Calling .transform() on a table with INTEGER PRIMARY KEY AUTOINCREMENT
recreated the table without the AUTOINCREMENT keyword, silently changing
the table's behaviour so that deleted row IDs could be reused.
The fix adds an autoincrement parameter to create_table_sql() and detects
AUTOINCREMENT in the existing schema before calling it from transform_sql().
> Add a test that covers what happens if you run transform against a table with ON CASCADE DELETE for one of its foreign keys - those records should not be deleted during the transform even though the table is dropped as part of that procedure
* Fix transform() corrupting TRUE/FALSE/NULL column defaults
quote_default_value() passed keyword-literal defaults (TRUE, FALSE, NULL)
through self.quote(), wrapping them in quotes. So transform() rebuilt a
column declared "INTEGER DEFAULT TRUE" as "INTEGER DEFAULT 'TRUE'", and a
later default insert stored the text 'TRUE' instead of the integer 1
(likewise 'FALSE' for 0 and 'NULL' for null) - silent value corruption.
Return these keyword literals unquoted, as already done for the
CURRENT_TIME/DATE/TIMESTAMP literals.
PR #764
* Recreate indexes when calling transform when possible and raise an error when they cannot be retained automatically
* Docs for sqlite_utils.db.TransformError
Co-authored-by: Simon Willison <swillison@gmail.com>
Closes#577
This should solve all sorts of problems seen by users of platforms that throw errors on writable_schema.
Also added `add_foreign_keys=` and `foreign_keys=` parameters to `table.transform()`.
