sqlalchemy-csv-normalise

SQLAlchemy utilities for normalising / denormalising table data, useful for CSV

Where a table is normalised to have “lookup tables” of values referred to by e.g. a numeric foreign-key ID, these functions will enable extraction of the data (or conversely, loading from such) with the looked-up values substituted in. Among other things, this allows more human-friendly data editing in e.g. a spreadsheet.

Example:

from sqlalchemy_csv_normalise import denormalise_prepare
q, col_names = denormalise_prepare(db.session, table)
filename = table_to_filename(table)
with open(filename, 'w', newline='') as csv_file:
    csv_file_writer = csv.writer(csv_file)
    csv_file_writer.writerow(col_names)
    csv_file_writer.writerows(q.all())

from sqlalchemy_csv_normalise import renormalise_prepare, empty_deleter,\
    type_coercer
row_maker = renormalise_prepare(db.session, table)
row_cleaner = empty_deleter(table)
row_coercer = type_coercer(table)
filename = table_to_filename(table)
with open(filename, newline='') as csv_file:
    for d in csv.DictReader(csv_file):
        row = row_coercer(row_cleaner(row_maker(d)))
        db.session.add(table(**row))
db.session.commit()

• Free software: MIT license

• Documentation: https://sqlalchemy-csv-normalise.readthedocs.io.

History