Synopsis
A Python package that provides a simple utility function for creating SQLite tables. The function will compare an existing table’s SQL statement from the sqlite_master
table to the one expected for a given SQL table creation statement. If the existing table does not match the expected SQL statement then it can either be dropped and recreated, or a ValueError
can be raised.
The primary purpose of this function is to facilitate the creation of tables for caching temporary data that can be discarded, such as memoizing calls to remote server APIs. When the format of the cached data changes over time, only the SQL table creation statement will need to be updated. The user will not need to manually alter or delete the previous table. Or course, if the prevously cached data should be kept rather than discarded when the table changes, the create_and_check_table
function will not help.
Links
GitLab
Other Repositories
Usage
import sqlite3
from sqlite3_create_and_check_table import create_and_check_table
sql = "CREATE TABLE test (prim TEXT PRIMARY KEY, int INTEGER, blob BLOB NON NULL)"
conn = sqlite3.connect("/path/to/some/sqlite3/db.sqlite")
with conn as cur:
# Create the table with the given SQL statement. If a table with the same
# name exists but differs from what would result from the given SQL table
# creation statement, it will be dropped and recreated to match the given
# statement.
# If drop is set to False then a ValueError will be raised instead.
create_and_check_table(cur, sql, drop=True)
The package also provides a row factory function for returning rows in the table as dict
s as well as an sqlite3 cursor context manager for temporarily setting the the row factory to the provided dict
factory:
import sqlite3
from sqlite3_create_and_check_table import dict_factory, dict_factory_cursor
conn = sqlite3.connect("/path/to/some/sqlite3/db.sqlite")
# Set the row factory within a context and restore the previous row factory when
# leaving the context.
with dict_factory_cursor(conn) as cur:
# cur.execute(...)
# Set the row factory for all queries.
conn.row_factory = dict_factory