. The CSV file is saved in the data folder using this name.

• https://purl.org/berg/csvw_functions_extra/vocab/sql_table_name: The name to be used for the database table when importing the CSV file into a SQLite database.

For CSV file downloads:

• https://purl.org/berg/csvw_functions_extra/vocab/csv_download_url: The url where the remote CSV file can be downloaded from.

For ZIP downloads:

• https://purl.org/berg/csvw_functions_extra/vocab/zip_download_url: The url where the remote ZIP file can be downloaded from.

• https://purl.org/berg/csvw_functions_extra/vocab/zip_file_name: The name for the newly downloaded ZIP file. The ZIP file is saved in the data folder using this name.

• https://purl.org/berg/csvw_functions_extra/vocab/csv_zip_extract_path: The path to extract the CSV file from the ZIP file.

Optional:

• https://purl.org/berg/csvw_functions_extra/vocab/metadata_download_url: The url where an associated metadata file for the CSV file can be downloaded from.

• https://purl.org/berg/csvw_functions_extra/vocab/metadata_file_suffix: A suffix to use when saving the associated metadata file.