SQL INSERT with multiple rows of values using Jython and zxJDBC

Question

I'm programming in Jython to an SQL database, using the com.ziclix.python.sql package. I'm wondering if the package has support in its prepared statements for SQL INSERT/UPDATE statements where you pass multiple rows of values rather than just one row, so that you're inserting multiple rows in one INSERT statement. Let me explain.

Here's a basic SQL INSERT statement:

INSERT INTO Names (FirstName, LastName) VALUES ("John", "Doe");

Using the zxJDBC prepared statement, this is pretty trivial:

# Assume cursor object
statement = 'INSERT INTO Names (FirstName, LastName) VALUES (?, ?)'
cursor.execute(statement, ('John', 'Doe'))

But the kind of INSERT statement I'm interested in is where you insert multiple rows of values, like so:

INSERT INTO Names (FirstName, LastName) VALUES ("John", "Doe"), ("Mary", "Smith");

Is there some way to do this using prepared statements? I would rather use prepared statements to build up an SQL query rather than string interpolation, if possible. Thanks.

Answer 1

Since zxJDBC implements PEP 249 " Python Database API Specification v2.0", you can use the method executemany() to accomplish this:

# Assume cursor object
statement = 'INSERT INTO Names (FirstName, LastName) VALUES (?, ?)'
cursor.executemany(statement, (
    ('John', 'Doe'),
    ('Mary', 'Smith'),
    # ...
    )
)

Update: Looks like this is too slow, since zxJDBC just calls execute many times without optimising the query. Instead you could do:

from itertools import chain
statement = 'INSERT INTO Names (FirstName, LastName) VALUES {values}'.format(
    values = ','.join('(?, ?)' for _ in input_seq))
input_seq = (
    ('John', 'Doe'),
    ('Mary', 'Smith'),
    # ...
    )
cursor.execute(statement, list(chain(*input_seq)))

This does involve some string manipulation, but it preserves the use of parameter lists and allows a single DB round-trip.